Apparatus and method for object pose estimation in a medical image

Abstract

Apparatus and method for object pose estimation are disclosed. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps, segment the plurality of echo depth maps, determine a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps, generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum and generate a pose datum of the object of interest as a function of the 3D point cloud.

Description

[0001] APPARATUS AND METHOD FOR OBJECT POSE ESTIMATION IN A MEDICAL IMAGE

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This application is an International PCT application which claims the benefit of priority of U.S. Nonprovisional Application Serial No. 18 / 787,196, filed on July 29, 2024, and entitled “APPARATUS AND METHOD FOR OBJECT POSE ESTIMATION IN A MEDICAL IMAGE,” and claims the benefit of priority of U.S. Non-provisional Application Serial No. 19 / 273,914, filed on July 18, 2025, and entitled “APPARATUS AND METHOD FOR OBJECT POSE ESTIMATION IN A MEDICAL IMAGE,” each which is incorporated by reference herein in its entirety.

[0004] FIELD OF THE INVENTION

[0005] The present invention generally relates to the field of medical imaging. In particular, the present invention is directed to an apparatus and method for object pose estimation in a medical image.

[0006] BACKGROUND

[0007] Currently, one of the most used methods for tracking an “object of interest” in the body uses an intricately placed magnetic sensor on the insertable devices such as magnetic sensor enabled catheters, magnetic sensor enabled intra-cardiac echo catheters. These magnetic sensors make insertable devices costly. These devices are used once in a patient and then disposed of due to the invasive nature of the surgery adding to the cost.

[0008] SUMMARY OF THE DISCLOSURE

[0009] In some aspects, the techniques described herein relate to an apparatus for object pose estimation in a medical image, the apparatus including: at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segment the plurality of echo depth maps; determine a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps; generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generate a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum includes: training a pose estimation model using pose estimation training data, wherein the pose estimation training data includes exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

[0010] In some aspects, the techniques described herein relate to an apparatus, wherein receiving the plurality of sets of echo data includes receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

[0011] In some aspects, the techniques described herein relate to an apparatus, wherein the plurality of sets of echo data is related to a cadaveric organ with a catheter inserted into it.

[0012] In some aspects, the techniques described herein relate to an apparatus, wherein segmenting the plurality of echo depth maps includes: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

[0013] In some aspects, the techniques described herein relate to an apparatus, wherein determining the depth datum includes: training a depth model using depth training data, wherein the depth training data includes exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

[0014] In some aspects, the techniques described herein relate to an apparatus, wherein the pose estimation model includes a regression model and the pose estimation training data includes synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

[0015] In some aspects, the techniques described herein relate to an apparatus, wherein: the pose datum includes a 6D pose datum; and generating the pose datum includes generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

[0016] In some aspects, the techniques described herein relate to an apparatus, further including: a catheter, wherein the catheter includes: the object of interest including a distinct shape; and an electrode including the second object of interest.

[0017] In some aspects, the techniques described herein relate to an apparatus, wherein the memory contains instructions further configuring the at least a processor to: generate a 3D model as a function of the 3D point cloud; and generate a user interface including the 3D model; and display the user interface including the 3D model on a display device.

[0018] In some aspects, the techniques described herein relate to an apparatus, wherein the memory contains instructions further configuring the at least a processor to generate an anatomical datum as a function of the pose datum, wherein the anatomical datum includes a dimension datum of the object of interest.

[0019] In some aspects, the techniques described herein relate to an apparatus, wherein the pose datum includes a five degree (5D) pose datum.

[0020] In some aspects, the techniques described herein relate to a method for object pose estimation in a medical image, the method including: receiving, using at least a processor, a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segmenting, using the at least a processor, the plurality of echo depth maps; determining, using the at least a processor, a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps; generating, using the at least a processor, a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generating, using the at least a processor, a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum includes: training a pose estimation model using pose estimation training data, wherein the pose estimation training data includes exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

[0021] In some aspects, the techniques described herein relate to a method, wherein receiving the plurality of sets of echo data includes receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

[0022] In some aspects, the techniques described herein relate to a method, wherein the plurality of sets of echo data is related to a cadaveric organ with an inserting device of a catheter inserted into it.

[0023] In some aspects, the techniques described herein relate to a method, wherein segmenting the plurality of echo depth maps includes: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features. In some aspects, the techniques described herein relate to a method, wherein determining the depth datum includes: training a depth model using depth training data, wherein the depth training data includes exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

[0024] In some aspects, the techniques described herein relate to a method, wherein the pose estimation model includes a regression model and the pose estimation training data includes synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

[0025] In some aspects, the techniques described herein relate to a method, wherein: the pose datum includes a 6D pose datum; and generating the pose datum includes generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

[0026] In some aspects, the techniques described herein relate to a method, wherein: the object of interest of a catheter includes a distinct shape; and an electrode of the catheter includes the second object of interest.

[0027] In some aspects, the techniques described herein relate to a method, further including: generating, by the at least a processor, a 3D model as a function of the 3D point cloud; and generating, by the at least a processor, a user interface including the 3D model; and displaying, by the at least a processor, the user interface including the 3D model on a display device.

[0028] In some aspects, the techniques described herein relate to a method, further including: generating, using the at least a processor, an anatomical datum as a function of the pose datum, wherein the anatomical datum includes a dimension datum of the object of interest.

[0029] In some aspects, the techniques described herein relate to a method, wherein the pose datum includes a five degree (5D) pose datum.

[0030] In some aspects, the techniques described herein relate to an apparatus for object pose estimation in a medical image, the apparatus including: one or more ultrasound imaging systems located on a surface of a subject; an object of interest configured to be placed within the subject; at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive a plurality of sets of echo data from the one or more ultrasound imaging systems, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segment the plurality of echo depth maps; determine a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps; generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generate a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum includes: training a pose estimation model using pose estimation training data, wherein the pose estimation training data includes exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

[0031] In some aspects, the techniques described herein relate to an apparatus, wherein receiving the plurality of sets of echo data includes receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

[0032] In some aspects, the techniques described herein relate to an apparatus, wherein the plurality of sets of echo data is related to a cadaveric organ with a catheter inserted into it.

[0033] In some aspects, the techniques described herein relate to an apparatus, wherein segmenting the plurality of echo depth maps includes: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

[0034] In some aspects, the techniques described herein relate to an apparatus, wherein determining the depth datum includes: training a depth model using depth training data, wherein the depth training data includes exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

[0035] In some aspects, the techniques described herein relate to an apparatus, wherein the pose estimation model includes a regression model and the pose estimation training data includes synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

[0036] In some aspects, the techniques described herein relate to an apparatus, wherein: the pose datum includes a 6D pose datum; and generating the pose datum includes generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

[0037] In some aspects, the techniques described herein relate to an apparatus, further including: a catheter, wherein the catheter includes: the object of interest including a distinct shape; and an electrode including the second object of interest.

[0038] In some aspects, the techniques described herein relate to an apparatus, wherein the memory contains instructions further configuring the at least a processor to: generate a 3D model as a function of the 3D point cloud; generate a user interface including the 3D model; and display the user interface including the 3D model on a display device.

[0039] In some aspects, the techniques described herein relate to an apparatus, wherein the memory contains instructions further configuring the at least a processor to generate an anatomical datum as a function of the pose datum, wherein the anatomical datum includes a dimension datum of the object of interest.

[0040] In some aspects, the techniques described herein relate to an apparatus, wherein the pose datum includes a five degree (5D) pose datum.

[0041] In some aspects, the techniques described herein relate to a method for object pose estimation in a medical image, the method including: placing, one or more ultrasound imaging systems on a surface of a subject; positioning an object of interest configured to be within the subject; receiving, using at least a processor and from the one or more ultrasound imaging systems, a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segmenting, using the at least a processor, the plurality of echo depth maps; determining, using the at least a processor, a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps; generating, using the at least a processor, a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generating, using the at least a processor, a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum includes: training a pose estimation model using pose estimation training data, wherein the pose estimation training data includes exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

[0042] In some aspects, the techniques described herein relate to a method, wherein receiving the plurality of sets of echo data includes receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

[0043] In some aspects, the techniques described herein relate to a method, wherein the plurality of sets of echo data is related to a cadaveric organ with an inserting device of a catheter inserted into it.

[0044] In some aspects, the techniques described herein relate to a method, wherein segmenting the plurality of echo depth maps includes: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

[0045] In some aspects, the techniques described herein relate to a method, wherein determining the depth datum includes: training a depth model using depth training data, wherein the depth training data includes exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

[0046] In some aspects, the techniques described herein relate to a method, wherein the pose estimation model includes a regression model and the pose estimation training data includes synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

[0047] In some aspects, the techniques described herein relate to a method, wherein: the pose datum includes a 6D pose datum; and generating the pose datum includes generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

[0048] In some aspects, the techniques described herein relate to a method, wherein: the object of interest of a catheter includes a distinct shape; and an electrode of the catheter includes the second object of interest.

[0049] In some aspects, the techniques described herein relate to a method, further including: generating, by the at least a processor, a 3D model as a function of the 3D point cloud; generating, by the at least a processor, a user interface including the 3D model; and displaying, by the at least a processor, the user interface including the 3D model on a display device.

[0050] In some aspects, the techniques described herein relate to a method, further including: generating, using the at least a processor, an anatomical datum as a function of the pose datum, wherein the anatomical datum includes a dimension datum of the object of interest.

[0051] In some aspects, the techniques described herein relate to a method, wherein the pose datum includes a five degree (5D) pose datum.

[0052] In some aspects, the techniques described herein relate to an apparatus for object pose estimation in a medical image, the apparatus including: one or more ultrasound imaging systems located on a surface of a subject; an object of interest configured to be placed within the subject; at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive a plurality of sets of echo data from the one or more ultrasound imaging systems, wherein the plurality of sets of echo data are configured for generation of a plurality of echo depth maps; generate a three dimensional (3D) point cloud related to the object of interest as a function of the plurality of sets of echo data; and generate a pose datum of the object of interest as a function of the 3D point cloud using a pose estimation model.

[0053] In some aspects, the techniques described herein relate to an apparatus, wherein the one or more ultrasound imaging systems includes: a first ultrasound imaging system located at a first position on the surface of the subject; and a second ultrasound imaging system located at a second position on the surface of the subject.

[0054] In some aspects, the techniques described herein relate to an apparatus, wherein a first set of echo data of the plurality of sets of echo data and a second set of echo data of the plurality of sets of echo data include differing views of the object of interest.

[0055] In some aspects, the techniques described herein relate to an apparatus, wherein the at least a processor is further configured to segment the plurality of echo depth maps to generate a plurality of segmented echo depth maps.

[0056] In some aspects, the techniques described herein relate to an apparatus, wherein segmenting the plurality of echo depth maps includes: extracting the plurality of echo depth maps as a function of the plurality of sets of echo data; identifying a spatial expanse of the object of interest as a function of at least an object feature; and segmenting the plurality of echo depth maps as a function of the spatial expanse.

[0057] In some aspects, the techniques described herein relate to an apparatus, wherein the at least a processor is further configured to determine a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps.

[0058] In some aspects, the techniques described herein relate to an apparatus, wherein the at least a processor is further configured to determine a depth datum using a depth model, wherein: the depth model includes a convolutional neural network (CNN); and the at least a processor is further configured to use the depth model to predict the depth datum at each pixel of the plurality of segmented echo depth maps.

[0059] In some aspects, the techniques described herein relate to an apparatus, wherein generating the 3D point cloud includes aggregating each 3D point of a plurality of 3D points of the object of interest, wherein each 3D point of the plurality of 3D points is generated by converting a 2D pixel coordinate of a segmented echo depth map into a 3D coordinate by adding a depth datum as a z-value.

[0060] In some aspects, the techniques described herein relate to an apparatus, wherein the at least a processor is further configured to generate a 3D model as a function of the 3D point cloud, wherein generating the 3D model includes applying at least a 3D reconstruction algorithm to the 3D point cloud.

[0061] In some aspects, the techniques described herein relate to an apparatus, wherein generating the pose datum includes determining a pose of a sub-part of the object of interest, wherein: the sub-part has a fixed spatial relationship to a plurality of electrodes on a catheter; and determining the pose of a sub-part of the object of interest includes calculating a pose of the plurality of electrodes as a function of the pose of the sub-part of the object of interest and a rigidity constraint between the sub-part of the object of interest and the plurality of electrodes.

[0062] In some aspects, the techniques described herein relate to a method for object pose estimation in a medical image, the method including: receiving, by at least a processor, a plurality of sets of echo data from one or more ultrasound imaging systems located on a surface of a subject, wherein the plurality of sets of echo data are configured for generation of a plurality of echo depth maps; generating, using the at least a processor, a three dimensional (3D) point cloud related to the object of interest as a function of the plurality of sets of echo data; and generating, using the at least a processor, a pose datum of the object of interest as a function of the 3D point cloud using a pose estimation model.

[0063] In some aspects, the techniques described herein relate to a method, wherein the one or more ultrasound imaging systems includes: a first ultrasound imaging system located at a first position on the surface of the subject; and a second ultrasound imaging system located at a second position on the surface of the subject.

[0064] In some aspects, the techniques described herein relate to a method, wherein a first set of echo data of the plurality of sets of echo data and a second set of echo data of the plurality of sets of echo data include differing views of the object of interest. In some aspects, the techniques described herein relate to a method, further including segmenting the plurality of echo depth maps to generate a plurality of segmented echo depth maps.

[0065] In some aspects, the techniques described herein relate to a method, wherein segmenting the plurality of echo depth maps includes: extracting the plurality of echo depth maps as a function of the plurality of sets of echo data; identifying a spatial expanse of the object of interest as a function of at least an object feature; and segmenting the plurality of echo depth maps as a function of the spatial expanse.

[0066] In some aspects, the techniques described herein relate to a method, further including determining a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps.

[0067] In some aspects, the techniques described herein relate to a method, further including determining a depth datum using a depth model, wherein: the depth model includes a convolutional neural network (CNN); and determining a depth datum further includes using the depth model to predict the depth datum at each pixel of the plurality of segmented echo depth maps.

[0068] In some aspects, the techniques described herein relate to a method, wherein generating the 3D point cloud includes aggregating each 3D point of a plurality of 3D points of the object of interest, wherein each 3D point of the plurality of 3D points is generated by converting a 2D pixel coordinate of a segmented echo depth map into a 3D coordinate by adding a depth datum as a z-value.

[0069] In some aspects, the techniques described herein relate to a method, further including generating a 3D model as a function of the 3D point cloud, wherein generating the 3D model includes applying at least a 3D reconstruction algorithm to the 3D point cloud.

[0070] In some aspects, the techniques described herein relate to a method, wherein generating the pose datum includes determining a pose of a sub-part of the object of interest, wherein: the sub-part has a fixed spatial relationship to a plurality of electrodes on a catheter; and determining the pose of a sub-part of the object of interest includes calculating a pose of the plurality of electrodes as a function of the pose of the sub-part of the object of interest and a rigidity constraint between the sub-part of the object of interest and the plurality of electrodes.

[0071] The details of one or more variations of the subject matter described herein are set forth in the accompanying drawings and the description below. Other features and advantages of the subject matter described herein will be apparent from the description and drawings, and from the claims.

[0072] DESCRIPTION OF DRAWINGS

[0073] For the purpose of illustrating the invention, the drawings show aspects of one or more embodiments of the invention. However, it should be understood that the present invention is not limited to the precise arrangements and instrumentalities shown in the drawings, wherein: FIG. 1 illustrates a block diagram of an exemplary apparatus for object pose estimation in a medical image;

[0074] FIG. 2 illustrates a configuration of an exemplary user interface displayed on a remote device; FIG. 3 illustrates a block diagram of an exemplary machine-learning module;

[0075] FIG. 4 illustrates a diagram of an exemplary neural network;

[0076] FIG. 5 illustrates a block diagram of an exemplary node in a neural network;

[0077] FIG. 6 illustrates a configuration of exemplary echo transducers examining an object of interest within a body;

[0078] FIG. 7 illustrates a flow diagram of an exemplary method for object pose estimation in a medical image;

[0079] FIG. 8 is a block diagram of an exemplary embodiment of an apparatus for generating a three- dimensional (3D) model of cardiac anatomy via machine-learning;

[0080] FIG. 9 shows an exemplary embodiment of an intracardiac echocardiography (ICE) image;

[0081] FIG. 10 is a flow diagram of an exemplary embodiment of an ICE image example generation process;

[0082] FIG. 11; illustrates an exemplary embodiment of a three-dimensional (3D) voxel occupancy representation; and

[0083] FIG. 12 is a block diagram of a computing system that can be used to implement any one or more of the methodologies disclosed herein and any one or more portions thereof.

[0084] The drawings are not necessarily to scale and may be illustrated by phantom lines, diagrammatic representations, and fragmentary views. In certain instances, details that are not necessary for an understanding of the embodiments or that render other details difficult to perceive may have been omitted. Like reference symbols in the various drawings indicate like elements.

[0085] DETAILED DESCRIPTION At a high level, aspects of the present disclosure are directed to apparatuses and methods for object pose estimation in a medical image are disclosed. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps, segment the plurality of echo depth maps, determine a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps, generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum and generate a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum includes training a pose estimation model using pose estimation training data, wherein the pose estimation training data includes exemplary 3D point clouds correlated to exemplary pose datums and generating the pose datum using the trained pose estimation model. Exemplary embodiments illustrating aspects of the present disclosure are described below in the context of several specific examples.

[0086] Referring now to FIG. 1, an exemplary embodiment of an apparatus 100 for object pose estimation in a medical image is illustrated. Apparatus 100 includes at least a processor 104. Processor 104 may include, without limitation, any processor described in this disclosure. Processor 104 may be included in a computing device. Processor 104 may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and / or system on a chip (SoC) as described in this disclosure. Processor 104 may include, be included in, and / or communicate with a mobile device such as a mobile telephone or smartphone. Processor 104 may include a single computing device operating independently, or may include two or more computing device operating in concert, in parallel, sequentially or the like; two or more computing devices may be included together in a single computing device or in two or more computing devices. Processor 104 may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting processor 104 to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone / voice provider (e.g., a mobile communications provider data and / or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and / or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and / or from a computer and / or a computing device. Processor 104 may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Processor 104 may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Processor 104 may distribute one or more computing tasks as described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing devices. Processor 104 may be implemented, as a non-limiting example, using a “shared nothing” architecture.

[0087] With continued reference to FIG. 1, processor 104 may be designed and / or configured to perform any method, method step, or sequence of method steps in any embodiment described in this disclosure, in any order and with any degree of repetition. For instance, processor 104 may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and / or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and / or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and / or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Processor 104 may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and / or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and / or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and / or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and / or parallel processing.

[0088] With continued reference to FIG. 1, apparatus 100 includes a memory 108 communicatively connected to processor 104. For the purposes of this disclosure, “communicatively connected” means connected by way of a connection, attachment or linkage between two or more relata which allows for reception and / or transmittance of information therebetween. For example, and without limitation, this connection may be wired or wireless, direct or indirect, and between two or more components, circuits, devices, systems, and the like, which allows for reception and / or transmittance of data and / or signal(s) therebetween. Data and / or signals therebetween may include, without limitation, electrical, electromagnetic, magnetic, video, audio, radio and microwave data and / or signals, combinations thereof, and the like, among others. A communicative connection may be achieved, for example and without limitation, through wired or wireless electronic, digital or analog, communication, either directly or by way of one or more intervening devices or components. Further, communicative connection may include electrically coupling or connecting at least an output of one device, component, or circuit to at least an input of another device, component, or circuit. For example, and without limitation, via a bus or other facility for intercommunication between elements of a computing device. Communicative connecting may also include indirect connections via, for example and without limitation, wireless connection, radio communication, low power wide area network, optical communication, magnetic, capacitive, or optical coupling, and the like. In some instances, the terminology “communicatively coupled” may be used in place of communicatively connected in this disclosure.

[0089] With continued reference to FIG. 1, memory 108 contains instructions configuring processor 104 to receive echo data 112, wherein the echo data 112 is configured for generation of a plurality of echo depth maps 116. For the purposes of this disclosure, “echo data” is data collected by an echo transducer during an imaging process. As a non-limiting example, echo data 112 may include two dimensional (2D) images of an object of interest and surroundings. In some embodiments, receiving a plurality of sets of echo data 112 may include receiving the plurality of sets of echo data 112 from a plurality of echo transducers 120. For the purposes of this disclosure, an “echo transducer” is a device that generates information related to sound waves traveled through various tissues and reflected back when the sound waves encounter an object of interest. As a non-limiting example, echo transducer 120 may include an ultrasound imaging system. In a non-limiting example, the ultrasound imaging system may include a point-of-care ultrasound (POCUS) (e.g., portable ultrasound) or a conventional ultrasound (e.g., ultrasound that a patient must travel to the ultrasound machine). In some embodiments, conventional ultrasound can assess an anatomical region using predefined parameters and measurements to provide a diagnosis while POCUS can assess one part of the body at a time; this may allow user to answer very specific questions in the context of a physical exam and patient history. In some embodiments, each of the plurality of echo transducers 120 may examine and capture echo data 112 of an object of interest 124 in different angles or views; therefore, processor 104 may generate a plurality of echo depth maps 116 that has different angles or views of object of interest 124. In some embodiments, one echo transducer 120 may capture a plurality of sets of echo data 112 of an object of interest 124 in different angles or views. For the purposes of this disclosure, an “object of interest” is a particular element or area within an image, dataset, or scene. As a non-limiting example, object of interest 124 may include specific tissue, cyst, or lumps in an organ 126, implant, inserting devices 128 within an organ 126, and the like. For the purposes of this disclosure, an “organ” is a structure within an organism that is composed of multiple types of tissues and performs a specific function or set of functions. As a non-limiting example, organ may include heart, lung, kidney, liver, stomach, brain, and the like. In some embodiments, the organ 126 may include a cadaveric organ as described below. For the purposes of this disclosure, an “inserting device” is any device that can be inserted into the body. As a non-limiting example, inserting device 128 may include a catheter. For the purposes of this disclosure, a “catheter” is a tube that is inserted into the body to perform a variety of medical procedures. In some embodiments, echo transducers 120 may operate by converting electrical signals into sound waves and vice versa, capturing 2D ultrasound images of an object of interest 124, such as cardiac structures or insertable devices. In some embodiments, inserting device 128 or object of interest 124 may include at least a sensor 131. For the purposes of this disclosure, a “sensor” is a device that produces an output signal for the purpose of sensing a physical phenomenon. As a non-limiting example, sensor 131 may include a location sensor which may be used to locate inserting device 128 or object of interest 124; such as a catheter or implant. In some embodiments, inserting device 128 and / or object of interest 124 may not include a location sensor.

[0090] With continued reference to FIG. 1, in some embodiments, echo data 112 may form the basis for generating echo depth maps 116, which can provide information about the relative distances of various points within an object of interest 124 from echo transducer 120. For the purposes of this disclosure, an “echo depth map” is a type of data visualization created using echo data, which provides a representation of the distances from an echo transducer to various points within an object of interest. As a non-limiting example, echo depth map 116 may include gradient colors, shades and contours, and the like. In a non-limiting example, echo depth map may include a 2D image of a catheter (e.g., inserting device 128 and object of interest 124) within a left ventricle of a heart that visualizes spatial relationships and depths of both the heart structures and the inserted catheter.

[0091] With continued reference to FIG. 1, in some embodiments, apparatus 100 may include an echo database 130. As used in this disclosure, “echo database” is a data store configured to store data associated with echo data. As a non-limiting example, echo database 130 may store echo data 112, echo depth map 116, object feature 132, depth datum 136, segmented echo depth map 140, three dimensional (3D) point cloud 144, pose datum 148, and the like. In one or more embodiments, echo database 130 may include inputted or calculated information and datum related to echo data 112. In some embodiments, a datum history may be stored in echo database 130. As a non-limiting example, the datum history may include real-time and / or previous inputted data related to echo data 112. As a non-limiting example, echo database 130 may include instructions from a user, who may be an expert user, a past user in embodiments disclosed herein, or the like, where the instructions may include examples of the data related to echo data 112.

[0092] With continued reference to FIG. 1, in some embodiments, processor 104 may be communicatively connected with echo database 130. For example, and without limitation, in some cases, echo database 130 may be local to processor 104. In another example, and without limitation, echo database 130 may be remote to processor 104 and communicative with processor 104 by way of one or more networks. The network may include, but is not limited to, a cloud network, a mesh network, and the like. By way of example, a “cloud-based” system can refer to a system which includes software and / or data which is stored, managed, and / or processed on a network of remote servers hosted in the “cloud,” e.g., via the Internet, rather than on local severs or personal computers. A “mesh network” as used in this disclosure is a local network topology in which the infrastructure processor 104 connect directly, dynamically, and non-hierarchically to as many other computing devices as possible. A “network topology” as used in this disclosure is an arrangement of elements of a communication network. The network may use an immutable sequential listing to securely store echo database 130. An “immutable sequential listing,” as used in this disclosure, is a data structure that places data entries in a fixed sequential arrangement, such as a temporal sequence of entries and / or blocks thereof, where the sequential arrangement, once established, cannot be altered or reordered. An immutable sequential listing may be, include and / or implement an immutable ledger, where data entries that have been posted to the immutable sequential listing cannot be altered.

[0093] With continued reference to FIG. 1, in some embodiments, echo database 130 may be implemented, without limitation, as a relational database, a key -value retrieval database such as a NOSQL database, or any other format or structure for use as a database that a person skilled in the art would recognize as suitable upon review of the entirety of this disclosure. Database may alternatively or additionally be implemented using a distributed data storage protocol and / or data structure, such as a distributed hash table or the like. Database may include a plurality of data entries and / or records as described in this disclosure. Data entries in a database may be flagged with or linked to one or more additional elements of information, which may be reflected in data entry cells and / or in linked tables such as tables related by one or more indices in a relational database. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which data entries in a database may store, retrieve, organize, and / or reflect data and / or records as used herein, as well as categories and / or populations of data consistently with this disclosure.

[0094] With continued reference to FIG. 1, in some embodiments, processor 104 may receive echo data 112 from remote device 152. For the purposes of this disclosure, a “remote device” is an external device to a processor 104. As a non-limiting example, remote device 152 may include a laptop, desktop, tablet, mobile phone, smart phone, smart watch, smart headset, or things of the like. In some embodiments, a user may use remote device 152 to input any data into processor 104 or receive or manipulate any data from processor 104. For the purposes of this disclosure, a “user” is any person, individual, organization or entity that is using or has used an apparatus. As a non-limiting example, user may include a physician, clinician, nurses, doctors, medical professionals, hospitals, medical organization, and the like. In some embodiments, remote device 152 may include an interface configured to receive inputs from user. In some embodiments, user may manually input any data into apparatus 100 using remote device 152. Tn some embodiments, user may have a capability to process, store or transmit any information independently.

[0095] With continued reference to FIG. 1, processor 104 may receive echo data 112 using an application programming interface (API). As used in the current disclosure, an “application programming interface” is a software interface for two or more computer programs to communicate with each other. As a non-limiting example, API may include EHR APIs, telemedicine APIs, and the like. An application programming interface may be a type of software interface, offering a service to other pieces of software. In contrast to a user interface, which connects a computer to a person, an application programming interface may connect computers or pieces of software to each other. An API may not be intended to be used directly by a person (e g., a user) other than a computer programmer who is incorporating it into the software. An API may be made up of different parts which act as tools or services that are available to the programmer. A program or a programmer that uses one of these parts is said to call that portion of the API. The calls that make up the API are also known as subroutines, methods, requests, or endpoints. An API specification may define these calls, meaning that it explains how to use or implement them. One purpose of API may be to hide the internal details of how a system works, exposing only those parts a programmer will find useful and keeping them consistent even if the internal details later change. An API may be custom-built for a particular pair of systems, or it may be a shared standard allowing interoperability among many systems. The term API may be often used to refer to web APIs, which allow communication between computers that are joined by the internet. API may be configured to query for web applications in order to retrieve echo data 112 to another web application, database (e.g., echo database 130), medical center patient portal, and the like. An API may be further configured to filter through web applications according to a filter criterion. In this disclosure, “filter criteria” are conditions the web applications must fulfill in order to qualify for API. Web applications may be filtered based on these filter criteria. Filter criteria may include, without limitation, types of medical facilities, location of the medical facility, and the like.

[0096] With continued reference to FIG. 1, memory 108 contains instructions configuring processor 104 to segment a plurality of echo depth maps 116 to generate a plurality of segmented echo depth maps. For the purposes of this disclosure, a “segmented echo depth map” is a depth map that is segmented by different objects or regions. In some embodiments, processor 104 may segment a plurality of echo depth maps 116 to identify a spatial expanse of an object of interest 124. As a non-limiting example, the spatial expanse may encompass all the pixels or points that define the boundaries and interior of the object of interest 124. In a non-limiting example, in a 2D image, this would include all the pixels that make up the object of interest 124, while in a 3D context, it would encompass all the voxels (3D pixels) that constitute the object of interest 124. In some embodiments, segmenting echo depth maps 116 may include extracting object features 132 of object of interest 124 from the plurality of echo depth maps 116 and segmenting the plurality of echo depth maps 116 as a function of the object features 132. For the purposes of this disclosure, an “object feature” is the distinct attributes or characteristics of an object of interest. As a non-limiting example, object feature 132 may include geometric features, such as edges, corners, surfaces, textures, and the like. As another non-limiting example, object feature 132 may include depth value (e.g., depth datum 136 as described below). In some embodiments, processor 104 may extract object feature 132 using a feature extraction model or machine vision system. In another embodiment, processor 104 may generate depth datum 136 without segmenting echo depth map 116.

[0097] With continued reference to FIG. 1, in some embodiments, processor 104 may be configured to analyze echo depth map 116 using machine vision system to determine object feature 132. For the purposes of this disclosure, a “machine vision system” is a type of technology that enables a computing device to inspect, evaluate and identify still or moving images. For example, in some cases a machine vision system may be used for world modeling or registration of objects within a space. In some cases, registration may include image processing, such as without limitation object recognition, feature detection, edge / corner detection, and the like. Non-limiting example of feature detection may include scale invariant feature transform (SIFT), Canny edge detection, Shi Tomasi corner detection, and the like. In some cases, a machine vision process may operate image classification and segmentation models 156, such as without limitation by way of machine vision resource (e.g., OpenMV or TensorFlow Lite). A machine vision process may detect motion, for example by way of frame differencing algorithms. A machine vision process may detect markers, for example blob detection, object detection (e.g., object of interest 124), face detection, and the like.

[0098] With continued reference to FIG. 1, in some cases, registration may include one or more transformations to orient a camera frame (or an image or video stream) relative a three- dimensional coordinate system; exemplary transformations include without limitation homography transforms and affine transforms. In an embodiment, registration of first frame to a coordinate system may be verified and / or corrected using object identification and / or computer vision, as described above. For instance, and without limitation, an initial registration to two dimensions, represented for instance as registration to the x and y coordinates, may be performed using a two-dimensional projection of points in three dimensions onto a first frame, however. A third dimension of registration, representing depth and / or a z axis, may be detected by comparison of two frames; for instance, where first frame includes a pair of frames captured using a pair of cameras (e.g., stereoscopic camera also referred to in this disclosure as stereocamera), image recognition and / or edge detection software may be used to detect a pair of stereoscopic views of images of an object; two stereoscopic views may be compared to derive z- axis values of points on object permitting, for instance, derivation of further z-axis points within and / or around the object using interpolation. This may be repeated with multiple objects in field of view, including without limitation environmental features of interest identified by object classifier and / or indicated by an operator. In an embodiment, x and y axes may be chosen to span a plane common to two cameras used for stereoscopic image capturing and / or an xy plane of a first frame; a result, x and y translational components and < / > may be pre-populated in translational and rotational matrices, for affine transformation of coordinates of object, also as described above. Initial x and y coordinates and / or guesses at transformational matrices may alternatively or additionally be performed between first frame and second frame, as described above. For each point of a plurality of points on object and / or edge and / or edges of object as described above, x and y coordinates of a first stereoscopic frame may be populated, with an initial estimate of z coordinates based, for instance, on assumptions about object, such as an assumption that ground is substantially parallel to an xy plane as selected above. Z coordinates, and / or x, y, and z coordinates, registered using image capturing and / or object identification processes as described above may then be compared to coordinates predicted using initial guess at transformation matrices; an error function may be computed using by comparing the two sets of points, and new x, y, and / or z coordinates, may be iteratively estimated and compared until the error function drops below a threshold level.

[0099] With continued reference to FIG. 1, in some embodiments, segmenting a plurality of echo depth maps 1 16 may include training a segmentation model 156 with segmentation training data 160, wherein the segmentation training data 160 may include exemplary plurality of echo depth maps correlated to exemplary segmented plurality of echo depth maps and segmenting a plurality of echo depth maps 116 using the trained segmentation model 156. For the purposes of this disclosure, a “segmentation model” is a machine learning or deep learning model designed to partition an image into multiple segments or regions, each corresponding to different objects or parts of an object within the image. In some embodiments, segmentation model 156 may assign a label to each pixel in an image (e.g., echo depth map 116) such that pixels with the same label share certain characteristics, such as belonging to the same object of interest 124 or region. As a non-limiting example, segmentation model 156 may include a neural network. For the purposes of this disclosure, “segmentation training data” is data containing correlations that a machinelearning process may use to model relationships between echo depth maps and segmented echo depth maps. In a non-limiting example, a segmentation model 156 may analyze echo depth map 116 to identify and delineate the boundaries of object of interest 124. This may include finding the set of coordinates {(X,, Fj] that represent the pixels or voxels making up the object of interest 124. In some embodiments, processor 104 may segment echo depth map 116 based on object feature 132.

[0100] With continued reference to FIG. 1, in some embodiments, processor 104 may be configured to generate segmentation training data 160. In a non-limiting example, segmentation training data 160 may include correlations between exemplary echo depth maps, exemplary object features and exemplary segmented echo depth maps. In some embodiments, segmentation training data 160 may be stored in echo database 130. In some embodiments, segmentation training data 160 may be received from one or more users, echo database 130, external computing devices, and / or previous iterations of processing. As a non-limiting example, segmentation training data 160 may include instructions from a user, who may be an expert user, a past user in embodiments disclosed herein, or the like, which may be stored in memory and / or stored in echo database 130, where the instructions may include labeling of training examples. In some embodiments, segmentation training data 160 may be updated iteratively on a feedback loop. As a non-limiting example, processor 104 may update segmentation training data 160 iteratively through a feedback loop as a function of output of feature extraction model, echo data 112, echo depth map 116, and the like. In some embodiments, processor 104 may be configured to generate segmentation model 156. Tn a non-limiting example, generating segmentation model 156 may include training, retraining, or fine-tuning segmentation model 156 using segmentation training data 160 or updated segmentation training data 160. In some embodiments, processor 104 may be configured to segment echo depth map 116 using segmentation model 156 (i.e. trained or updated segmentation model 156). In some embodiments, echo data 112 or echo depth map 116 may be classified to a user cohort using a cohort classifier. Cohort classifier may be consistent with any classifier discussed in this disclosure. Cohort classifier may be trained on cohort training data, wherein the cohort training data may include echo data 112 or echo depth map 116 correlated to user cohorts. In some embodiments, a user may be classified to a user cohort and processor 104 may determine segmented echo depth map 116 or object feature 132 based on the user cohort and the resulting output may be used to update segmentation training data 160. In some embodiments, generating training data and training machine-learning models may be simultaneous.

[0101] With continued reference to FIG. 1, memory 108 contains instructions configuring processor 104 to determine a depth datum 136 related to pixels of an object of interest 124 as a function of a plurality of segmented echo depth maps 140. For the purposes of this disclosure, a “depth datum” is information or value that represents a distance from an echo transducer to a particular point within an object of interest or echo depth map. In a non-limiting example, depth datum 136 may include 20mm, indicating that the distance from echo transducer 120 to a point on object of interest 124 is 20 millimeters. In another non-limiting example, each pixel in the 2D image (e.g., echo depth map 116 or segmented echo depth map 140) may include an (x, y) coordinate and depth datum 136 (z) can be added to the pixel coordinates to convert them into 3D points (x, y, z). In some embodiments, determining depth datum 136 may include training a depth model 164 using depth training data 168, wherein the depth training data 168 may include exemplary plurality of sets of echo data correlated to exemplary depth datums and determining depth datum 136 using the trained depth model 164. In some embodiments, depth datum 136 may be stored in echo database 130. In some embodiments, processor 104 may retrieve depth datum 136. In some embodiments, user may manually input depth datum 136.

[0102] With continued reference to FIG. 1, some embodiments, processor 104 may be configured to generate depth training data 168. In a non-limiting example, depth training data 168 may include correlations between exemplary segmented echo depth maps or exemplary depth maps and exemplary depth datums. In some embodiments, depth training data 168 may be stored in echo database 130. In some embodiments, depth training data 168 may be received from one or more users, echo database 130, external computing devices, and / or previous iterations of processing. As a non-limiting example, depth training data 168 may include instructions from a user, who may be an expert user, a past user in embodiments disclosed herein, or the like, which may be stored in memory and / or stored in echo database 130, where the instructions may include labeling of training examples. In some embodiments, depth training data 168 may be updated iteratively on a feedback loop. As a non-limiting example, processor 104 may update depth training data 168 iteratively through a feedback loop as a function of output of feature extraction model, echo data 112, echo depth map 116, output of segmentation model 156, and the like. In some embodiments, processor 104 may be configured to generate depth model 164. In a non-limiting example, generating depth model 164 may include training, retraining, or fine-tuning depth model 164 using depth training data 168 or updated depth training data 168. In some embodiments, processor 104 may be configured to segment echo depth map 116 using depth model 164 (i.e. trained or updated depth model 164). As a nonlimiting example, depth model 164 may include a convolutional neural network (CNN). In some embodiments, depth model 164 may predict depth datum 136 at each pixel. In some embodiments, echo data 112 or echo depth map 116 may be classified to a user cohort using a cohort classifier. Cohort classifier may be consistent with any classifier discussed in this disclosure. Cohort classifier may be trained on cohort training data, wherein the cohort training data may include echo data 112 or echo depth map 116 correlated to user cohorts. In some embodiments, a user may be classified to a user cohort and processor 104 may determine segmented echo depth map 116 or object feature 132 based on the user cohort and the resulting output may be used to update depth training data 168. In some embodiments, generating training data and training machine-learning models may be simultaneous.

[0103] With continued reference to FIG. 1, in some embodiments, depth datum 136 related to pixels of object of interest 124 may be determined as a function of segmented echo depth maps 140. In some embodiments, depth model 164 may be trained depth training data 168 including echo 2D or 3D frames and corresponding data from 3D imaging modalities, such as computed tomography (CT), magnetic resonance imaging (MRI), 3D cardiography or electroanatomical mapping (EAM) system. In some embodiments, data from imaging modalities may provide depth datums 136 that can be used as a reference. For the purposes of this disclosure, an “electroanatomical mapping system” is a technology used to create three-dimensional maps using the heart's electrical activity of anatomical structure. In some embodiments, EAM system may include catheters with multiple electrodes to detect electrical signals from different parts of the heart. In some embodiments, EAM system may incorporate mapping software, which processes the electrical signals and positional data (e.g., depth datum 136) to construct and update the electroanatomical maps (e.g., echo depth map 116). These maps are displayed on monitors, allowing clinicians to visualize the heart’s activity and structure in great detail. In a non-limiting example, 3D cardiography may use ultrasound equipment to produce detailed three- dimensional images of the heart (e.g., echo data 112), providing precise depth information (e.g., depth datum 136) for each point within the cardiac structures. In another non-limiting example, EAM system may combine electrical signals with anatomical data to create detailed maps (e g., echo depth map 116) of the heart’s structure and function, providing spatial depth information (e.g., depth datum 136) and correlating electrical activity with anatomical features. In some embodiments, processor 104 may determine depth datum 136 using algorithms and techniques that do not rely on machine learning. These methods might include geometric calculations based on echo data 112 and heuristic approaches that use domain-specific knowledge to infer depth. In some embodiments, processor 104 may determine depth datum 136 using self-supervision where the model learns to predict depth datum 136 without data from imaging modalities for each frame; instead, it may use consistency between frames or other implicit signals in the data to guide the learning process. In some embodiments, electroanatomical mapping derived depth (e.g., depth datum 136) can be used as part of depth training data 168; this may utilize the detailed anatomical maps created by EAM to enhance the model’s accuracy. To illustrate the application of this process, in a non-limiting example, in an echocardiographic procedure where the goal is to segment the left ventricle and determine its spatial characteristics, echo depth map 116 may be created from echo data 112, showing the distances from echo transducer 120 to various points within the heart. A segmentation model 156 may isolate the left ventricle in the echo depth map 116, and for each pixel in the segmented region (e.g., segmented echo depth map 140), the depth datum 136 may be calculated, providing detailed depth information about the ventricle. The depth model 164, trained on 2D echo frames and data from 3D cardiography or EAM, may predict the depth datum 136 for each pixel, creating a new depth map that accurately represents the left ventricle’s structure. This estimated depth map can then be used to create a 3D model of the left ventricle, which can be analyzed for volume, shape, and other characteristics or to generate pose datum 148.

[0104] With continued reference to FIG. 1, memory 108 contains instructions configuring processor 104 to generate a three dimensional (3D) point cloud 144 related to object of interest 124 as a function of depth datum 136. In some embodiments, depth datum 136 may be combined with the spatial coordinates of the segmented pixels (e.g., segmented echo depth map 140) to create a 3D point cloud 144. For the purposes of this disclosure, a “three dimensional point cloud” is a collection of data points in space, each represented by its x, y, and z coordinates. In some embodiments, 3D point cloud 144 may capture the geometry of object of interest 124, providing a comprehensive 3D representation. In some embodiments, the construction of 3D point cloud 144 may integrate depth datum 136 from multiple echo frames. In a non-limiting example, when depth datum 136 (z) is added to pixel coordinates to convert them into 3D points (x, y, z), all the 3D points can be aggregated to form 3D point cloud 144.

[0105] With continued reference to FIG. 1, in some embodiments, processor 104 may be configured to generate a three dimensional (3D) model 170 as a function of 3D point cloud 144. For the purposes of this disclosure, a “three dimensional model” is a digital representation of an object of interest. As a non-limiting example, Processor 104 may be configured to apply one or more 3D reconstruction algorithms, such as without limitation, marching cubes, contour detection and segmentation, active contour models, and / or the like to create a coherent 3D representation e.g., 3D model 170. In some cases, 3D modeling techniques may include surface modeling, solid modeling, or parametric modeling, among others. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various 3D reconstruction algorithms may be used by processor 104 to generate 3D model 170. Additional disclosure related to generating 3D model 170 is described with respect to FIGS. 8-11.

[0106] With continued reference to FIG. 1, memory 108 contains instructions configuring processor 104 to generate a pose datum 148. For the purposes of this disclosure, a “pose datum” is a data element that describes the physical stance, positioning, or orientation of an object in space. In some embodiments, pose datum 148 may include a six degree (6D) pose datum 149. For the purposes of this disclosure, a “six degree pose datum” is a data element that describes an object's position and orientation in three-dimensional space using six parameters: three translational coordinates that indicate the object's location and three rotational coordinates that describe the object's orientation. In some embodiments, pose datum 148 may include a five degree (5D) pose datum 150. A “five degree pose datum,” for the purposes of this disclosure, is a data element that describes an object’s position and orientation in three-dimensional space using five parameters: three translational coordinates and two rotational coordinates that describe the object's orientation. In some embodiments, processor 104 may generate a 5D pose datum 150, opposed to a 6D pose datum 149, for objects that are symmetric along one of their axes. For example, a catheter may be symmetric along its longitudinal axis. For this, a 5D pose datum 150 may be determined as it may be difficult or impossible to determine the catheter’s rotation about its longitudinal axis. In a non-limiting example, in procedures involving surgical navigation, pose datum 148 may provide information about the position and orientation of surgical instruments (e.g., inserting device 128) relative to the patient’s anatomy. During minimally invasive surgeries, accurate pose datum 148 can help surgeons navigate instruments with precision, reducing the risk of damaging surrounding tissues and improving surgical outcomes. For instance, in echocardiographic procedures, knowing the 6D or 5D pose of a catheter tip within the heart may allow for precise targeting of specific cardiac structures, essential for effective interventions. Similarly, accurate pose datum 148 may aid in the diagnosis and treatment planning (e.g., anatomical datum 172) by providing detailed 3D models of organs 126, which can be invaluable for visualizing complex anatomical relationships and planning surgical approaches. In some embodiments, pose datum 148 may be stored in echo database 130. In some embodiments, processor 104 may retrieve pose datum 148 from echo database 130. In some embodiments, user may manually input pose datum 148.

[0107] With continued reference to FIG. 1, in some embodiments, processor 104 may determine pose datum 148 using various methods. In some embodiments, processor 104 may determine pose datum 148 using instance-level 6D or 5D pose estimation by determining pose datum 148 of object of interest 124 using pre-existing CAD models. In a non-limiting example, processor 104 may determine pose datum 148 using red, green, and blue (RGB)-based methods using RGB images to estimate pose datum 148 through various techniques. In some embodiments, RGBbased methods may include regression-based methods, template-based methods, feature-based methods, and the like. As a non-limiting example, regression-based methods may include PoseNet, PoseCNN, Deep-6DPose, YOLO-6D, and the like. PoseNet and PoseCNN may use convolutional neural networks (CNNs) to directly regress the pose datum 148 from RGB images. These methods may predict orientation and position without intermediate keypoint representations. These methods may demonstrate the feasibility of deep learning for pose estimation but often require refinement for higher accuracy. Deep-6DPose may extend Mask R- CNN to include a pose prediction branch, simplifying the process and improving efficiency. nY0L0-6D may transform pose estimation into a keypoint regression task using the YOLO framework, offering real-time performance but limited effectiveness in complex environments. As a non-limiting example, template-based methods may include matching the input image with a set of pre-defined templates. SSD-6D and LatentFusion may use deep learning to extend traditional 2D detection networks to 3D pose estimation, leveraging large datasets of 3D shapes to improve generalization to unseen objects. DPOD may combine detection and matching using a dense matching approach, robust to occlusion and lighting changes. As another non-limiting example, feature-based methods may extract distinctive features (e.g., object feature 132) from the image (e.g., echo data 112), such as scale-invariant feature transform (SIFT) or speeded-up robust features (SURF), and may match them to corresponding features on the CAD model. These methods may extract features from images and match them with a 3D model using algorithms like Perspective-n-Point (PnP). PVNet and BB8 can employ segmentation and keypoint voting to handle occlusion and symmetry. EPOS and Pix2Pose can use deep learning to predict pixel-level 3D coordinates, improving robustness to symmetry and occlusion.

[0108] With continued reference to FIG. 1, as another non-limiting example, deep learningbased methods may involve training convolutional neural networks (CNNs) (e.g., pose estimation model 176) to directly predict pose datum 148 from RGB images. As another nonlimiting example, point cloud or depth-based methods may use 3D point clouds 144 or echo depth maps 116, providing geometric information about the object of interest 124. In a nonlimiting example, algorithms like ICP (Iterative Closest Point) may iteratively align the 3D points from the depth sensor with the CAD model by minimizing the distance between corresponding points. These methods can utilize 3D point clouds to infer object pose. PointNet and PointNet++ may be foundational models that process point clouds directly, extracting global and local features for segmentation and classification. PPFNet can combine point-pair features with deep learning to enhance 3D shape retrieval and matching, while PPR-Net integrates instance segmentation and pose estimation for real-time applications. Depth-based methods may convert depth images into point clouds for pose estimation. SwinDePose can use the Swin Transformer for high accuracy by leveraging depth information, handling occlusions effectively. 0VE6D can decompose pose estimation into viewpoint, in-plane rotation, and translation tasks, suitable for synthetic training data.

[0109] With continued reference to FIG. 1, in some embodiments, processor 104 may determine pose datum 148 using RGB-D-based methods. RGB-D-based methods may include fusion-based Methods. These may combine RGB and depth data to leverage both appearance and geometric information. DenseFusion can extract and fuse features from both modalities, achieving accurate pose estimation through pixel-level voting. MoreFusion can use volumetric maps to represent space occupancy, enabling multi-object pose estimation in occluded scenarios. RGB-D-based methods may include keypoints-based methods. These methods may detect keypoints in objects and establish correspondences for pose prediction. PVN3D can integrate feature extraction, keypoint detection, and semantic segmentation for robust pose estimation in occluded environments. G2L-Net may follow a global-to-local approach, enhancing accuracy by considering rotation residuals and viewpoint perception. GB-D-based methods may include Uni6D that unifies RGB and depth information extraction within a single network framework, achieving high efficiency and approximate accuracy on standard datasets and StablePose that introduces geometric stability for pose estimation, focusing on stable portions of the point cloud to enhance robustness in occluded scenes.

[0110] With continued reference to FIG. 1, in some embodiments, processor 104 may determine pose datum 148 using category-level 6D or 5D pose estimation. In some emb catheter moves 50 mm forward along the depth axis. Additionally, the roll, represented by a 30-degree rotation around the x-axis, indicates odiments, category-level 6D or 5D pose estimation may predict the pose datum 148 of object of interest 124 within a category without requiring known CAD models. As a non-limiting example, regression-based methods may directly regress pose datum 148 from input data, utilizing neural networks to map input images (e.g., echo data 112) or 3D point clouds 144 to pose datum 148, using fully connected layers after feature extraction layers to predict the translation and rotation. Regression-based methods may include NOCS that introduces a Normalized Object Coordinate Space for handling different object instances within a category, robustly estimating pose and size through direct regression and DualPoseNet that combines implicit and explicit pose decoders for consistent pose prediction, utilizing spherical fusion to efficiently learn appearance and shape features. As another non-limiting example, prior-based methods may incorporate prior knowledge about object categories to enhance pose estimation. Prior-based methods may include SPD and ACR-Pose that incorporate prior knowledge to handle intra-class variations, employing adversarial training to reconstruct canonical representations for improved estimation accuracy and DPDN that uses a deep prior deformation network to minimize domain gaps with synthetic data, improving sensitivity to pose changes. As another non-limiting example, anchor-based approaches may use a set of anchor poses as references and may refine the predictions based on the nearest anchors. As another nonlimiting example, latent space models may learn a compact representation of possible object shapes within a category and predict the pose within this latent space. As another non-limiting example, category-level 6D or 5D pose estimation may further include a method that leverages RGB-D images for single-stage object pose and shape estimation using semantic primitives within a generative model and OnePose that constructs object representations from video scans without requiring CAD models, suitable for real-time applications but challenging with untextured objects.

[0111] With continued reference to FIG. 1, in some embodiments, processor 104 may determine pose datum 148 using segmentation-based approaches. In some embodiments, segmentationbased approaches may segment object of interest 124 from the background and occlusions before estimating pose datum 148. In a non-limiting example, estimators like RANSAC can handle outliers and occlusions in the data, ensuring more accurate pose predictions. In some embodiments, sequential models like recurrent neural networks (RNNs) or temporal convolutional networks (TCNs) may maintain consistent pose estimations across frames, leveraging the temporal information to refine predictions. In some embodiments, multi-view approaches, such as multi-view fusion, combine information from multiple views of the object to improve pose accuracy (e.g., a plurality of sets of echo data 112 from a plurality of echo transducers 120 in different angles.

[0112] With continued reference to FIG. 1, in some embodiments, processor 104 may determine pose datum 148 using self-supervised or unsupervised learning methods leveraging unlabeled data to learn useful features for pose estimation. Self-supervised learning approaches may use auxiliary tasks, such as predicting future frames or reconstructing the input image, to train pose estimation models 176 without explicit labels. Unsupervised learning methods, including generative models like GANs (Generative Adversarial Networks), may synthesize pose estimation training data 180 or predict pose datum 148 without explicit labels, reducing the dependency on labeled datasets. In some embodiments, hybrid approaches may combine multiple techniques, such as integrating RGB-based and depth-based methods, to leverage the strengths of each approach.

[0113] With continued reference to FIG. 1, processor 104 is configured to train a pose estimation model 176 using pose estimation training data 180, wherein the pose estimation training data 180 includes exemplary 3D point clouds correlated to exemplary 6D or 5D pose data and generate pose datum 148 using the trained pose estimation model 176. As a non-limiting example, pose estimation training data 180 may include exemplary 3D point clouds correlated to exemplary 6D pose datums. As a non-limiting example, pose estimation training data 180 may include exemplary 3D point clouds correlated to exemplary 5D pose data. For the purposes of this disclosure, a “pose estimation model” is a machine-learning model that generates a 6D or 5D pose datum. For the purposes of this disclosure, “pose estimation training data” is data containing correlations that a machine-learning process may use to model relationships between a 3D point cloud and 6D or 5D pose datum. In some embodiments, pose estimation model 176 may include a regression model and pose estimation training data 180 may include synthetic 3D point clouds generated from a computer aided design (CAD) model of object of interest 124. For the purposes of this disclosure, a “computer aided design model” is a digital representation of an object created using specialized software that enables the design, visualization, and simulation of products in a virtual environment. For the purposes of this disclosure, a “regression model” is a model used in machine learning or data science to predict a continuous outcome variable (dependent variable) based on one or more predictor variables (independent variables). As a nonlimiting example, regression model may include linear regression, polynomial regression, logistic regression, and the like. In a non-limiting example, regression model may be applied to 3D point cloud 144 to estimate pose datum 148, trained on synthetic 3D point clouds and their associated 6D or 5D poses. In some embodiments, rigidity constraints 184 may be applied to ensure that the relative positions and orientations of the cardiac structures remain consistent with anatomical constraints. This step may involve using physical models of the heart's anatomy to guide the pose estimation process. With continued reference to FIG. 1, some embodiments, processor 104 may be configured to generate pose estimation training data 180. In a non-limiting example, pose estimation training data 180 may include correlations between exemplary segmented echo depth maps or exemplary 3D point clouds and exemplary 6D or 5D pose datums. In some embodiments, pose estimation training data 180 may be stored in echo database 130. In some embodiments, pose estimation training data 180 may be received from one or more users, echo database 130, external computing devices, and / or previous iterations of processing. As a nonlimiting example, pose estimation training data 180 may include instructions from a user, who may be an expert user, a past user in embodiments disclosed herein, or the like, which may be stored in memory and / or stored in echo database 130, where the instructions may include labeling of training examples. In some embodiments, pose estimation training data 180 may be updated iteratively on a feedback loop. As a non-limiting example, processor 104 may update pose estimation training data 180 iteratively through a feedback loop as a function of echo data 112, output of feature extraction model, echo data 112, output of depth model 164, output of segmentation model 156, and the like. In some embodiments, processor 104 may be configured to generate pose estimation model 176. In a non-limiting example, generating pose estimation model 176 may include training, retraining, or fine-tuning pose estimation model 176 using pose estimation training data 180 or updated pose estimation training data 180. In some embodiments, processor 104 may be configured to segment echo pose estimation model 176 using pose estimation model 176 (i.e. trained or updated pose estimation model 176). In some embodiments, generating training data and training machine-learning models may be simultaneous.

[0114] With continued reference to FIG. 1, in some embodiments, generating pose datum 148 may include generating the pose datum 148 related to a second object of interest 188 relative to object of interest 124 as a function of a rigidity constraint 184 between object of interest 124 and a second object of interest 188. For the purposes of this disclosure, a “second object of interest” is a particular element or area within an image, dataset, or scene other than an object of interest. As a non-limiting example, object of interest 124 may include a catheter and second object of interest 188 may include an electrode on the catheter. For the purposes of this disclosure, an “electrode” is a conductor through which electric current enters or exits a medium. As another non-limiting example, object of interest 124 may include a distinct shape of an inserting device 128, wherein the inserting device 128 may include a catheter and second object of interest 188 may include an electrode within the catheter. For the purposes of this disclosure, a “distinct shape” of an inserting device is a part of the inserting device that has different or recognizable geometric configuration compared to the other parts of the inserting device. For example, and without limitation, a balloon catheter may feature a cylindrical tube with an inflatable balloon (e g., distinct shape) near the tip. For the purposes of this disclosure, a “rigidity constraint” is a condition imposed on a set of objects or points to ensure that the relative distances between them remain constant over time or transformations. For example, and without limitation, in the case of cardiac structures, the spatial relationships between different parts of the heart may be preserved due to the connective tissue and muscle structure. For example, and without limitation, when estimating pose datum 148 of a rigid body like a catheter within the heart, the distances between electrodes on the catheter may remain constant. For example, and without limitation, inflation balloon part of a balloon catheter and an electrode at a tip of the catheter may have rigidity constraint between them. In some embodiments, rigidity constraint 184 may be stored in echo database 130. In some embodiments, processor 104 may retrieve rigidity constraint 184 from echo database 130. In some embodiments, a user may manually input rigidity constraint 184.

[0115] With continued reference to FIG. 1, in some embodiments, processor 104 may be configured to generate an anatomical datum 172 as a function of pose datum 148, wherein the anatomical datum 172 may include a dimension datum of object of interest 124. For the purposes of this disclosure, an “anatomical datum” is a data element that describes anatomical structures and features of a patient. As a non-limiting example, anatomical datum 172 may include information of anatomical features, including shapes, boundaries, and relationships between different structures. For the purposes of this disclosure, a “dimension datum” is a data element about the size or measurements of anatomical structures. As a non-limiting example, dimension datum may include the size and measurements of abnormal tissue masses, cysts, or lumps. For example, and without limitation, dimension datum may include measurements such as volume, surface area, and linear dimensions (length, width, height) of the anatomical structures. These measurements may be used for assessing the extent of abnormalities and planning appropriate medical interventions.

[0116] With continued reference to FIG. 1, in some embodiments, generating pose datum 148 may include localizing a rigid catheter (object of interest 124) within a body by finding the position of a known sub-part (second object of interest 188) on the catheter. This sub-part may have fixed distances (rigidity constraint 184) from multiple electrodes on the catheter, which do not change. By determining the 3D coordinate and 6D or 5D pose of the sub-part, the pose datum 148 of all electrodes can be calculated, facilitating the generation of an electro-anatomical map. In some embodiments, generating pose datum 148 may include tracking surgically inserted devices. In a non-limiting example, generating pose datum 148 may aid in tracking surgically inserted devices, such as those used in colonoscopy, to navigate and guide the device to specific locations within the anatomy. The accurate 6D or 5D pose estimation may allow for precise control and monitoring of the device's position and orientation, enhancing the safety and effectiveness of the procedure. In some embodiments, generating pose datum 148 may include size / dimension estimation of abnormal tissue masses (e.g., dimension datum of anatomical datum 172). In a non-limiting example, generating pose datum 148 may allow processor 104 to estimate the size and dimensions of abnormal tissue masses, cysts, or lumps in dense organs 126 such as the liver or breast. By generating 3D point clouds 144 and pose datum 148, clinicians can obtain precise measurements, aiding in diagnosis and treatment planning. The size and dimensions of an abnormal tissue mass can be derived by integrating the volume within the segmented region in the 3D point cloud 144. In some embodiments, generating pose datum 148 may incorporate using point-of-care ultrasound (POCUS) with cadaveric organs that contain artificial objects, such as catheters or implanted elements and generating depth datum 136, pose datum 148, voltage, and the like related to the cadaveric organs. For the purposes of this disclosure, a “point-of-care ultrasound” refers to a use of a portable echo transducer at the patient's bedside or in various clinical settings to quickly diagnose, monitor, and guide treatment. For the purposes of this disclosure, a “cadaveric organ” is an organ that has been removed from a deceased body. In some embodiments, processor 104 may receive echo data 112 that is related to a cadaveric organ from echo transducers 120. This data can be used to build deep learning models and algorithms, reducing the need for animal or human studies to validate medical systems. By systematically varying the positions and orientations (pose datum 148) of the implanted objects (object of interest 124), a diverse and extensive training dataset can be created, enhancing the robustness and accuracy of the developed models.

[0117] With continued reference to FIG. 1, processor 104 may be configured to transmit pose datum 148 to a remote device 152 to display pose datum 148, anatomical datum 172, and the like to a user. In some embodiments, at least a processor 104 may be further configured to generate a user interface displaying pose datum 148, 3D model 170, 3D point cloud 144, anatomical datum 172, and the like. For the purposes of this disclosure, a “user interface” is a means by which a user and a computer system interact; for example through the use of input devices and software. In some embodiments, user interface may be displayed on a display device. A “display device,” for the purposes of this disclosure, is a device that presents visual information to a user. A display device may include an LCD, CRT, OLED, LED, plasma, and the like. Display device may include a computer monitor, television, and the like. A user interface may include a graphical user interface (GUI), command line interface (CLI), menu-driven user interface, touch user interface, voice user interface (VUI), form-based user interface, any combination thereof and the like. In some embodiments, user interface may operate on and / or be communicatively connected to a decentralized platform, metaverse, and / or a decentralized exchange platform associated with the user. For example, a user may interact with user interface in virtual reality. In some embodiments, a user may interact with the use interface using a computing device distinct from and communicatively connected to at least a processor 104. For example, a smart phone, smart, tablet, or laptop operated by a user. In an embodiment, user interface may include a graphical user interface. A “graphical user interface,” as used herein, is a graphical form of user interface that allows users to interact with electronic devices. In some embodiments, GUI may include icons, menus, other visual indicators or representations (graphics), audio indicators such as primary notation, and display information and related user controls. A menu may contain a list of choices and may allow users to select one from them. A menu bar may be displayed horizontally across the screen such as pull-down menu. When any option is clicked in this menu, then the pull-down menu may appear. A menu may include a context menu that appears only when the user performs a specific action. An example of this is pressing the right mouse button. When this is done, a menu may appear under the cursor. Files, programs, web pages and the like may be represented using a small picture in a graphical user interface. For example, links to decentralized platforms as described in this disclosure may be incorporated using icons. Using an icon may be a fast way to open documents, run programs etc. because clicking on them yields instant access.

[0118] Referring now to FIG. 2, a configuration of an exemplary user interface 200 of a remote device 152. In some embodiments, user interface 200 may display pose datum 148, a plurality of sets of echo data 112a-b, and the like. As a non-limiting example, plurality of sets of echo data 112a-b may show echo data 112a-b generated by different echo transducers 120 at different angles. As a non-limiting example, as shown in FIG. 2, user interface 200 may display pose datum 148 including shifting the catheter (object of interest 124) along the x, y, and z axes. The x-coordinate, which is 10 mm, indicates that the catheter moves 10 mm to the right of its initial position along the horizontal axis. Next, the y-coordinate, which is 15 mm, signifies that the catheter moves 15 mm upwards along the vertical axis. Finally, the z-coordinate, which is 50 mm, means that the catheter moves 50 mm forward along the depth axis. Additionally, the roll, represented by a 30-degree rotation around the x-axis, indicates how the catheter tilts sideways. The pitch, described by a 45-degree rotation around the y-axis, shows how the catheter tilts upwards or downwards. Finally, the yaw, with a 60-degree rotation around the z-axis, represents how the catheter turns left or right.

[0119] Referring now to FIG. 3, an exemplary embodiment of a machine-learning module 300 that may perform one or more machine-learning processes as described in this disclosure is illustrated. Machine-learning module may perform determinations, classification, and / or analysis steps, methods, processes, or the like as described in this disclosure using machine learning processes. A “machine learning process,” as used in this disclosure, is a process that automatedly uses training data 304 to generate an algorithm instantiated in hardware or software logic, data structures, and / or functions that will be performed by a computing device / module to produce outputs 308 given data provided as inputs 312; this is in contrast to a non-machine learning software program where the commands to be executed are determined in advance by a user and written in a programming language.

[0120] Still referring to FIG. 3, “training data,” as used herein, is data containing correlations that a machine-learning process may use to model relationships between two or more categories of data elements. For instance, and without limitation, training data 304 may include a plurality of data entries, also known as “training examples,” each entry representing a set of data elements that were recorded, received, and / or generated together; data elements may be correlated by shared existence in a given data entry, by proximity in a given data entry, or the like. Multiple data entries in training data 304 may evince one or more trends in correlations between categories of data elements; for instance, and without limitation, a higher value of a first data element belonging to a first category of data element may tend to correlate to a higher value of a second data element belonging to a second category of data element, indicating a possible proportional or other mathematical relationship linking values belonging to the two categories. Multiple categories of data elements may be related in training data 304 according to various correlations; correlations may indicate causative and / or predictive links between categories of data elements, which may be modeled as relationships such as mathematical relationships by machine-learning processes as described in further detail below. Training data 304 may be formatted and / or organized by categories of data elements, for instance by associating data elements with one or more descriptors corresponding to categories of data elements. As a nonlimiting example, training data 304 may include data entered in standardized forms by persons or processes, such that entry of a given data element in a given field in a form may be mapped to one or more descriptors of categories. Elements in training data 304 may be linked to descriptors of categories by tags, tokens, or other data elements; for instance, and without limitation, training data 304 may be provided in fixed-length formats, formats linking positions of data to categories such as comma-separated value (CSV) formats and / or self-describing formats such as extensible markup language (XML), JavaScript Object Notation (JSON), or the like, enabling processes or devices to detect categories of data.

[0121] Alternatively or additionally, and continuing to refer to FIG. 3, training data 304 may include one or more elements that are not categorized; that is, training data 304 may not be formatted or contain descriptors for some elements of data. Machine-learning algorithms and / or other processes may sort training data 304 according to one or more categorizations using, for instance, natural language processing algorithms, tokenization, detection of correlated values in raw data and the like; categories may be generated using correlation and / or other processing algorithms. As a non-limiting example, in a corpus of text, phrases making up a number “n” of compound words, such as nouns modified by other nouns, may be identified according to a statistically significant prevalence of n-grams containing such words in a particular order; such an n-gram may be categorized as an element of language such as a “word” to be tracked similarly to single words, generating a new category as a result of statistical analysis. Similarly, in a data entry including some textual data, a person’s name may be identified by reference to a list, dictionary, or other compendium of terms, permitting ad-hoc categorization by machinelearning algorithms, and / or automated association of data in the data entry with descriptors or into a given format. The ability to categorize data entries automatedly may enable the same training data 304 to be made applicable for two or more distinct machine-learning algorithms as described in further detail below. Training data 304 used by machine-learning module 300 may correlate any input data as described in this disclosure to any output data as described in this disclosure. As a non-limiting illustrative example, input data may include echo data 112, echo depth map 116, segmented echo depth map 140, depth datum 136, 3D point cloud 144, object features 132, and the like. As another non-limiting illustrative example, output data may include echo depth map 116, segmented echo depth map 140, depth datum 136, 3D point cloud 144, pose datum 148, object features 132, anatomical datum 172, and the like.

[0122] Further referring to FIG. 3, training data may be filtered, sorted, and / or selected using one or more supervised and / or unsupervised machine-learning processes and / or models as described in further detail below; such models may include without limitation a training data classifier 316. Training data classifier 316 may include a “classifier,” which as used in this disclosure is a machine-learning model as defined below, such as a data structure representing and / or using a mathematical model, neural net, or program generated by a machine learning algorithm known as a “classification algorithm,” as described in further detail below, that sorts inputs into categories or bins of data, outputting the categories or bins of data and / or labels associated therewith. A classifier may be configured to output at least a datum that labels or otherwise identifies a set of data that are clustered together, found to be close under a distance metric as described below, or the like. A distance metric may include any norm, such as, without limitation, a Pythagorean norm. Machine-learning module 300 may generate a classifier using a classification algorithm, defined as a processes whereby a computing device and / or any module and / or component operating thereon derives a classifier from training data 304. Classification may be performed using, without limitation, linear classifiers such as without limitation logistic regression and / or naive Bayes classifiers, nearest neighbor classifiers such as k-nearest neighbors classifiers, support vector machines, least squares support vector machines, fisher’s linear discriminant, quadratic classifiers, decision trees, boosted trees, random forest classifiers, learning vector quantization, and / or neural network-based classifiers. As a non-limiting example, training data classifier 316 may classify elements of training data to user cohorts. For example, and without limitation, training data classifier 316 may classify elements of training data to user cohorts related to a user’s age, weight, existing conditions, surgery, treatment or medication history, gender, and the like.

[0123] Still referring to FIG. 3, Computing device may be configured to generate a classifier using a Naive Bayes classification algorithm. Naive Bayes classification algorithm generates classifiers by assigning class labels to problem instances, represented as vectors of element values. Class labels are drawn from a finite set. Naive Bayes classification algorithm may include generating a family of algorithms that assume that the value of a particular element is independent of the value of any other element, given a class variable. Naive Bayes classification algorithm may be based on Bayes Theorem expressed as P(A / B)= P(B / A) P(A)^-P(B), where P(A / B) is the probability of hypothesis A given data B also known as posterior probability; P(B / A) is the probability of data B given that the hypothesis A was true; P(A) is the probability of hypothesis A being true regardless of data also known as prior probability of A; and P(B) is the probability of the data regardless of the hypothesis. A naive Bayes algorithm may be generated by first transforming training data into a frequency table. Computing device may then calculate a likelihood table by calculating probabilities of different data entries and classification labels. Computing device may utilize a naive Bayes equation to calculate a posterior probability for each class. A class containing the highest posterior probability is the outcome of prediction. Naive Bayes classification algorithm may include a gaussian model that follows a normal distribution. Naive Bayes classification algorithm may include a multinomial model that is used for discrete counts. Naive Bayes classification algorithm may include a Bernoulli model that may be utilized when vectors are binary.

[0124] With continued reference to FIG. 3, Computing device may be configured to generate a classifier using a K-nearest neighbors (KNN) algorithm. A “K-nearest neighbors algorithm” as used in this disclosure, includes a classification method that utilizes feature similarity to analyze how closely out-of-sample- features resemble training data to classify input data to one or more clusters and / or categories of features as represented in training data; this may be performed by representing both training data and input data in vector forms, and using one or more measures of vector similarity to identify classifications within training data, and to determine a classification of input data. K-nearest neighbors algorithm may include specifying a K-value, or a number directing the classifier to select the k most similar entries training data to a given sample, determining the most common classifier of the entries in the database, and classifying the known sample; this may be performed recursively and / or iteratively to generate a classifier that may be used to classify input data as further samples. For instance, an initial set of samples may be performed to cover an initial heuristic and / or “first guess” at an output and / or relationship, which may be seeded, without limitation, using expert input received according to any process as described herein. As a non-limiting example, an initial heuristic may include a ranking of associations between inputs and elements of training data. Heuristic may include selecting some number of highest-ranking associations and / or training data elements.

[0125] With continued reference to FIG. 3, generating k-nearest neighbors algorithm may generate a first vector output containing a data entry cluster, generating a second vector output containing an input data, and calculate the distance between the first vector output and the second vector output using any suitable norm such as cosine similarity, Euclidean distance measurement, or the like. Each vector output may be represented, without limitation, as an n- tuple of values, where n is at least two values. Each value of n-tuple of values may represent a measurement or other quantitative value associated with a given category of data, or attribute, examples of which are provided in further detail below; a vector may be represented, without limitation, in n-dimensional space using an axis per category of value represented in n-tuple of values, such that a vector has a geometric direction characterizing the relative quantities of attributes in the n-tuple as compared to each other. Two vectors may be considered equivalent where their directions, and / or the relative quantities of values within each vector as compared to each other, are the same; thus, as a non-limiting example, a vector represented as [5, 10, 15] may be treated as equivalent, for purposes of this disclosure, as a vector represented as [1, 2, 3], Vectors may be more similar where their directions are more similar, and more different where their directions are more divergent; however, vector similarity may alternatively or additionally be determined using averages of similarities between like attributes, or any other measure of similarity suitable for any n-tuple of values, or aggregation of numerical similarity measures for the purposes of loss functions as described in further detail below. Any vectors as described herein may be scaled, such that each vector represents each attribute along an equivalent scale of values. Each vector may be “normalized,” or divided by a “length” attribute, such as a length attribute I as derived using a Pythagorean norm: I = SF=oai2where a, is attribute number i of the vector. Scaling and / or normalization may function to make vector comparison independent of absolute quantities of attributes, while preserving any dependency on similarity of attributes; this may, for instance, be advantageous where cases represented in training data are represented by different quantities of samples, which may result in proportionally equivalent vectors with divergent values. With further reference to FIG. 3, training examples for use as training data may be selected from a population of potential examples according to cohorts relevant to an analytical problem to be solved, a classification task, or the like. Alternatively or additionally, training data may be selected to span a set of likely circumstances or inputs for a machine-learning model and / or process to encounter when deployed. For instance, and without limitation, for each category of input data to a machine-learning process or model that may exist in a range of values in a population of phenomena such as images, user data, process data, physical data, or the like, a computing device, processor, and / or machine-learning model may select training examples representing each possible value on such a range and / or a representative sample of values on such a range. Selection of a representative sample may include selection of training examples in proportions matching a statistically determined and / or predicted distribution of such values according to relative frequency, such that, for instance, values encountered more frequently in a population of data so analyzed are represented by more training examples than values that are encountered less frequently. Alternatively or additionally, a set of training examples may be compared to a collection of representative values in a database and / or presented to a user, so that a process can detect, automatically or via user input, one or more values that are not included in the set of training examples. Computing device, processor, and / or module may automatically generate a missing training example; this may be done by receiving and / or retrieving a missing input and / or output value and correlating the missing input and / or output value with a corresponding output and / or input value collocated in a data record with the retrieved value, provided by a user and / or other device, or the like.

[0126] Continuing to refer to FIG. 3, computer, processor, and / or module may be configured to preprocess training data. “Preprocessing” training data, as used in this disclosure, is transforming training data from raw form to a format that can be used for training a machine learning model. Preprocessing may include sanitizing, feature selection, feature scaling, data augmentation and the like.

[0127] Still referring to FIG. 3, computer, processor, and / or module may be configured to sanitize training data. “Sanitizing” training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and / or process to a useful result. For instance, and without limitation, a training example may include an input and / or output value that is an outlier from typically encountered values, such that a machine-learning algorithm using the training example will be adapted to an unlikely amount as an input and / or output; a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated. Alternatively or additionally, one or more training examples may be identified as having poor quality data, where “poor quality” is defined as having a signal to noise ratio below a threshold value. Sanitizing may include steps such as removing duplicative or otherwise redundant data, interpolating missing data, correcting data errors, standardizing data, identifying outliers, and the like. In a nonlimiting example, sanitization may include utilizing algorithms for identifying duplicate entries or spell-check algorithms.

[0128] As a non-limiting example, and with further reference to FIG. 3, images used to train an image classifier or other machine-learning model and / or process that takes images as inputs or generates images as outputs may be rejected if image quality is below a threshold value. For instance, and without limitation, computing device, processor, and / or module may perform blur detection, and eliminate one or more Blur detection may be performed, as a non-limiting example, by taking Fourier transform, or an approximation such as a Fast Fourier Transform (FFT) of the image and analyzing a distribution of low and high frequencies in the resulting frequency-domain depiction of the image; numbers of high-frequency values below a threshold level may indicate blurriness. As a further non-limiting example, detection of blurriness may be performed by convolving an image, a channel of an image, or the like with a Laplacian kernel; this may generate a numerical score reflecting a number of rapid changes in intensity shown in the image, such that a high score indicates clarity and a low score indicates blurriness. Blurriness detection may be performed using a gradient-based operator, which measures operators based on the gradient or first derivative of an image, based on the hypothesis that rapid changes indicate sharp edges in the image, and thus are indicative of a lower degree of blurriness. Blur detection may be performed using Wavelet -based operator, which takes advantage of the capability of coefficients of the discrete wavelet transform to describe the frequency and spatial content of images. Blur detection may be performed using statistics-based operators take advantage of several image statistics as texture descriptors in order to compute a focus level. Blur detection may be performed by using discrete cosine transform (DCT) coefficients in order to compute a focus level of an image from its frequency content.

[0129] Continuing to refer to FIG. 3, computing device, processor, and / or module may be configured to precondition one or more training examples. For instance, and without limitation, where a machine learning model and / or process has one or more inputs and / or outputs requiring, transmitting, or receiving a certain number of bits, samples, or other units of data, one or more training examples’ elements to be used as or compared to inputs and / or outputs may be modified to have such a number of units of data. For instance, a computing device, processor, and / or module may convert a smaller number of units, such as in a low pixel count image, into a desired number of units, for instance by upsampling and interpolating. As a non-limiting example, a low pixel count image may have 100 pixels, however a desired number of pixels may be 128. Processor may interpolate the low pixel count image to convert the 100 pixels into 128 pixels. It should also be noted that one of ordinary skill in the art, upon reading this disclosure, would know the various methods to interpolate a smaller number of data units such as samples, pixels, bits, or the like to a desired number of such units. In some instances, a set of interpolation rules may be trained by sets of highly detailed inputs and / or outputs and corresponding inputs and / or outputs downsampled to smaller numbers of units, and a neural network or other machine learning model that is trained to predict interpolated pixel values using the training data. As a non-limiting example, a sample input and / or output, such as a sample picture, with sample- expanded data units (e.g., pixels added between the original pixels) may be input to a neural network or machine-learning model and output a pseudo replica sample-picture with dummy values assigned to pixels between the original pixels based on a set of interpolation rules. As a non-limiting example, in the context of an image classifier, a machine-learning model may have a set of interpolation rules trained by sets of highly detailed images and images that have been downsampled to smaller numbers of pixels, and a neural network or other machine learning model that is trained using those examples to predict interpolated pixel values in a facial picture context. As a result, an input with sample-expanded data units (the ones added between the original data units, with dummy values) may be run through a trained neural network and / or model, which may fill in values to replace the dummy values. Alternatively or additionally, processor, computing device, and / or module may utilize sample expander methods, a low-pass filter, or both. As used in this disclosure, a “low-pass filter” is a filter that passes signals with a frequency lower than a selected cutoff frequency and attenuates signals with frequencies higher than the cutoff frequency. The exact frequency response of the filter depends on the filter design. Computing device, processor, and / or module may use averaging, such as luma or chroma averaging in images, to fill in data units in between original data units.

[0130] In some embodiments, and with continued reference to FIG. 3, computing device, processor, and / or module may down-sample elements of a training example to a desired lower number of data elements. As a non-limiting example, a high pixel count image may have 256 pixels, however a desired number of pixels may be 128. Processor may down-sample the high pixel count image to convert the 256 pixels into 128 pixels. In some embodiments, processor may be configured to perform downsampling on data. Downsampling, also known as decimation, may include removing every Nth entry in a sequence of samples, all but every Nth entry, or the like, which is a process known as “compression,” and may be performed, for instance by an N-sample compressor implemented using hardware or software. Anti-aliasing and / or anti-imaging filters, and / or low-pass filters, may be used to clean up side-effects of compression.

[0131] Further referring to FIG. 3, feature selection includes narrowing and / or filtering training data to exclude features and / or elements, or training data including such elements, that are not relevant to a purpose for which a trained machine-learning model and / or algorithm is being trained, and / or collection of features and / or elements, or training data including such elements, on the basis of relevance or utility for an intended task or purpose for a trained machine-learning model and / or algorithm is being trained. Feature selection may be implemented, without limitation, using any process described in this disclosure, including without limitation using training data classifiers, exclusion of outliers, or the like.

[0132] With continued reference to FIG. 3, feature scaling may include, without limitation, normalization of data entries, which may be accomplished by dividing numerical fields by norms thereof, for instance as performed for vector normalization. Feature scaling may include absolute maximum scaling, wherein each quantitative datum is divided by the maximum absolute value of all quantitative data of a set or subset of quantitative data. Feature scaling may include min-max scaling, in which each value X has a minimum value Xminin a set or subset of values subtracted therefrom, with the result divided by the range of the values, give maximum value in the set or subset XmaxXnew= -mm. Feature scaling may include mean normalization, which

[0133] ^max~ min involves use of a mean value of a set and / or subset of values, Xmeanwith maximum and — minimum values: Xnew= -mean. Feature scaling may include standardization, where a

[0134] ^max~^min difference between X and Xmeanis divided by a standard deviation a of a set or subset of values: Xnew= —^ Scaling may be performed using a median value of a set or subset Xmedianand / or interquartile range (IQR), which represents the difference between the 25thpercentile value and the 50111percentile value (or closest values thereto by a rounding protocol), such as: Xnew=Persons skilled in the art, upon reviewing the entirety of this disclosure, will IQR be aware of various alternative or additional approaches that may be used for feature scaling.

[0135] Further referring to FIG. 3, computing device, processor, and / or module may be configured to perform one or more processes of data augmentation. “Data augmentation” as used in this disclosure is addition of data to a training set using elements and / or entries already in the dataset. Data augmentation may be accomplished, without limitation, using interpolation, generation of modified copies of existing entries and / or examples, and / or one or more generative Al processes, for instance using deep neural networks and / or generative adversarial networks; generative processes may be referred to alternatively in this context as “data synthesis” and as creating “synthetic data.” Augmentation may include performing one or more transformations on data, such as geometric, color space, affine, brightness, cropping, and / or contrast transformations of images.

[0136] Still referring to FIG. 3, machine-learning module 300 may be configured to perform a lazy-learning process 320 and / or protocol, which may alternatively be referred to as a “lazy loading” or “call-when-needed” process and / or protocol, may be a process whereby machine learning is conducted upon receipt of an input to be converted to an output, by combining the input and training set to derive the algorithm to be used to produce the output on demand. For instance, an initial set of simulations may be performed to cover an initial heuristic and / or “first guess” at an output and / or relationship. As a non-limiting example, an initial heuristic may include a ranking of associations between inputs and elements of training data 304. Heuristic may include selecting some number of highest-ranking associations and / or training data 304 elements. Lazy learning may implement any suitable lazy learning algorithm, including without limitation a K-nearest neighbors algorithm, a lazy naive Bayes algorithm, or the like; persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various lazy- learning algorithms that may be applied to generate outputs as described in this disclosure, including without limitation lazy learning applications of machine-learning algorithms as described in further detail below.

[0137] Alternatively or additionally, and with continued reference to FIG. 3, machine-learning processes as described in this disclosure may be used to generate machine-learning models 324. A “machine-learning model,” as used in this disclosure, is a data structure representing and / or instantiating a mathematical and / or algorithmic representation of a relationship between inputs and outputs, as generated using any machine-learning process including without limitation any process as described above, and stored in memory; an input is submitted to a machine-learning model 324 once created, which generates an output based on the relationship that was derived. For instance, and without limitation, a linear regression model, generated using a linear regression algorithm, may compute a linear combination of input data using coefficients derived during machine-learning processes to calculate an output datum. As a further non-limiting example, a machine-learning model 324 may be generated by creating an artificial neural network, such as a convolutional neural network comprising an input layer of nodes, one or more intermediate layers, and an output layer of nodes. Connections between nodes may be created via the process of "training" the network, in which elements from a training data 304 set are applied to the input nodes, a suitable training algorithm (such as Levenberg-Marquardt, conjugate gradient, simulated annealing, or other algorithms) is then used to adjust the connections and weights between nodes in adjacent layers of the neural network to produce the desired values at the output nodes. This process is sometimes referred to as deep learning.

[0138] Still referring to FIG. 3, machine-learning algorithms may include at least a supervised machine-learning process 328. At least a supervised machine-learning process 328, as defined herein, include algorithms that receive a training set relating a number of inputs to a number of outputs, and seek to generate one or more data structures representing and / or instantiating one or more mathematical relations relating inputs to outputs, where each of the one or more mathematical relations is optimal according to some criterion specified to the algorithm using some scoring function. For instance, a supervised learning algorithm may include echo data 112, echo depth map 116, segmented echo depth map 140, depth datum 136, 3D point cloud 144, object features 132, and the like as described above as inputs, echo depth map 116, segmented echo depth map 140, depth datum 136, 3D point cloud 144, pose datum 148, object features 132, anatomical datum 172, and the like as outputs, and a scoring function representing a desired form of relationship to be detected between inputs and outputs; scoring function may, for instance, seek to maximize the probability that a given input and / or combination of elements inputs is associated with a given output to minimize the probability that a given input is not associated with a given output. Scoring function may be expressed as a risk function representing an “expected loss” of an algorithm relating inputs to outputs, where loss is computed as an error function representing a degree to which a prediction generated by the relation is incorrect when compared to a given input-output pair provided in training data 304. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various possible variations of at least a supervised machine-learning process 328 that may be used to determine relation between inputs and outputs. Supervised machine-learning processes may include classification algorithms as defined above.

[0139] With further reference to FIG. 3, training a supervised machine-learning process may include, without limitation, iteratively updating coefficients, biases, weights based on an error function, expected loss, and / or risk function. For instance, an output generated by a supervised machine-learning model using an input example in a training example may be compared to an output example from the training example; an error function may be generated based on the comparison, which may include any error function suitable for use with any machine-learning algorithm described in this disclosure, including a square of a difference between one or more sets of compared values or the like. Such an error function may be used in turn to update one or more weights, biases, coefficients, or other parameters of a machine-learning model through any suitable process including without limitation gradient descent processes, least-squares processes, and / or other processes described in this disclosure. This may be done iteratively and / or recursively to gradually tune such weights, biases, coefficients, or other parameters. Updating may be performed, in neural networks, using one or more back-propagation algorithms. Iterative and / or recursive updates to weights, biases, coefficients, or other parameters as described above may be performed until currently available training data is exhausted and / or until a convergence test is passed, where a “convergence test” is a test for a condition selected as indicating that a model and / or weights, biases, coefficients, or other parameters thereof has reached a degree of accuracy. A convergence test may, for instance, compare a difference between two or more successive errors or error function values, where differences below a threshold amount may be taken to indicate convergence. Alternatively or additionally, one or more errors and / or error function values evaluated in training iterations may be compared to a threshold. Still referring to FIG. 3, a computing device, processor, and / or module may be configured to perform method, method step, sequence of method steps and / or algorithm described in reference to this figure, in any order and with any degree of repetition. For instance, a computing device, processor, and / or module may be configured to perform a single step, sequence and / or algorithm repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and / or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and / or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and / or division of a larger processing task into a set of iteratively addressed smaller processing tasks. A computing device, processor, and / or module may perform any step, sequence of steps, or algorithm in parallel, such as simultaneously and / or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and / or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and / or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and / or parallel processing.

[0140] Further referring to FIG. 3, machine learning processes may include at least an unsupervised machine-learning processes 332. An unsupervised machine-learning process, as used herein, is a process that derives inferences in datasets without regard to labels; as a result, an unsupervised machine-learning process may be free to discover any structure, relationship, and / or correlation provided in the data. Unsupervised processes 332 may not require a response variable; unsupervised processes 332may be used to find interesting patterns and / or inferences between variables, to determine a degree of correlation between two or more variables, or the like.

[0141] Still referring to FIG. 3, machine-learning module 300 may be designed and configured to create a machine-learning model 324 using techniques for development of linear regression models. Linear regression models may include ordinary least squares regression, which aims to minimize the square of the difference between predicted outcomes and actual outcomes according to an appropriate norm for measuring such a difference (e.g. a vector-space distance norm); coefficients of the resulting linear equation may be modified to improve minimization. Linear regression models may include ridge regression methods, where the function to be minimized includes the least-squares function plus term multiplying the square of each coefficient by a scalar amount to penalize large coefficients. Linear regression models may include least absolute shrinkage and selection operator (LASSO) models, in which ridge regression is combined with multiplying the least-squares term by a factor of 1 divided by double the number of samples. Linear regression models may include a multi-task lasso model wherein the norm applied in the least-squares term of the lasso model is the Frobenius norm amounting to the square root of the sum of squares of all terms. Linear regression models may include the elastic net model, a multi-task elastic net model, a least angle regression model, a LARS lasso model, an orthogonal matching pursuit model, a Bayesian regression model, a logistic regression model, a stochastic gradient descent model, a perceptron model, a passive aggressive algorithm, a robustness regression model, a Huber regression model, or any other suitable model that may occur to persons skilled in the art upon reviewing the entirety of this disclosure. Linear regression models may be generalized in an embodiment to polynomial regression models, whereby a polynomial equation (e.g. a quadratic, cubic or higher-order equation) providing a best predicted output / actual output fit is sought; similar methods to those described above may be applied to minimize error functions, as will be apparent to persons skilled in the art upon reviewing the entirety of this disclosure.

[0142] Continuing to refer to FIG. 3, machine-learning algorithms may include, without limitation, linear discriminant analysis. Machine-learning algorithm may include quadratic discriminant analysis. Machine-learning algorithms may include kernel ridge regression. Machine-learning algorithms may include support vector machines, including without limitation support vector classification-based regression processes. Machine-learning algorithms may include stochastic gradient descent algorithms, including classification and regression algorithms based on stochastic gradient descent. Machine-learning algorithms may include nearest neighbors algorithms. Machine-learning algorithms may include various forms of latent space regularization such as variational regularization. Machine-learning algorithms may include Gaussian processes such as Gaussian Process Regression. Machine-learning algorithms may include cross-decomposition algorithms, including partial least squares and / or canonical correlation analysis. Machine-learning algorithms may include naive Bayes methods. Machinelearning algorithms may include algorithms based on decision trees, such as decision tree classification or regression algorithms. Machine-learning algorithms may include ensemble methods such as bagging m eta-estimator, forest of randomized trees, AdaBoost, gradient tree boosting, and / or voting classifier methods. Machine-learning algorithms may include neural net algorithms, including convolutional neural net processes.

[0143] Still referring to FIG. 3, a machine-learning model and / or process may be deployed or instantiated by incorporation into a program, apparatus, system and / or module. For instance, and without limitation, a machine-learning model, neural network, and / or some or all parameters thereof may be stored and / or deployed in any memory or circuitry. Parameters such as coefficients, weights, and / or biases may be stored as circuit-based constants, such as arrays of wires and / or binary inputs and / or outputs set at logic “1” and “0” voltage levels in a logic circuit to represent a number according to any suitable encoding system including twos complement or the like or may be stored in any volatile and / or non-volatile memory. Similarly, mathematical operations and input and / or output of data to or from models, neural network layers, or the like may be instantiated in hardware circuitry and / or in the form of instructions in firmware, machine-code such as binary operation code instructions, assembly language, or any higher- order programming language. Any technology for hardware and / or software instantiation of memory, instructions, data structures, and / or algorithms may be used to instantiate a machinelearning process and / or model, including without limitation any combination of production and / or configuration of non-reconfigurable hardware elements, circuits, and / or modules such as without limitation ASICs, production and / or configuration of reconfigurable hardware elements, circuits, and / or modules such as without limitation FPGAs, production and / or of non- reconfigurable and / or configuration non-rewritable memory elements, circuits, and / or modules such as without limitation non-rewritable ROM, production and / or configuration of reconfigurable and / or rewritable memory elements, circuits, and / or modules such as without limitation rewritable ROM or other memory technology described in this disclosure, and / or production and / or configuration of any computing device and / or component thereof as described in this disclosure. Such deployed and / or instantiated machine-learning model and / or algorithm may receive inputs from any other process, module, and / or component described in this disclosure, and produce outputs to any other process, module, and / or component described in this disclosure.

[0144] Continuing to refer to FIG. 3, any process of training, retraining, deployment, and / or instantiation of any machine-learning model and / or algorithm may be performed and / or repeated after an initial deployment and / or instantiation to correct, refine, and / or improve the machinelearning model and / or algorithm. Such retraining, deployment, and / or instantiation may be performed as a periodic or regular process, such as retraining, deployment, and / or instantiation at regular elapsed time periods, after some measure of volume such as a number of bytes or other measures of data processed, a number of uses or performances of processes described in this disclosure, or the like, and / or according to a software, firmware, or other update schedule. Alternatively or additionally, retraining, deployment, and / or instantiation may be event-based, and may be triggered, without limitation, by user inputs indicating sub-optimal or otherwise problematic performance and / or by automated field testing and / or auditing processes, which may compare outputs of machine-learning models and / or algorithms, and / or errors and / or error functions thereof, to any thresholds, convergence tests, or the like, and / or may compare outputs of processes described herein to similar thresholds, convergence tests or the like. Event-based retraining, deployment, and / or instantiation may alternatively or additionally be triggered by receipt and / or generation of one or more new training examples; a number of new training examples may be compared to a preconfigured threshold, where exceeding the preconfigured threshold may trigger retraining, deployment, and / or instantiation.

[0145] Still referring to FIG. 3, retraining and / or additional training may be performed using any process for training described above, using any currently or previously deployed version of a machine-learning model and / or algorithm as a starting point. Training data for retraining may be collected, preconditioned, sorted, classified, sanitized or otherwise processed according to any process described in this disclosure. Training data may include, without limitation, training examples including inputs and correlated outputs used, received, and / or generated from any version of any system, module, machine-learning model or algorithm, apparatus, and / or method described in this disclosure; such examples may be modified and / or labeled according to user feedback or other processes to indicate desired results, and / or may have actual or measured results from a process being modeled and / or predicted by system, module, machine-learning model or algorithm, apparatus, and / or method as “desired” results to be compared to outputs for training processes as described above.

[0146] Redeployment may be performed using any reconfiguring and / or rewriting of reconfigurable and / or rewritable circuit and / or memory elements; alternatively, redeployment may be performed by production of new hardware and / or software components, circuits, instructions, or the like, which may be added to and / or may replace existing hardware and / or software components, circuits, instructions, or the like.

[0147] Further referring to FIG. 3, one or more processes or algorithms described above may be performed by at least a dedicated hardware unit 336. A “dedicated hardware unit,” for the purposes of this figure, is a hardware component, circuit, or the like, aside from a principal control circuit and / or processor performing method steps as described in this disclosure, that is specifically designated or selected to perform one or more specific tasks and / or processes described in reference to this figure, such as without limitation preconditioning and / or sanitization of training data and / or training a machine-learning algorithm and / or model. A dedicated hardware unit 336 may include, without limitation, a hardware unit that can perform iterative or massed calculations, such as matrix-based calculations to update or tune parameters, weights, coefficients, and / or biases of machine-learning models and / or neural networks, efficiently using pipelining, parallel processing, or the like; such a hardware unit may be optimized for such processes by, for instance, including dedicated circuitry for matrix and / or signal processing operations that includes, e.g., multiple arithmetic and / or logical circuit units such as multipliers and / or adders that can act simultaneously and / or in parallel or the like. Such dedicated hardware units 336 may include, without limitation, graphical processing units (GPUs), dedicated signal processing modules, FPGA or other reconfigurable hardware that has been configured to instantiate parallel processing units for one or more specific tasks, or the like, A computing device, processor, apparatus, or module may be configured to instruct one or more dedicated hardware units 336 to perform one or more operations described herein, such as evaluation of model and / or algorithm outputs, one-time or iterative updates to parameters, coefficients, weights, and / or biases, and / or any other operations such as vector and / or matrix operations as described in this disclosure.

[0148] Referring now to FIG. 4, an exemplary embodiment of neural network 400 is illustrated. A neural network 400 also known as an artificial neural network, is a network of “nodes,” or data structures having one or more inputs, one or more outputs, and a function determining outputs based on inputs. Such nodes may be organized in a network, such as without limitation a convolutional neural network, including an input layer of nodes 404, one or more intermediate layers 408, and an output layer of nodes 412. Connections between nodes may be created via the process of "training" the network, in which elements from a training dataset are applied to the input nodes, a suitable training algorithm (such as Levenberg-Marquardt, conjugate gradient, simulated annealing, or other algorithms) is then used to adjust the connections and weights between nodes in adjacent layers of the neural network to produce the desired values at the output nodes. This process is sometimes referred to as deep learning. Connections may run solely from input nodes toward output nodes in a “feed-forward” network, or may feed outputs of one layer back to inputs of the same or a different layer in a “recurrent network.” As a further nonlimiting example, a neural network may include a convolutional neural network comprising an input layer of nodes, one or more intermediate layers, and an output layer of nodes. A “convolutional neural network,” as used in this disclosure, is a neural network in which at least one hidden layer is a convolutional layer that convolves inputs to that layer with a subset of inputs known as a “kernel,” along with one or more additional layers such as pooling layers, fully connected layers, and the like.

[0149] Referring now to FIG. 5, an exemplary embodiment of a node 500 of a neural network is illustrated. A node may include, without limitation, a plurality of inputs xt that may receive numerical values from inputs to a neural network containing the node and / or from other nodes. Node may perform one or more activation functions to produce its output given one or more inputs, such as without limitation computing a binary step function comparing an input to a threshold value and outputting either a logic 1 or logic 0 output or something equivalent, a linear activation function whereby an output is directly proportional to the input, and / or a non-linear activation function, wherein the output is not proportional to the input. Non-linear activation functions may include, without limitation, a sigmoid function of the form (%) = -j— — given eX_e-X input x, a tanh (hyperbolic tangent) function, of the formeX+e-x, a tanh derivative function such as f (x) = tanh2(x), a rectified linear unit function such as f(x) = max (0, x), a “leaky” and / or “parametric” rectified linear unit function such as (x) = max (ax, x) for some a, an exponential linear units function such as Q forsomevalue of a (this function may be replaced and / or weighted by its own derivative in some embodiments), a softmax function such as f(xt) — where the inputs to an instant layer are xt, a swish function such as (x) = x * sigmoid(x), a Gaussian error linear unit function such as f(x) = a(l + tanh for some values of a, b, and r, and / or a scaled exponential linear unit function such as . Fundamentally, there is no limit to the nature of functions of inputs Xi that may be used as activation functions. As a non-limiting and illustrative example, node may perform a weighted sum of inputs using weights w, that are multiplied by respective inputs;. Additionally or alternatively, a bias b may be added to the weighted sum of the inputs such that an offset is added to each unit in the neural network layer that is independent of the input to the layer. The weighted sum may then be input into a function (p, which may generate one or more outputs y. Weight w> applied to an input x, may indicate whether the input is “excitatory,” indicating that it has strong influence on the one or more outputs^, for instance by the corresponding weight having a large numerical value, and / or a “inhibitory,” indicating it has a weak effect influence on the one more inputs y, for instance by the corresponding weight having a small numerical value. The values of weights wt may be determined by training a neural network using training data, which may be performed using any suitable process as described above.

[0150] Referring now to FIG. 6, a configuration of exemplary echo transducers 120a-c examining an object of interest 124 and / or second object of interest 188 within a body though a surface of a body 600. As a non-limiting example, echo transducers 120a-c may examine object of interest 124 and / or second object of interest 188 through surface of chest, pelvic, abdomen, and the like. In some embodiments, organ within surface of body 600 may include a cadaveric organ. In some embodiments, each of the plurality of echo transducers 120a-c may examine and capture a plurality of sets of echo data 112 of an object of interest 124 and / or second object of interest 188 in different angles or views; therefore, processor 104 may generate a plurality of echo depth maps 116 that has different angles or views of object of interest 124 and / or second object of interest 188. In some embodiments, one echo transducer 120 may capture a plurality of sets of echo data 112 of an object of interest 124 and / or second object of interest 188 in different angles or views. In some embodiments, object of interest 124 may include a catheter 604. As another non-limiting example, object of interest 124 may include a distinct shape 608 of a catheter 604 and second object of interest 188 may include an electrode 612 within the catheter 604. For example, and without limitation, a balloon catheter may feature a cylindrical tube with an inflatable balloon (e.g., distinct shape 608) near the tip.

[0151] Referring now to FIG. 7, a flow diagram of an exemplary method 700 for object pose estimation in a medical image is illustrated. Method 700 contains a step 705 of receiving, using at least a processor, a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps. In some embodiments, receiving the plurality of sets of echo data may include receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers may be located at a different angle. In some embodiments, the plurality of sets of echo data may be related to a cadaveric organ with an inserting device of a catheter inserted into it. These may be implemented as disclosed, without limitation, with reference to FIGS. 1-6 and 8-11.

[0152] With continued reference to FIG. 7, method 700 contains a step 710 of segmenting, using at least a processor, a plurality of echo depth maps. In some embodiments, segmenting the plurality of echo depth maps may include extracting object features of the object of interest from the plurality of echo depth maps and segmenting the plurality of echo depth maps as a function of the object features. These may be implemented as disclosed, without limitation, with reference to FIGS. 1-6 and 8-11.

[0153] With continued reference to FIG. 7, method 700 contains a step 715 of determining, using at least a processor, a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps. In some embodiments, determining the depth datum may include training a depth model using depth training data, wherein the depth training data may include exemplary segmented echo depth maps correlated to exemplary depth datums and determining the depth datum using the trained depth model. These may be implemented as disclosed, without limitation, with reference to FIGS. 1-6 and 8-11.

[0154] With continued reference to FIG. 7, method 700 contains a step 720 of generating, using at least a processor, a three dimensional (3D) point cloud related to an object of interest as a function of a depth datum. These may be implemented as disclosed, without limitation, with reference to FIGS. 1-6 and 8-11.

[0155] With continued reference to FIG. 7, method 700 contains a step 725 of generating, using at least a processor, a pose datum of an object of interest as a function of a 3D point cloud, wherein generating the pose datum includes training a pose estimation model using pose estimation training data, wherein the pose estimation training data includes exemplary 3D point clouds correlated to exemplary pose datums and generating the pose datum using the trained pose estimation model. In some embodiments, the pose estimation model may include a regression model and the pose estimation training data may include synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest. In some embodiments, generating the pose datum may include generating the pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest. In some embodiments, pose datum may include a 6D pose datum. In some embodiments, pose datum may include a 5D pose datum. In some embodiments, the object of interest of a catheter may include a distinct shape and an electrode of the catheter may include the second object of interest. In some embodiments, method 700 may further include generating, using the at least a processor, a 3D model as a function of the 3D point cloud and generating, using the at least a processor, a user interface displaying the 3D model to a remote device. In some embodiments, method 700 may further include generating, using the at least a processor, an anatomical datum as a function of the pose datum, wherein the anatomical datum may include a dimension datum of the object of interest. These may be implemented as disclosed, without limitation, with reference to FIGS. 1-6 and 8- 11.

[0156] Referring now to FIG. 8, an exemplary embodiment of an apparatus 800 for generating 3D model of a cardiac anatomy via machine-learning is illustrated. System includes at least a processor 804. Processor 804 may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and / or system on a chip (SoC) as described in this disclosure. Computing device may include, be included in, and / or communicate with a mobile device such as a mobile telephone or smartphone. Processor 804 may include a single computing device operating independently, or may include two or more computing device operating in concert, in parallel, sequentially or the like; two or more computing devices may be included together in a single computing device or in two or more computing devices. Processor 804 may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting processor 804 to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g, a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone / voice provider ( .g., a mobile communications provider data and / or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and / or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and / or from a computer and / or a computing device. Processor 804 may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Processor 804 may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Processor 804 may distribute one or more computing tasks as described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing devices. Processor 804 may be implemented using a “shared nothing” architecture in which data is cached at the worker, in an embodiment, this may enable scalability of system and / or computing device.

[0157] With continued reference to FIG. 8, processor 804 may be designed and / or configured to perform any method, method step, or sequence of method steps in any embodiment described in this disclosure, in any order and with any degree of repetition. For instance, processor 804 may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and / or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and / or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and / or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Processor 804 may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and / or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and / or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and / or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and / or parallel processing.

[0158] With continued reference to FIG. 8, apparatus includes a memory 808 communicatively connected to at least a processor 804, wherein the memory 808 contains instructions configuring at least a processor 804 to perform any processing steps described herein. As used in this disclosure, “communicatively connected” means connected by way of a connection, attachment or linkage between two or more relata which allows for reception and / or transmittance of information therebetween. For example, and without limitation, this connection may be wired or wireless, direct or indirect, and between two or more components, circuits, devices, systems, and the like, which allows for reception and / or transmittance of data and / or signal(s) therebetween. Data and / or signals therebetween may include, without limitation, electrical, electromagnetic, magnetic, video, audio, radio and microwave data and / or signals, combinations thereof, and the like, among others. A communicative connection may be achieved, for example and without limitation, through wired or wireless electronic, digital or analog, communication, either directly or by way of one or more intervening devices or components. Further, communicative connection may include electrically coupling or connecting at least an output of one device, component, or circuit to at least an input of another device, component, or circuit. For example, and without limitation, via a bus or other facility for intercommunication between elements of a computing device. Communicative connecting may also include indirect connections via, for example and without limitation, wireless connection, radio communication, low power wide area network, optical communication, magnetic, capacitive, or optical coupling, and the like. In some instances, the terminology “communicatively coupled” may be used in place of communicatively connected in this disclosure.

[0159] With continued reference to FIG. 8, processor 804 may perform determinations, classification, and / or analysis steps, methods, processes, or the like as described in this disclosure using machine-learning processes. A “machine-learning process,” as used in this disclosure, is a process that automatedly uses a body of data known as “training data” and / or a “training set” (described further below in this disclosure) to generate an algorithm that will be performed by a processor 804 / module to produce outputs given data provided as inputs; this is in contrast to a non-machine learning software program where the commands to be executed are determined in advance by a user and written in a programming language. Machine-learning process may utilize supervised, unsupervised, lazy-leaming processes and / or neural networks, described further below.

[0160] With continued reference to FIG. 8, processor is configured to receive a set of images 812 of a cardiac anatomy 816 pertaining to a subject 820. As used in this disclosure, a “set of images” refers to a collection or group of visual representations captured using any imaging modality or technique described herein. Set of images 812 may include, without limitation, two- dimensional images. In an embodiment, set of images 812 may include a set of intracardiac echocardiography (ICE) images, wherein the "set of ICE images” is a collection of ultrasound images obtained from within the heart’s chambers or blood vessels. In some cases, ICE images may be captured using a specialized catheter equipped with an ultrasound transducer that is inserted into the body and guided to the heart of subject 820. In an embodiment, set of images 812 may provide a detailed and real-time visualizations of “cardiac anatomy,” which refers to the structural composition of the heart and its associated blood vessels. Set of images 812 may also include internal structures, functions, and bold flow patterns of the heart of subject 820. Other exemplary embodiments of set of images 812 may include, without limitation, X-ray images, magnetic resonance imaging (MRI) scans, computed tomography (CT) scans, ultrasound images, optical images, digital photographs, or any other form of visual data. Additionally, images within set of images 812 may be related in terms of content, time of capture, sequence, or any other relevant parameters described herein. In a non-limiting example, each image of set of images 812 may represent a particular view, angle, or perspective of an object, subject, or scene, and may be in two-dimensional (2D) or 3D format. Images of set of images 812 may include, without limitation, any two-dimensional or three-dimensional images of any anatomy or anatomical structure, including without limitation images of any internal organ, tissue including without limitation muscular, connective tissue, epithelial tissue, and / or nervous tissue, bone, and / or any other element that may be imaged within a human and / or animal body.

[0161] Still referring to FIG. 8, in a non-limiting example, cardiac anatomy 816 may include chambers (e.g., four chambers including left and right atria and left and right ventricles), valves (i.e., the structures that regulate blood flow between chambers and vessels, including mitral, tricuspid, aortic, and pulmonary valves), vessels (e.g., aorta, pulmonary arteries and veins, and coronary arteries), conduction system (i.e., a network of specialized cells that control the heart’s electrical activity and rhythm), muscular and connective tissues (e.g., heart’s muscular walls, septa, any other connective tissues that provide structural integrity and enable contraction), LAA and other appendages, pathological features (e.g., any abnormalities, defects, and / or the like), among others.

[0162] Still referring to FIG. 8, as used in this disclosure, a “subject” refers to an individual organism. In an embodiment, subject 820 may include a human, on whom or on which the procedure, study, or otherwise experiment, such as without limitation, AF ablation described herein, is being conducted. In some cases, subject 820 may include a provider of set of images 812 described herein. In other cases, subject 820 may include a recipient or a participant in a clinical trial or research study. In a non-limiting example, subject 820 may include a human patient with AF who is undergoing a procedure, an individual undergoing cardiac screening, a participant in a clinical trial, patient with congenital heart disease, heart transplant candidate, patient receiving follow-up care after cardiac surgery, healthy volunteer, patient with heart failure, or the like. Additionally, or alternatively, subject 820 may include an animal models (i.e., animal used to model AF such as a laboratory rat).

[0163] Still referring to FIG. 8, in an embodiment, each ICE image of set of ICE images may include a particular view of subject’s 820 heart’s chambers, valves, vessel, and / or the like. In a non-limiting example, set of images 812 may include multiple views e.g., different angles and perspectives of subject’s 820 heart. In another embodiment, set of images 812 may be arranged in a temporal sequence. In a non-limiting example, set of images 812 may include a series of images captured over time, allowing for an observation of dynamic cardiac functions such as beating, blood flow, and / or the like. In some cases, each ICE image of set of images 812 may include a corresponding timestamp, wherein the timestamp may include an indicator showing a date and time of when the corresponding ICE image was taken.

[0164] Additionally, or alternatively, and still referring to FIG. 8, various imaging techniques or settings may be applied to set of images 812 that provide specific insights into cardiac anatomy 816. In some cases, cardiac anatomy 816 may include a plurality of physical characteristics, spatial relationships, and function aspects of the heart’s component; for instance, and without limitation, receiving set of images 812 may include applying a doppler imaging technique, wherein the “doppler imaging technique” is a specialized ultrasound technique used to assess the movement of blood within the body, particularly within the heart and blood vessels. Processor 804 may configure a transducer to send high-frequency sound waves into the subject’s 820 body, wherein the sound waves may bounce off moving blood cells and other structures. Reflected waves may then be picked up by the transducer and frequency of the reflected waves changes (Doppler shift) depending on the speed and direction of blood flow may be analyzed to determine one or more blood flow characteristics. In some cases, one or more ICE images within set of images 812 may include visual representations translated based on one or more blood flow characteristics. Such visual representations may be further color-coded, showing the speed and direction of blood flow. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will aware other exemplary modalities of ICE imaging such as, without limitation, computed tomography (CT) scans, magnetic resonance imaging MRI, positron emission tomography (PET) scan, angiography, electrocardiogram (ECG or EKG), single-photon emission computed tomography (SPECT), optical coherence tomography (OCT), thermography, tactile imaging, and / or the like.

[0165] With continued reference to FIG. 8, in one or more embodiments, receiving set of images 812 of cardiac anatomy 816 may include receiving a patient profile pertaining to subject 820. As used in this disclosure, a “patient profile” is a comprehensive collection of information related to an individual patient. In some cases, patient profile may include a variety of different types of data that, when combined, provide a detailed picture of a patient's overall health. In an embodiment, patient profile may include demographic data of patient, for example, and without limitation, patient profile may include basic information about the patient such as name, age, gender, ethnicity, socioeconomic status, and / or the like. In another embodiment, each patient profile may also include a patient’s medical history, for example, and without limitation, patient profile may include a detailed record of the patient's past health conditions, medical procedures, hospitalizations, and illnesses such as surgeries, treatments, medications, and / or the like. In another embodiment, each patient profile may include lifestyle Information of patient, for example, and without limitation, patient profile may include details about the patient's diet, exercise habits, smoking and alcohol consumption, and other behaviors that could impact health. In a further embodiment, patient profile may include patient’s family history, for example, and without limitation, patient profile may include a record of hereditary diseases. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various type of data within patient profiles apparatus 800 may receive and process in consistent with this disclosure.

[0166] In a non-limiting example, and still referring to FIG. 8, patient profile may include one or more ICE images or set of images 812. Receiving set of images 812 may include extracting set of images 812 from patient profile (subsequent to patient identity verification and obtaining consent from subject 820). In some cases, patient profile of subject 820 may be obtained through hospital information system (HIS) or any other data acquisition platform to securely access patient’s electronic medical record (EMR) or other relevant databases. Set of images 812 may be directly or indirectly downloaded or exported. In some cases, each ICE image of set of images 812 may be in a usable and / or computer-readable format such as, without limitation, DICOM format, and necessary metadata (e.g., patient information described above) may be included. Further, receiving set of images 812 may include recording the access and extraction of set of images 812; for instance, and without limitation, this process may be documented, by processor 804, in the patient’ s / subject’s 820 medical record, databases, or other appropriate logs.

[0167] Further, and still referring to FIG. 8, in other embodiments, patient profile may include electrocardiogram (ECG) data, wherein the “ECG data,” for the purpose of this disclosure, refers to data related to an electrocardiogram of the patient that corresponds to the patient profile. A “electrocardiogram,” as described herein, is a medical test that records the electrical activity of subject’s heart over a period of time. In an embodiment, ECG data may include one or more recordings captured by a plurality of electrodes placed on patient’s skin. In one or more embodiments, ECG data may include information regarding a P wave, T wave, QRS complex, PR interval, ST segment, and / or the like. Processor 804 may associate set of images 812 with ECG data, or in other cases, receiving set of images 812 may include receiving ECG data pertaining to subject 820 associated with set of images 812. Such ECG data may be collected simultaneously during ICE imaging. In some cases, set of images 812 may be linked with ECG data by one or more unique identifiers, such as without limitations, timestamps or other metadata described herein. In a non-limiting example, ECG data may be used to identify specific cardiac events or phases of the cardiac cycle, and the corresponding ICE images may be analyzed to see how heart’s structure changes during those times.

[0168] With continued reference to FIG. 8, in other embodiments, receiving set of images 812 may include receiving set of ICE images from an Image database 824. In some cases, Image database 824 may be implemented, without limitation, as a relational database, a key -value retrieval database such as a NOSQL database, or any other format or structure for use as a database that a person skilled in the art would recognize as suitable upon review of the entirety of this disclosure. Image database 824 may alternatively or additionally be implemented using a distributed data storage protocol and / or data structure, such as a distributed hash table or the like. Image database 824 may include a plurality of data entries and / or records as described above. Data entries in Image database 824 database may be flagged with or linked to one or more additional elements of information, which may be reflected in data entry cells and / or in linked tables such as tables related by one or more indices in Image database 824 or another relational database. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which data entries in a database may store, retrieve, organize, and / or reflect data and / or records as used herein, as well as categories and / or populations of data consistently with this disclosure.

[0169] In a further embodiment, and still referring to FIG. 8, receiving set of images 812 may involve one or more image preprocessing steps. In some cases, processor 804 may be configured to calibrate one or more ICE images of set of images 812 by correct for distortions and ensure accurate spatial representation of cardiac anatomy 816 pertaining to subject 820. In a nonlimiting example, processor 804 may select one or more reference objects within ICE image that needs calibration to correct spatial distortions. In some cases, processor 804 may be configured to place a phantom with pre-determine dimensions in such ICE image and adjust ICE image until the phantom’s dimensions are accurately represented. In another non-limiting example, one or more ICE images’ brightness and contrast may be adjusted, by processor 804 to ensure that echogenicity (reflectivity) of the tissues is accurately represented. One or more tissues with known echogenicity may be selected by processor 804 as reference tissues to adjust corresponding portions of the one or more ICE images. In other cases, standardized correction curves may be applied in or der to correct the echogenicity of ICE images. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, may be aware of various calibration techniques, such as, without limitation, temporal calibration, geometric calibration, among others that can be used by processor 804 to preprocess set of images 812.

[0170] Additionally, or alternatively, and still referring to FIG. 8, receiving set of images 812 may include perform image segmentation on or more ICE images of set of images 812. In some cases, image segmentation may include separating specific structures or regions of interest (ROI) from the background or other structures in a given ICE image. In a non-limiting example, processor 804 may be configured to use edge detection algorithms to outline the heart chambers, separating them from surrounding tissues. One or more filters may be applied to highlight the boundaries between different types of tissues during the segmentation. In another non-limiting examples, valves and vessels may also be segmented by applying thresholding techniques. Processor 804 may be configured to set an intensity threshold based on the known echogenicity of blood and vessel walls and select pixels or regions having intensity below or above the intensity threshold from the given ICE image. In some cases, one or more machine learning models may be used to perform image segmentations, for example, and without limitation, a U- net (i.e., a convolution neural network containing a contracting path as an encoder and an expansive path as a decoder, wherein the encoder and the decoder forms a U-shaped structure).

[0171] With continued reference to FIG. 8, processor is configured to generate a 3D data structure 828 representing cardiac anatomy 816 as a function of set of images 812. In a nonlimiting example, 3D data structure 828 may include a 3D voxel occupancy representation (VOR). As used in this disclosure, a "3D voxel occupancy representation (VOR)" of a cardiac anatomy is a 3D digital representation of a spatial structure of the cardiac anatomy, wherein the representation is composed of a plurality of discrete volumetric elements known as voxels 832. A “voxel,” for the purpose of this disclosure, is a 3D equivalent of a pixel in 2D imaging. While a pixel represents a point in a 2D image and may include properties such as color and / or brightness, a voxel may represent a volume in a 3D space and may include additional properties such density / occupancy as described below. In an embodiment, each voxel of plurality of voxels 832 within 3D VOR may represent a specific portion of cardiac anatomy 816. In some cases, voxel may be a smallest distinguishable box-shaped part (i.e., Ipx- Ipx Ipx) of a three- dimensional image. In some cases, each voxel of plurality of voxels 832 within VOR may be represented as a cube or rectangular prism (although other shapes may be used in specialized applications). Each voxel may include a size that determines a resolution of the 3D image or model. In an embodiment, smaller voxels may provide higher resolution; however, it may require more computational resources (e.g., RAM) for processor 804 to process.

[0172] In an embodiment, and still referring to FIG. 8, each voxel of plurality of voxels 832 within VOR may include one or more embedded values. As used herein, “embedded values” refers to specific numerical or categorical data associated with each voxel. In some cases, embedded values may represent various attributes or characteristics of the corresponding portion of cardiac anatomy 816 that voxel represents. In a non-limiting example, embedded values may include density values, intensity values, texture information, or any other quantitative measures that provide insights into the underlying cardiac tissue. Such embedded values may be derived from set of ICE images or other imaging modalities used to generate data structure 828. In some cases, embedded values may be utilized, by processor 804, to differentiate between different types of cardiac tissues, such as myocardial tissue, blood vessels, or chambers. Embedded values may also facilitate the visualization of dynamic cardiac functions, for example, and without limitation, blood flow or heart beating by encoding temporal information such as timestamps within plurality of voxels 832.

[0173] Still referring to FIG. 8, in an embodiment, each voxel of plurality of voxels 832 may include a presence indicator 836. As used in this disclosure, a “presence indicator” refers to a data element that indicates a presence or absence (i.e., occupancy) of cardiac tissue within that portion. In some cases, and without limitation, presence indicator 836 may include an occupancy status as one of the embedded values described herein. Portion may include a specific location within 3D space where data structure 828 is generated; for instance, and without limitation, a coordinate in 3D space represented in a tuple such as (x, y, z). In an embodiment, 3D VOR may provide a spatial framework that allows for the modeling and visualization of cardiac anatomy 816 in 3D space. In some cases, 3D data structure 828 may include a plurality of layers or slices (either horizontal [e.g., xy plane] or vertical [e.g., xz or yz plane depends on the view direction]), wherein each layer or slices of the plurality of layers or slices is corresponding to a different cross-sectional view of subject’s 820 heart, and collectively forming a comprehensive 3D depiction of the cardiac structure. In a non-limiting example, 3D VOR having plurality of voxels 832 with presence indicators 836 may indicate whether each voxel in 3D space may be occupied by a part of subject’s 820 heart. A binary value such as 0 or 1 may be configured as presence indicator to show ether a pixel of 3D space is occupied (e.g., 1) or empty (e.g., 0). In should be noted that other values may be used as presence indicator 836 such as a Boolean value e.g., TRUE or FALSE.

[0174] In some cases, and still reference to FIG. 8, one or more embedded values, such as, without limitations, occupancy, or density, may be derived from set of images 812 described herein by processor 804. In a non-limiting example, determining occupancy status of each voxel of plurality of voxels 832 may include converting set of ICE images to a set of binary images and determining occupancy status of each voxel as a function of the structure of interest’s binary value. Tn some cases, occupancy status may include a value representing the likelihood of occupancy of the corresponding heart tissue. In another non-limiting example, density may be calculated, by processor 804, for each voxel as a function of the echogenicity of one or more pixels on a given ICE image, wherein, the brightness of the given ICE image may be analyzed since different tissues reflect ultrasound waves differently.

[0175] With continued reference to FIG. 8, generating 3D data structure 828 of cardiac anatomy 816 may include generating a 3D array. In some cases, processor 804 may divide 3D space into a grid of plurality of voxels 832, each with specific x, y, and z coordinates as embedded values. Each element of 3D array may correspond to a voxel. In some cases, 3D array may allow for easy access and manipulation of plurality of voxels 832, enabling various analyses, visualizations, and transformations either described or not described herein. In a non-limiting example, embedded values may include a density of the tissue at a specific location of a patient’s body derived from one or more ICE images of set of images 812.

[0176] Additionally, or alternatively, and still referring to FIG. 8, 3D data structure 828 of cardiac anatomy 816 may include a 3D grid configured to map presence indicators 836 and / or other embedded values described herein of plurality of voxels 832 (e.g., tissue density, blood flow velocity, echogenicity or acoustic properties, and any other biophysical properties). As used in this disclosure, a “3D grid” refers to a 3D data structure that divides a given volume (e.g., volume of a heart) into a plurality of discrete units called cells (i.e., volume elements). In an embodiment, each cell within 3D grid may be associated with a distinct voxel. Mapping presence indicators 836 or other embedded values may include assigning each presence indicator or embedded value to each points within 3D grid such as corners of each corresponding cell. Such values may be derived from set of images 812 as described above.

[0177] In yet another embodiment, and still referring to FIG. 8, cells may be continuous, meaning that one or more cells may represent one or more continuous regions of space rather than discreate, separate units. In a non-limiting example, instead of being uniform, mapped presence indicator and / or other embedded values may vary continuously across different cells or cell’s volume. In such embodiment, processor 804 may use interpolation to estimate other (unknown) embedded values within a range based on existing values such as known embedded values at specific points, thereby allowing for smooth transitions between cells. Exemplary interpolation methods may include, without limitation, linear interpolation, cubic interpolation, and / or the like. For example, and without limitation, if the corners of a cell have known values interpolation can be used to estimate the values at any point within the cell based on those corner values.

[0178] In a non-limiting example, and still referring to FIG. 8, 3D data structure 828 of cardiac anatomy 816 may include a 3D grid having a plurality of cells e.g., voxels, wherein each cell may contain a continuous range of values representing tissue density, blood flow velocity, or other properties (i.e., embedded values). Processor 804 may be configured to apply trilinear or tricubic interpolation to estimate tissue density within each cell based on presence indicator or other known values at the cell’s boundaries, since tissue densities change gradually; Such 3D grid may provide a smooth, continuous representation of heat’s internal structures, allowing for more nuanced analysis and visualization as described below. In a further embodiment, 3D grid with continuous cells may be additionally used in fluid dynamics simulations.

[0179] With continued reference to FIG. 8, in some case, presence indicators 836 and / or other embedded values may be mapped to 3D grid as a function of array masking, wherein specific array or grid may be selected to modify based on one or more pre-defined criteria. In a nonlimiting example, processor 804 may generate a mask e.g., a binary array that defines which voxels or cells are affected. Mask may be used to select or modify specific voxels or cells based on certain attributes; for instance, and without limitation, processor 804 may use mask to isolate the LA within the heart focusing the analysis on that specific region. Such mask may include a criteria defined by specific density thresholds that distinguish the LA’s tissue (i.e., voxels representing LA in 3D grid) from surrounding structures (i.e., neighboring voxels). In some cases, such mask may further include a binary mask, wherein each voxel in the 3D gird may be assigned a first presence indicator such as 1 if the voxel meets the criteria for the LA and a second presence indicator such as 0 if it does not. In some embodiments, mask may be directly applied to 3D grid, selecting, or modifying voxels or cells, thereby enabling processor 804 to highlight, exclude, or otherwise manipulate specific parts of cardiac anatomy 815 within 3D grid. Processor 804 may then perform an element-wise multiplication between 3D grid and the mask. Continuing from the previous non-limiting example, voxels corresponding to the LA (wherein the mask value is 1) may retain their original values, while other voxels (where the mask value is 0) may be set to 0 or other specific value (i.e., excluded or masked out).

[0180] With continued reference to FIG. 8, in some embodiments, 3D grid may include one or more spatial features 840 extracted from set of images 812 of cardiac anatomy 816. As used in this disclosure, “spatial features” are specific characteristics or attributes related to the spatial arrangement, shape, size, texture, or orientation of structures within a 3D space. In some cases, spatial features may include one or more embedded values described herein and their combinations thereof. In a non-limiting example, spatial feature may be represented numerically as a vector, a metric or other mathematical constructs that capture specific spatial characteristics. In some cases, spatial features 840 may also be visualized as contours, surfaces, or other geometric representations. In an embodiment, spatial features 840 may be extracted using edge detection, texture analysis, or other image processing techniques (e.g., cleaning and enhancing images, image segmentation, and / or the like). In another embodiment, one or more machine learning models, such as convolutional neural networks (CNNs) as described in further detail below, may be used to extract complex spatial features 840.

[0181] Still referring to FIG. 8, as used in this disclosure, a “vector” is a data structure that represents one or more a quantitative values and / or measures of one or more spatial features 840. A vector may be represented as an n-tuple of values, where n is one or more values, as described in further detail below; a vector may alternatively or additionally be represented as an element of a vector space, defined as a set of mathematical objects that can be added together under an operation of addition following properties of associativity, commutativity, existence of an identity element, and existence of an inverse element for each vector, and can be multiplied by scalar values under an operation of scalar multiplication compatible with field multiplication, and that has an identity element is distributive with respect to vector addition, and is distributive with respect to field addition. Each value of n-tuple of values may represent a measurement or other quantitative value associated with a given category of data, or attribute, examples of which are provided in further detail below; a vector may be represented, without limitation, in n- dimensional space using an axis per category of value represented in n-tuple of values, such that a vector has a geometric direction characterizing the relative quantities of attributes in the n-tuple as compared to each other. Two vectors may be considered equivalent where their directions, and / or the relative quantities of values within each vector as compared to each other, are the same; thus, as a non-limiting example, a vector represented as [5, 10, 15] may be treated as equivalent, for purposes of this disclosure, as a vector represented as [1, 2, 3], Vectors may be more similar where their directions are more similar, and more different where their directions are more divergent, for instance as measured using cosine similarity as computed using a dot product of two vectors; however, vector similarity may alternatively or additionally be determined using averages of similarities between like attributes, or any other measure of similarity suitable for any n-tuple of values, or aggregation of numerical similarity measures for the purposes of loss functions as described in further detail below. Any vectors as described herein may be scaled, such that each vector represents each attribute along an equivalent scale of values. Each vector may be “normalized,” or divided by a “length” attribute, such as a length attribute / as derived using a Pythagorean norm: where a, is attribute number i of the vector. Scaling and / or normalization may function to make vector comparison independent of absolute quantities of attributes, while preserving any dependency on similarity of attributes.

[0182] Still referring to FIG. 8, in a non-limiting example, one or more spatial features 840 may include one or more shape features (i.e., characteristics related to the shape of specific cardiac structures), such as curvature, surface area, volume, and / or the like. In another non-limiting example, one or more spatial features 840 may include one or more texture features (i.e., characteristics related to the texture or pattern within cardiac tissues, as seen set of images 812), such as gray-level co-occurrence matrix (GLCM) features representing the texture of heart muscle tissue. In another non-limiting example, one or more spatial features 840 may include one or more orientation features (i.e., characteristics related to the orientation or alignment of cardiac structures), such as the angle or alignment of the septum within the heart. In a further non-limiting example, one or more spatial features 840 may include one or more edge and boundary features (i.e., Characteristics related to the edges or boundaries between different cardiac structures or tissues), such as edge detection features highlighting the boundary between the myocardium and the cardiac chambers. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various spatial features extracted from set of images 812 in consistent with this disclosure.

[0183] With continued reference to FIG. 8, in some embodiments, apparatus 800 may include a computer vision model 844 configured to generate 3D data structure 828 of cardiac anatomy 816. A “computer vision model,” for the purpose of this disclosure, is a computation model designed to interpret and make determinations based on visual data. In an embodiment, computer vision model 844 may process set of images 812, to make a determination about a scene, space, and / or object in cardiac anatomy 816. In a non-limiting example, computer vision model 844 may be used for registration of plurality of voxels 832 within a 3D space. Tn some cases, registration may include image processing described herein, such as without limitation object recognition, feature detection, edge / corner detection, and the like. Non-limiting example of feature detection may include scale invariant feature transform (SIFT), Canny edge detection, Shi Tomasi corner detection, and the like. In some cases, registration may include one or more transformations to orient an ICE image relative a 3D coordinate system; exemplary transformations include without limitation homography transforms and affine transforms. In an embodiment, registration of ICE image to a coordinate system may be verified and / or corrected using object identification and / or computer vision, as described above. For instance, and without limitation, an initial registration to two dimensions, represented for instance as registration to the x and y coordinates, may be performed using a two-dimensional projection of points in three dimensions onto the ICE image; however, a third dimension of registration, representing depth and / or a z axis, may be detected by utilizing depth-sensing techniques such as Doppler imaging. Alternatively, the third dimension may be inferred from the known geometry and orientation of the imaging device (e.g., ICE catheter), or through the application of one or more machine learning models trained to interpret depth from the two-dimensional projection.

[0184] With continued reference to FIG. 8, processor 804 may use a machine learning module 848 to implement one or more algorithms or generate one or more machine learning models, such as a cardiac anatomy modeling model 852 to generate data structure 828 of cardiac anatomy 816. However, the machine learning module is exemplary and may not be necessary to generate one or more machine learning models and perform any machine learning described herein. In one or more embodiments, one or more machine-learning models may be generated using training data. Training data may include inputs and corresponding predetermined outputs so that a machine-learning model may use correlations between the provided exemplary inputs and outputs to develop an algorithm and / or relationship that then allows machine-learning model to determine its own outputs for inputs. Training data may contain correlations that a machinelearning process may use to model relationships between two or more categories of data elements. Exemplary inputs and outputs may come from a database, such as any database described in this disclosure, or be provided by a user. In other embodiments, a machine-learning module may obtain a training set by querying a communicatively connected database that includes past inputs and outputs. Training data may include inputs from various types of databases, resources, and / or user inputs and outputs correlated to each of those inputs so that a machine-learning model may determine an output. Correlations may indicate causative and / or predictive links between data, which may be modeled as relationships, such as mathematical relationships, by machine-learning models, as described in further detail below. In one or more embodiments, training data may be formatted and / or organized by categories of data elements by, for example, associating data elements with one or more descriptors corresponding to categories of data elements. As a non-limiting example, training data may include data entered in standardized forms by persons or processes, such that entry of a given data element in a given field in a form may be mapped to one or more descriptors of categories. Elements in training data may be linked to descriptors of categories by tags, tokens, or other data elements. In a further embodiment, training data may include previous outputs such that one or more machine learning models iteratively produces outputs.

[0185] Still referring to FIG. 8, machine learning module 848 may be used to generate cardiac anatomy modeling model and / or any other machine learning models, such as, shape identification model as described in further detail below, using training data. Cardiac anatomy modeling model 852 may be trained by correlated inputs and outputs of training data. Training data may be data sets that have already been converted from raw data whether manually, by machine, or any other method. In an embodiment, generating data structure 828 of cardiac anatomy 816 includes receiving cardiac anatomy training data 856, wherein the cardiac anatomy training data 856 may include a plurality of image sets as input and a plurality of computed tomography (CT) based cardiac anatomy models as output, and wherein each image set of plurality of image sets may include any images described in this disclosure. In some cases, cardiac anatomy training data 856 may be received from Image database 824 or other databases. In other cases, cardiac anatomy training data 856 may be collected by a data acquisition unit from external sources such as one or more medical equipment’s e.g., imaging devices or diagnostic tools, wherein the data acquisition may be configured as an intermediary between the data source and machine learning module 848.

[0186] Still referring to FIG, 1, as used in this disclosure, a “computed tomography (CT) based cardiac anatomy model” refers to a 3D representation of the heart and surrounding structures that is created using data from CT scans. Computed Tomography is a medical imaging technique that uses X-rays to capture cross-sectional images (slices) of the body. By taking a plurality of slices, a CT scan creates a detailed 3D representation of the internal structure. In an embodiment, CT- based cardiac anatomy model may include 3D representations of the heart including chambers, valves, blood vessels, and surrounding tissues. In some cases, CT-based cardiac anatomy model may be interactive; for instance, medical professionals may rotate, zoom, and / or explore CT- based cardiac anatomy model from various angles. In some cases, plurality of CT-based cardiac anatomy models may be generated prior to the training of the cardiac anatomy modeling model 852. Plurality of CT-based cardiac anatomy models may be generated using existing techniques in the field as described above such as, without limitation, FAM, cardiac CT merging, among others. In a non-limiting example, plurality of CT-based cardiac anatomy models may provide ground through or references models against cardiac anatomy modeling model 852 that is being trained. In a non-limiting example, generating data structure 828 of cardiac anatomy 816 further includes training cardiac anatomy modeling model 852 using cardiac anatomy training data described herein. Cardiac anatomy modeling model 852 trained using cardiac anatomy training data 856 may be able to interpret ICE images by learning relationships between ICE images and corresponding CT-based cardiac anatomy models. Processor 804 is further configured to generate data structure 828 of cardiac anatomy 816 as a function of set of images 812 using trained cardiac anatomy modeling model 852. In some cases, data structure 828 e.g., 3D VOR may be interpreted, visualized, and analyzed by processor 804 in similar manner to CT-based cardiac anatomy models, wherein both are 3D structures that correspond to ICE images.

[0187] With continued reference to FIG. 8, in an embodiment, cardiac anatomy modeling model comprises a deep neural network (DNN). As used in this disclosure, a “deep neural network” is defined as a neural network with two or more hidden layers. Neural network is described in further detail below with reference to FIGS. 4-5. In a non-limiting example, cardiac anatomy modeling model may include a convolutional neural network (CNN). Generating 3D data structure 828 of cardiac anatomy 816 may include training CNN using cardiac anatomy training data and generating 3D data structure 828 as a function of set of images 812 using trained CNN. A “convolutional neural network,” for the purpose of this disclosure, is a neural network in which at least one hidden layer is a convolutional layer that convolves inputs to that layer with a subset of inputs known as a “kernel,” along with one or more additional layers such as pooling layers, fully connected layers, and the like. In some cases, CNN may include, without limitation, a deep neural network (DNN) extension. Mathematical (or convolution) operations performed in the convolutional layer may include convolution of two or more functions, where the kernel may be applied to input data e.g., set of images 812 through a sliding window approach. In some cases, convolution operations may enable processor 804 to detect local / global patterns, edges, textures, and any other spatial features 840 described herein within each ICE image of set of images 812. Spatial features 840 may be passed through one or more activation functions, such as without limitation, Rectified Linear Unit (ReLU), to introduce non-linearities into the processing step of generating 3D data structure 828 of cardiac anatomy 816. Additionally, or alternatively, CNN may also include one or more pooling layers, wherein each pooling layer is configured to reduce the dimensionality of input data while preserving essential features within the input data. In a non-limiting example, CNN may include one or more pooling layer configured to reduce the spatial dimensions of spatial feature maps by applying downsampling, such as max-pooling or average pooling, to small, non-overlapping regions of one or more spatial features 840.

[0188] Still referring to FIG. 8, CNN may further include one or more fully connected layers configured to combine spatial features 840 extracted by the convolutional and pooling layers as described above. In some cases, one or more fully connected layers may allow for higher-level pattern recognition. In a non-limiting example, one or more fully connected layers may connect every neuron (i.e., node) in its input to every neuron in its output, functioning as a traditional feedforward neural network layer. In some cases, one or more fully connected layers may be used at the end of CNN to perform high-level reasoning and produce the final output such as, without limitation, a 3D data structure 828 of cardiac anatomy 816. Further, each fully connected layer may be followed by one or more dropout layers configured to prevent overfitting, and one or more normalization layers to stabilize the learning process described herein.

[0189] With continued reference to FIG. 8, CNN may further include a 3D CNN, wherein the 3D CNN, unlike standard 2D CNN, may include utilization of one or more 3D convolutions which allow them to directly process 3D data, thereby enabling processor 804 to generate 3D structures such as 3D data structure 828 of cardiac anatomy 816 using the 3D CNN. In a non-limiting example, 3D CNN may include one or more 3D filters (i.e., kernels) that move through the set of images 812 in three dimensions and capturing spatial relationships in x, y, and z axis. Similar to 3D convolutions, 3D CNN may further include one or more 3D pooling layers that may be used to reduce the dimensionality of ICE images while preserving spatial features 840 as described above. Additionally, or alternatively, an encoder-decoder structure may be implemented (extended to 3D), by processor 804, in 3D CNN, wherein the encoder-decoder structure includes an encoding path that captures the context and a decoding path that enables precise localization in a same manner as U-net as described above. Such encoder-decoder structures may also include a plurality of skip connections, allowing 3D CNN to use information from multiple resolutions to improve the process of generating 3D data structure 828 of cardiac anatomy 816.

[0190] With continued reference to FIG. 8, in an embodiment, training the cardiac anatomy modeling model 852 (i.e., CNN) may include selecting a suitable loss function to guide the training process. In a non-limiting example, a loss function that measures the difference between the predicted 3D VORs and the ground truth 3D structure e.g., CT-based cardiac anatomy models may be used, such as, without limitation, mean squared error (MSE) or a custom loss function may be designed for one or more embodiments described herein. Additionally, or alternatively, optimization algorithms, such as stochastic gradient descent (SGD), may then be used to adjust the cardiac anatomy modeling model’s parameters to minimize such loss. In a further non-limiting embodiment, instead of directly predicting 3D data structure 828, cardiac anatomy modeling model 852 may be trained as a regression model to predict presence indicators 836 and / or other embedded values described herein for each voxel of plurality of voxels 832 within a 3D grid. Additionally, CNN may be extended with additional deep learning techniques, such as recurrent neural networks (RNNs) or attention mechanism, to capture additional features and / or data relationships within input data. These extensions may further enhance the accuracy and robustness of the cardiac anatomy modeling.

[0191] With continued reference to FIG. 8, alternatively, processor 804 may generate a set of shape parameters 860 based on set of images 812. As used in this disclosure, a “set of shape parameters” refers to a collection of numerical values or descriptors that quantitatively represent the geometric or morphological characteristics of a structure e.g., a heart. In a non-limiting example, set of shape parameters 860 may include information and / or metadata calculated, determined, and / or extracted from set of ICE images, such as, dimensions, angles, curvatures, surface areas, texture, symmetry, and / or the like. In other embodiments, processor 804 may be configured to parameterize features (e.g., edges, textures, contours, and any other characteristics that describe the shape cardiac anatomy 816) extracted from set of images 812 using CNN described herein. Such parameterization may involve processor 804 to derive one or more shape parameters including one or more morphological descriptors that quantitatively describe cardiac anatomy 816 based on extracted features. In some cases, processor 804 may be configured to use principal component analysis (PCA) to reduce the dimensionality of set of shape parameters 860, allowing processor 804 to focusing on the most informative shape parameters of set of shape parameters 860 in further processing steps described below.

[0192] With continued reference to FIG. 8, in a non-limiting example, set of shape parameters 860 may be generated based on set of images 812 using machine learning model such as, without limitation, a shape identification model 864. Generating set of shape parameters 860 may include receiving cardiac geometry training data 868, wherein the cardiac geometry training data 868 may include a plurality of image sets as input correlated to a plurality of shape parameter sets as output. In some cases, cardiac geometry training data may be received from Image database 824 described herein. For example, and without limitation, cardiac geometry training data 868 may be used to show each ICE image may indicate a particular set of shape parameters. Shape identification model 864 may be trained, by processor 804, using cardiac geometry training data 868. Additionally, cardiac geometry training data 868 may include previously input image sets and their corresponding shape parameters output. Shape identification model 864 may be iterative such that outputs may be used as future inputs of shape identification model 864. This may allow the shape identification model 864 to evolve. Processor 804 may be further configured to generate set of shape parameters 860 as a function of set of images 812 using the trained shape identification model 864.

[0193] With continued reference to FIG. 8, processor 804 is configured to generate an initial 3D model 872 of cardiac anatomy 816. As used in this disclosure, an “initial 3D model” is a foundational representation, capturing the basic geometric and spatial characteristics of the heart in 3D space. In an embodiment, initial 3D model 872 may provide a “starting point” for further refinement and customization as described in further detail below, allowing for the incorporation of more detailed and patient-specific information. In some cases, initial 3D model 872 may be generated through a direct 3D reconstruction from a series of (2D) ICE images. In a non-limiting example, set of images 812 may include a plurality of ICE images captured from different angles and positions within the heart. Processor 804 may be configured to apply one or more 3D reconstruction algorithms, such as without limitation, marching cubes, contour detection and segmentation, active contour models, and / or the like to create a coherent 3D representation e.g., initial model 872 of cardiac anatomy 816. In some cases, such direct 3D reconstruction may leverage the inherent spatial information within set of images 812, providing a direct and intuitive way to model the initial model 872 of the heart's structure. In a further embodiment, generic 3D modeling techniques may be applied to create the initial 3D model. In some cases, generic 3D modeling techniques may include surface modeling, solid modeling, or parametric modeling, among others. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various 3D reconstruction algorithms may be used by processor 804 to generate initial 3D model 872 of cardiac anatomy 816.

[0194] Additionally, or alternatively, and still referring to FIG. 8, initial 3D model 872 may be generated based on a plurality of standard anatomical templates, wherein the “plurality of standard anatomical templates,” for the purpose of this disclosure, refers to predefined and commonly accepted representations of the human body’s anatomical structures. In some cases, plurality of standard anatomical templates may be selected from Image database 824 as described herein based on statistical averages or shared characteristics. In a non-limiting example, initial 3D model 872 may include a template model 876 selected from a plurality of pre-determined template models. Plurality of pre-determined template model may be generated by processor 804 based on plurality of standard anatomical templates prior to the generation of initial 3D model 872 using 3D reconstruction / modeling algorithms / techniques as listed above. In an embodiment, generating initial 3D model 872 may include selecting template model 876 from plurality of template models based on set of ICE images. In some cases, template model 876 may represent a typical or average cardiac anatomy that is most similar to cardiac anatomy 816 pertaining to subject 820. Such similarity may be determined based on one or more similarity metrics, such as without limitation, structural similarity index (SSI), MSE, peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC), Pearson correlation coefficient, and / or the like between set of images 812 and each image sets stored in Image database 824. Template model 876 may be adjusted and customized to fit the specific patient's ICE images as described below in further detail.

[0195] With continued reference to FIG. 8, processor 804 is configured to refine generated initial 3D model 872 of cardiac anatomy 816 as a function of 3D data structure 828 of cardiac anatomy 816. In a non-limiting embodiment, refining initial 3D model 872 of cardiac anatomy 816 may include utilizing a statistical shape model (SSM) 180. It should be noted that SSM may not be the only method for refining initial 3D model 872. A person of ordinary skill in the art, upon reviewing the entirety of this disclosure, will be aware of various methods, such as, without limitation, mesh smoothing techniques, level set method, physics-based simulation, among others may be implemented, by processor 804, to refine initial 3D model 872 described herein. As used in this disclosure, a “statistical shape model (SSM)” is a data structure representing, including, and / or utilizing a mathematical model that captures principal modes of variation in shape across a population of cardiac anatomies. In some cases, SSM may be constructed by analyzing one or more datasets of shapes and identifying, for example, mean shapes and main modes of variation within the one or more datasets. In a non-limiting example, SSM may start with calculation of at least one mean shape, which represents an average geometry of all the heart shapes in a given dataset, wherein the at least one mean shape may be served as a central reference point for processor 804 to understand different variations. In some cases, dataset may include, without limitation, cardiac anatomy training data 856, cardiac geometry training data 868, and / or any datasets within ICE image databases described herein. SSM may also identify one or more principal modes of variation within given datasets described herein, wherein the “principal modes of variations,” for the purpose of this disclosure, refer to main patterns or directions along which data points vary within dataset. In a non-limiting example, identifying principal modes of variations may include applying principal component analysis (PCA) on given dataset. Additionally, or alternatively, shapes may be described directly using plurality of shape parameter sets (in cardiac geometry training data 868). In some cases, shape parameter sets may correspond to a plurality of modes of variations. Further, one or more statistical constraints (e.g., mean, variance, correlation, boundary, proportion constraint and / or the like) may be introduced into SSM 880 based on the distribution of shape parameters within plurality of shape parameter sets.

[0196] With continued reference to FIG. 8, refining initial 3D model 872 of cardiac anatomy 816 may include aligning initial 3D model 872 with 3D VOR of cardiac anatomy 816. In an embodiment, aligning initial 3D model 872 with 3D VOR may include matching template model 876 to 3D VOR; for instance, and without limitation, this may involve adjusting the position, orientation, and scale of template model 876 to match the spatial distribution captured in 3D VOR. In some cases, matching template model 876 to 3D VOR may include matching spatial features 840, wherein matching the spatial features 840 may further include aligning the surface, boundaries and internal structures of template model 876 with corresponding features in 3D VOR. In some embodiments, processor 804 may utilize one or more optimization techniques to achieve a desired alignment; for instance, and without limitation, processor may be configured to minimizing the difference between template model 876 and 3D VOR using iterative closest point (ICP) algorithms, gradient descent, or any other optimization strategies. Additionally, alignment of template model 876 with 3D VOR may also allow incorporation of patient-specific details (e.g., patient profile) into initial 3D model 872 to form a final model as described in further detail below.

[0197] In a non-limiting example, and still referring to FIG. 8, refining initial 3D model 872 of cardiac anatomy 816 may include deforming, using processor 804, template model 876 to match 3D data structure 828 of cardiac anatomy 816. As used in this disclosure, “deforming” means altering the geometric structure of a structure e g., template model 876 in a systematic and controlled manner to align the structure with the spatial characteristics captured in another structure e.g., 3D VOR. In some cases, processor 804 may utilize one or more mathematical deformation models such as, without limitation, B-splines, radial basis functions, or other deformation functions to control and guide the deformation process of template model 876. In some cases, one or more constraints listed above may be applied, by processor 804, based on anatomical knowledge, biomechanical properties, or other relevant factors to ensure that the deformation of template model 876 is realistic and consistent with physiological principles as would be understood and / or expected by an ordinary person skilled in the art.

[0198] Still referring to FIG. 8, additionally, or alternatively, refining initial 3D model 872 of cardiac anatomy 816 may also include validating template model or deformed template model against 3D data structure 828 or additional data such as, without limitation, expert input, adjust parameters, and / or the like. Such validation process may ensure that the refined model accurately represents the underlaying cardiac anatomy 816. In some cases, expert input may include any user input entered via a user interface as described in further detail below. In a non-limiting example, expert input may include, without limitation, clinical assessment, anatomical knowledge, or other professional insights that guide and evaluate the refinement process inputted to apparatus 800 by one or more users including medical professionals, subjects, patients, and / or any other related individuals. In a further embodiment, validating template model or deformed template model against 3D data structure 828 may also include fine-tuning defamation controls, alignment settings, or other model characteristics or properties to achieve desired alignment with 3D VOR or additional data. In some cases, other information that is incorporated and codified within template model 876 / deformed template model and / or 3D data structure 828 such as medical imaging, biomechanical simulations, patient-specific data / metadata may be validated and cross-verified. At least a machine-learning process, for example a machine-learning model described herein, may be used to validate by processor 804. Processor 804 may use any machinelearning process described in this disclosure for this or any other functions.

[0199] With continued reference to FIG. 8, in some embodiments, embedded values described herein may be employed in the refinement process of initial 3D model 872 of cardiac anatomy 816. In a non-limiting example, the embedded values may contribute to SSM 880 by providing additional parameters that guide the deformation and alignment of the template to match 3D VOR. Embedded values such as, without limitation, presence indicators 836 may be used by processor 804 to guide the deformation process by providing targets for alignment; for instance, and without limitation, SSM may be configured to identify specific target areas where initial 3D model e.g., a 3D LA model that needs to be deformed. Presence indicators 836, in this case, may reveal a bulge in LA wall that is not present in initial 3D model 872. In some cases, presence indicators 836 may define the exact shape of the bulge in LA wall. Processor 804 may then deform initial 3D model 872, particularly the wall to match the bulge defined by presence indicators 836 in 3D VOR.

[0200] With continued reference to FIG. 8, processor 804 is configured to generate a subsequent 3D model 884 of cardiac anatomy 816 as a function of the refinement. As used in this disclosure, a “subsequent 3D model” refers to a more detailed and accurate 3D representation of cardiac anatomy 816. In an embodiment, subsequent 3D model 884 may be derived from initial 3D model 872 and / or template model 876 and adjusted based on 3D data structure 828 ad described above. In such embodiment, subsequent 3D model 884 may include a deformed initial 3D model 872 and / or template model 876. In a non-limiting example, 3D VOR may indicate a need of adjustment to initial 3D model 872 of left ventricle to match subject’s 820 unique geometry. SSM 880 may then be configured to generate subsequent 3D model 884 that accurately captures such specific cardiac anatomy based on initial 3D model 872 and 3D VOR. In other cases, initial 3D model 872 may not need any refinement; for instance, and without limitation, if initial 3D model 872 already align perfectly with 3D data structure representing subject’s 820 right atrium (RA), no deformation or adjustment would be necessary, thereby resulting in subsequent 3D model 884 that is identical to initial 3D model 872.

[0201] Still referring to FIG. 8, in some cases, the refinement process may also include the incorporation of more detailed features and textures based on 3D data structure 828 and embedded values thereof, enhancing the realism and specificity of initial 3D model 872. In an embodiment, SSM 880 may be integrated with one or more additional models such as, without limitation, texture models, appearance models, or functional models to generate subsequent 3D model 884. In some cases, such integration may result in subsequent 3D model 884 that reflects not just the geometry but also the biomechanical properties or blood flow dynamics within cardiac anatomy 816. In a non-limiting example, texture of the myocardium may be modeled, by integrating texture models with SSM 880, to represent the fibrous nature of the heart muscle. In another non-limiting example, appearance of blood vessels, including color variations and translucency, may be modeled, by integrating appearance models with SSM 880.

[0202] With continued reference to FIG. 8, alternatively, refining initial 3D model 872 of cardiac anatomy 816 may include adjusting template model 876 based on set of shape parameters 860. In an embodiment, processor 804 may be configured to map set of shape parameters 860 to SSM 880. The mapping process may define how template model 876 should be adjusted to represent specific subject’s 820 cardiac anatomy. In a non-limiting example, shape parameters may include one or more numeric values indicating a particular thick ventricular wall, processor 804 may configure SSM 880 to adjust template model 876 to reflect such characteristic. In an embodiment, generating subsequent 3D model 884 may involve generating a 3D mesh or grid that accurately represents the shape defined by set of shape parameters; for instance, and without limitation, processor 804 may be configured to generate a 3D mesh for left ventricle with vertices and edges positioned according to specific curvature and thickness defined by set of shape parameters 860 using SSM 880.

[0203] With continued reference to FIG. 8, in some embodiments, processor 804 may be configured to input subsequent 3D model 884 back into cardiac anatomy modeling model 852 and / or shape identification model 864 for continuous learning. In some cases, training data for these models such as, without limitation, cardiac anatomy training data 856, cardiac geometry training data 868, and / or the like may be updated, by replacing, appending or otherwise inserting subsequent 3D model 884 (and corresponding set of ICE images) into the dataset. This iterative process may allow machine learning module 848 to evolve over time, adapting to new set of ICE images and improving the accuracy of machine learning models generated by machine learning module 848. Incorporation of subsequent 3D models as additional training data may enable apparatus 800 to capture more variations and nuances in cardiac anatomy modeling, enhancing its ability to generalize across different patients and conditions.

[0204] Still referring to FIG. 8, additionally, processor 804 may use user feedback to train the machine-learning models described above. For example, cardiac anatomy modeling model 852 and / or shape identification model 864 may be trained using past inputs and outputs of cardiac anatomy modeling model 852 and / or shape identification model 864. In some embodiments, if user feedback indicates that a subsequent 3D model outputted by SSM 880 was “bad,” then that output and the corresponding input e.g., set of ICE images, corresponding CT-based cardiac anatomy model, and / or template model, may be removed from training data used to train cardiac anatomy modeling model 852 and / or shape identification model 864, and / or may be replaced with a value entered by, e.g., another user that represents an ideal 3D model of the heart given the input the machine learning models originally received, permitting use in retraining, and adding to training data as described above; in either case, machine learning models described herein may be retrained with modified training data. In some embodiments, training data such as cardiac anatomy training data 856 and / or cardiac geometry training data 868 may include user feedback. Further, apparatus 800 may be configured to validate one or more machine learning models described herein against real-world data, identifying areas where machine learning models may be underperforming or misaligned with clinical needs. Such feedback may also be used to guide model training, ensuring that machine learning models are not only accurate but also clinically meaningful and aligned with healthcare or medical professional’s needs and priorities.

[0205] With continued reference to FIG. 8, apparatus 800 may further include a display device 888. As used in this disclosure, a “display device” is an electronic device that visually presents information to a user. In an embodiment, display device may include an output interface that translates data such as, without limitation, subsequent 3D model 884 from processor 804 or other computing devices into a visual form that can be easily understood by user. In some cases, subsequent 3D model 884 and / or other data described herein such as, without limitation, ICE images, 3D VOR, shape parameters initial model and / or template model may also be displayed through display device 888 using a user interface 892. User interface 892 may include a graphical user interface (GUI), wherein the GUI may include a window in which subsequent 3D model 884 and / or other data described herein may be displayed. In an embodiment, user interface 892 may include one or more graphical locator and / or cursor facilities allowing user to interact with subsequent 3D model 884 and / or any other data, or even process described herein; for instance, and without limitation, by using a touchscreen, touchpad, mouse, keyboard, and / or other manual data entry device, user may enter user input containing selecting specific regions, adding comments, adjusting parameter, and / or the like. In a non-limiting example, user interface 892 may include one or more menus and / or panels permitting selection of measurements, models, visualization of data / model to be displayed and / or used, elements of data, functions, or other aspects of data / model to be edited, added, and / or manipulated, options for importation of and / or linking to application programmer interfaces (APIs), exterior services, data source, machine-learning models, and / or algorithms, or the like. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which a visual interface and / or elements thereof may be implemented and / or used as described in this disclosure.

[0206] Now referring to FIG. 9, an exemplary embodiment of an ICE image 900 is illustrated. As described above with reference to FIG. 8, set of images 812 may include a plurality of ICE images, wherein each ICE image of the plurality of ICE images is a specialized form of echocardiography that may provides detailed image of heart’s (i.e., cardiac anatomy 816) interior structures. In a non-limiting example, plurality of ICE images may include an ICE video (e.g., plurality of ICE images arranged in a corresponding time sequence). In an embodiment, ICE image 900 may be real-time, dynamic ultrasound image that provide a (detailed) view 904 of heart’s interior structures, including, without limitation, right atrium (RA) 908, anterior descending (AD) 912, pulmonary atresia (PA) 916, and right ventricular (RV) 920.

[0207] With continued reference to FIG. 9, in some cases, ICE image 900 may include gray scaled image. It should be noted that, in some cases, ICE image 900 may be configured to visualize blood flow and / or blood flow patterns within the heart via color doppler as described above with FIG. 8. In some cases, resolution and / or clarity of ICE image 900 as described herein may be superior to transthoracic or transesophageal echocardiography due to the ICE catheter may be positioned inside the heart, closer to the structures being imaged. Still referring to FIG. 9, in a non-limiting example, heart chambers 908 may appear as dark, anechoic (black) areas since they are filled with blood, which doesn’t reflect ultrasound waves well. Heart walls, valves, and / or other structures may appear as varying shades of gray, depending on their density and composition, in some cases, Color Doppler overlays may show blood flow in different colors, indicating the direction and speed of blood flow. For instance, and without limitation, red may indicate flow towards the probe, while blue may indicate flow away from the probe.

[0208] With continued reference to FIG. 9, in a non-limiting embodiment, ICE image 900 may be synchronized with ECG data as described above with reference to FIG. 8, allowing for precise timing of cardiac events with anatomical visualization provided by ICE. In some cases, ICE image 900 may include an ECG display 924 configured to display ECG waveform as a continuous line graph at the top, bottom, or side of ICE image 900. In some cases, specific parts of the cardiac cycle e.g., systole or diastole, may be correlated with visual data from ICE image 900.

[0209] Additionally, or alternatively, and still referring to FIG. 9, ICE image 900 may come with accompanying metadata 928 displayed on the side or comers of ICE image 900 as described herein. In some cases, metadata 928 may provide essential contextual information about ICE image 900 and / or the corresponding patient. In a non-limiting example, metadata 928 may include patient information (e.g., patient ID, name, DOB, age, gender, and the like), image acquisition details (e.g., date and time, probe type, frequency, depth, gain, and the like), procedure-related information (e.g., procedure name, operator, location, and the like), ECG trace (e.g., ECG data as described above), measurement annotations (e.g., any measurements taken directly on the image e.g., diameter, a value of thickness of a heart wall and the like), image sequence information (e.g., image number, total number of frames, and the like), comments or notes, hospital or clinic information, and / or the like. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of ICE image 900 and various components thereof may be incorporated by apparatus 800 for generating 3D model of cardiac anatomy.

[0210] Now referring to FIG. 10, a flow diagram of an exemplary embodiment of an ICE example generation process 1000. In an embodiment, cardiac anatomy training data 856 may be generated, at least in part, via ICE example generation process 1000. In some cases, processor 804 may be configured to receive a 3D model of the heart, such as, without limitation, template model 876, initial model 872, subsequent 3D model 884, and / or any 3D model of cardiac anatomy 816 as described herein and identify an ICE view 1004 (i.e., visual representation of image obtained using intracardiac echocardiography as described above e.g., ICE image 900) based on the received 3D model. In some cases, 3D model received by processor 804 may be derived from CT scans as described above with reference to FIG. 8. In other cases, processor may receive CT scans directly instead of 3D models. A synthetic ICE frame 1008 may then be generated, by processor 804, as a function of identified ICE view 1004, wherein the synthetic ICE frame 1008 may be used as one or the training examples in cardiac anatomy training data 856.

[0211] With continued reference to FIG. 10, in some cases, processor 804 may interface with one or more 3D models (i.e., detailed representation of heart’s anatomy in a 3D space, capturing intricate structures, chambers, vessels, valves, among others) as described above, or other imaging modalities and / or databases, and equipped with algorithms e.g., CNN, gradient boosting machines, SVM, PC A, and / or the like to analyze model’s geometry and spatial relationships upon receiving the 3D models. In some cases, 3D models may be received from SSM 880 as described above with reference to FIG. 8 via a communicative connection between processor 804 and SSM 880. In a non-limiting example, processor 804 may be configured to determine an optimal viewpoints or angles from which ICE view 1004 would provide a desired diagnostic value or procedural guidance.

[0212] Still referring to FIG. 10, in some cases, identification and selection of ICE view 1004 may be automatically identified, using one or more machine learning models as described herein. In a non-limiting example, processor 804 may utilize one or more machine learning models trained on cardiac anatomy viewpoints identification training data, wherein the cardiac anatomy viewpoints identification training data may include a plurality of cardiac anatomies as input correlated to a plurality of ICE images as output and identify at least one ICE view 1004 (most informative) for a given cardiac anatomy using the trained machine learning models.

[0213] Still referring to FIG. 10, in other cases, ICE view 1004 may be defined by a user such as a medical professional. In a non-limiting example user interface 892 of display device 888 may allow a user (e.g., a clinician) to manually rotate, pan, and zoom displayed 3D model and / or corresponding CT scans. As user do so, processor 804 may dynamically calculate and displays potential ICE views 1004 based on user’s chosen perspective. Additionally, or alternatively, depending on cardiac procedure being planned or executed, processor 804 may prioritize certain ICE views 1004. For instance, and without limitation, ICE view 1004 may be pre-defined. For atrial fibrillation ablation, ICE view 1004 may showcase the pulmonary veins’ entrances into the LA may be emphasized. In other cases, ICE view 1004 may be automatically identified, by processor 804, using one or more machine learning models as described herein, such as, without limitation, synthetic ICE data generator as described in detail below.

[0214] With continued reference to FIG. 10, as used in this disclosure, a “synthetic ICE frame” refers to a digitally generated or simulated image that emulates a visual representation obtained from ICE view 1004. In some cases, synthetic ICE frames 1008 may be produced using computational methods and / or models such as, without limitation, a synthetic ICE data generator 1012 based on pre-existing data, models, or simulations e.g., identified ICE views 1004. In a non-limiting example, synthetic ICE frames 1008 may include a simplified version e.g., an image illustrating heart anatomy via a plurality of lines indicating contours of heart’s structure as shown in FIG. 10. One or more image processing techniques and / or computer vision algorithms such as, without limitation, histogram equalization, adaptive filtering, edge detection (e.g., Canny or Sobel operators), contour extraction, and / or the like may be applied, by processor 804, on a segmented CT scan and / or 3D models based on identified ICE view 1004. Synthetic ICE frame 1008 may be rendered on a blank canvas or background that mimics the echogenicity of an ICE image according to extracted contours, wherein the extracted contours may be represented as a bold lines and enhanced with shading to give depth. In some cases, synthetic ICE frame 1008 may be validated and verified by overlaying synthetic ICE frame 1008 onto original ICE view 1004, ensuring accuracy and resemblance.

[0215] Still referring to FIG. 10, in some cases, generating synthetic ICE frames 1008 may include implementations of one or more aspects of “generative artificial intelligence(AI),” a type of Al that uses machine learning algorithms to create, establish, or otherwise generate data such as, without limitation, ICE images, ICE videos, and / or the like that is similar to one or more provided training examples. In an embodiment, machine learning module described herein may generate one or more generative machine learning models that are trained on one or more set of CT scans and / or 3D models in ICE image view 1004 as described above. Synthetic ICE data generator 1012 may include one or more generative machine learning models may be configured to generate new examples that are similar to the training data of the one or more generative machine learning models but are not exact replicas; for instance, and without limitation, data quality or attributes of the generated examples may bear a resemblance to the training data provided to one or more generative machine learning models, wherein the resemblance may pertain to underlying patterns, features, or structures found within the provided training data.

[0216] Still referring to FIG. 10, in some cases, generative machine learning models within synthetic ICE data generator may include one or more generative models. As described herein, “generative models” refers to statistical models of the joint probability distribution P(X, K) on a given observable variable x, representing features or data that can be directly measured or observed (e.g. CT scans and / or 3D models derived from CT scans) and target variable y, representing the outcomes or labels that one or more generative models aims to predict or generate (e.g., synthetic ICE frames 1008). In some cases, generative models may rely on Bayes theorem to find joint probability; for instance, and without limitation, Naive Bayes classifiers may be employed by computing device to categorize input data such as, without limitation, CT scans and / or 3D models derived from CT scans into different views.

[0217] In a non-limiting example, and still referring to FIG. 10, one or more generative machine learning models may include one or more Naive Bayes classifiers generated, by processor 804, using a Naive bayes classification algorithm. Naive Bayes classification algorithm generates classifiers by assigning class labels to problem instances, represented as vectors of element values. Class labels are drawn from a finite set. Naive Bayes classification algorithm may include generating a family of algorithms that assume that the value of a particular element is independent of the value of any other element, given a class variable. Naive Bayes classification algorithm may be based on Bayes Theorem expressed as P(A / B)= P(B / A) P(A)-^P(B), where P(A / B) is the probability of hypothesis A given data B also known as posterior probability; P(B / A) is the probability of data B given that the hypothesis A was true; P(A) is the probability of hypothesis A being true regardless of data also known as prior probability of A; and P(B) is the probability of the data regardless of the hypothesis. A naive Bayes algorithm may be generated by first transforming training data into a frequency table. Processor 804 may then calculate a likelihood table by calculating probabilities of different data entries and classification labels. Processor 804 may utilize a naive Bayes equation to calculate a posterior probability for each class. A class containing the highest posterior probability is the outcome of prediction. Still referring to FIG. 10, although Naive Bayes classifier may be primarily known as a probabilistic classification algorithm; however, it may also be considered a generative model described herein due to its capability of modeling the joint probability distribution P(X, Y) over observable variables X and target variable Y. In an embodiment, Naive Bayes classifier may be configured to make an assumption that the features X are conditionally independent given class label Y, allowing generative model to estimate the joint distribution as P(X, F) = P(Yy [iP(Xi I F), wherein P(Y) may be the prior probability of the class, and P(X |F) is the conditional probability of each feature given the class. One or more generative machine learning models containing Naive Bayes classifiers may be trained on labeled training data, estimating conditional probabilities P(X[ |F) and prior probabilities P(F) for each class; for instance, and without limitation, using techniques such as Maximum Likelihood Estimation (MLE). One or more generative machine learning models containing Naive Bayes classifiers may select a class label y according to prior distribution P(F), and for each feature X sample at least a value according to conditional distribution P(X y). Sampled feature values may then be combined to form one or more new data instance with selected class label y. In a non-limiting example, one or more generative machine learning models may include one or more Naive Bayes classifiers to generate new examples of ICE images based on CT scans and / or 3D models derived from CT scans (e.g., identified ICE views 1004), wherein the models may be trained using training data containing a plurality of features of input data as described herein and / or the like correlated to a plurality of ICE views.

[0218] Still referring to FIG. 10, in some cases, one or more generative machine learning models may include generative adversarial network (GAN). As used in this disclosure, a “generative adversarial network” is a type of artificial neural network with at least two sub models (e.g., neural networks), a generator, and a discriminator, that compete against each other in a process that ultimately results in the generator learning to generate new data samples, wherein the “generator” is a component of the GAN that learns to create hypothetical data by incorporating feedbacks from the “discriminator” configured to distinguish real data from the hypothetical data. In some cases, generator may learn to make discriminator classify its output as real. In an embodiment, discriminator may include a supervised machine learning model while generator may include an unsupervised machine learning model as described in further detail with reference to FIGS. 5-7. With continued reference to FIG. 10, in an embodiment, discriminator may include one or more discriminative models, i.e., models of conditional probability P(F|X = x) of target variable Y, given observed variable X. In an embodiment, discriminative models may learn boundaries between classes or labels in given training data. In a non-limiting example, discriminator may include one or more classifiers as described in further detail below with reference to FIG. 5 to distinguish between different categories e.g., real vs. fake, or states e.g., TRUE vs. FALSE within the context of generated data such as, without limitations, synthetic ICE frames 1008, and / or the like. In some cases, processor 804 may implement one or more classification algorithms such as, without limitation, Support Vector Machines (SVM), Logistic Regression, Decision Trees, and / or the like to define decision boundaries.

[0219] In a non-limiting example, and still referring to FIG. 10, generator of GAN may be responsible for creating synthetic data that resembles real ICE images. In some cases, GAN may be configured to receive CT scans and / or 3D models derived from CT scans as input and generates corresponding examples of ICE images containing information describing heart anatomy in different ICE views. On the other hand, discriminator of GAN may evaluate the authenticity of the generated content by comparing it to true ICE images, for example, discriminator may distinguish between genuine and generated content and providing feedback to generator to improve the model performance. Additionally, or alternatively, GAN may include a conditional GAN as an extension of the basic GAN as described herein that allows for generation of ICE images using pre-existing CT scans and / or 3D models derived from CT scans based on certain conditions or labels. In standard GAN, generator may produce samples from random noise, while in a conditional GAN, generator may produce samples based on random noise and a given condition or label.

[0220] With continued reference to FIG. 10, in other embodiments, one or more generative models may also include a variational autoencoder (VAE). As used in this disclosure, a “variational autoencoder” is an autoencoder (i.e., an artificial neural network architecture) whose encoding distribution is regularized during the model training process in order to ensure that its latent space includes desired properties allowing new data sample generation. In an embodiment, VAE may include a prior and noise distribution respectively, trained using expectationmaximization meta-algorithms such as, without limitation, probabilistic PC A, sparse coding, among others. In a non-limiting example, VEA may use a neural network as an amortized approach to jointly optimize across input data and output a plurality of parameters for corresponding variational distribution as it maps from a known input space to a low-dimensional latent space. Additionally, or alternatively, VAE may include a second neural network, for example, and without limitation, a decoder, wherein the “decoder” is configured to map from the latent space to the input space.

[0221] In a non-limiting example, and still referring to FIG. 10, VAE may be used by processor 804 to model complex relationships between CT scans and / or 3D models derived from CT scans. In some cases, VAE may encode input data into a latent space, capturing example ICE images. Such encoding process may include learning one or more probabilistic mappings from observed CT scans and / or 3D models derived from CT scans to a lower-dimensional latent representation. Latent representation may then be decoded back into the original data space, therefore reconstructing the 3D models representing example ICE images. In some cases, such decoding process may allow VAE to generate new examples or variations that are consistent with the learned distributions.

[0222] Additionally, or alternatively, and still referring to FIG. 10, processor 804 may be configured to continuously monitor synthetic ICE data generator. In an embodiment, processor 804 may configure discriminator to provide ongoing feedback and further corrections as needed to subsequent input data. An iterative feedback loop may be created as processor 804 continuously receive real-time data, identify errors (e.g., distance between synthetic ICE frame 1008 and real ICE images) as a function of real-time data, delivering corrections based on the identified errors, and monitoring subsequent model outputs and / or user feedbacks on the delivered corrections. In an embodiment, processor 804 may be configured to retrain one or more generative machine learning models within synthetic ICE data generator based on user modified ICE frames or update training data of one or more generative machine learning models within synthetic ICE data generator by integrating validated synthetic ICE frames (i.e., subsequent model output) into the original training data. In such embodiment, iterative feedback loop may allow synthetic ICE data generator to adapt to the user’s needs and performance requirements, enabling one or more generative machine learning models described herein to learn and update based on user responses and generated feedbacks.

[0223] With continued reference to FIG. 10, other exemplary embodiments of generative machine learning models may include, without limitation, long short-term memory networks (LSTMs), (generative pre-trained) transformer (GPT) models, mixture density networks (MDN), and / or the like. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various generative machine learning models may be used generating synthetic ICE frames 1008.

[0224] Still referring to FIG. 10, in a further non-limiting embodiment, synthetic ICE data generator 1012 may be further configured to generate a multi-model neural network that combines various neural network architectures described herein. In a non-limiting example, multi-model neural network may combine LSTM for time-series analysis with GPT models for natural language processing. Such fusion may be applied by computing device to generate synthetic ICE frames 1008. In some cases, multi-model neural network may also include a hierarchical multi-model neural network, wherein the hierarchical multi-model neural network may involve a plurality of layers of integration; for instance, and without limitation, different models may be combined at various stages of the network. Convolutional neural network (CNN) may be used for image feature extraction, followed by LSTMs for sequential pattern recognition, and a MDN at the end for probabilistic modeling. Other exemplary embodiments of multi-model neural network may include, without limitation, ensemble-based multi-model neural network, cross-modal fusion, adaptive multi-model network, among others. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various generative machine learning models may be used to generating synthetic ICE frames 1008 as described herein. As an ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various multi-model neural network and combination thereof that may be implemented by apparatus 800 in consistent with this disclosure.

[0225] Now referring to FIG. 11, an exemplary embodiment of a 3D VOR 1100 is illustrated. 3D VOR 1100 may be used to represent 3D object 1104. In an embodiment, 3D VOR 1100 may divide a 3D space 1108 into a grid of one or more cubic units e.g., voxels 1112, wherein each voxel 1112 represents a specific volume within 3D space 1108. In a non-limiting example, 3D object 1104 may include a cardiac anatomy pertaining to a subject.

[0226] Still referring to FIG. 11, in some cases, each voxel 1112 may act as a basic building block. In a non-limiting example, each voxel 1112 may be configured to represent a discrete portion of 3D space 1108. In an embodiment, each voxel 1112 may include a presence indicator as described above with reference to FIG. 8, which denotes whether the voxel is occupied or unoccupied. In such embodiment, the binary or continuous value may allow 3D VOR 1100 to map the presence or absence of material within each voxel 1112, creating a granular representation of 3D object 1104.

[0227] With continued reference to FIG. 11, in some cases, the resolution of 3D VOR 1100 may be determined by the size and number of voxels within the grid. In a non-limiting example, smaller voxel may provide a higher resolution, capturing finer details, while larger voxels offer a more generalized representation.

[0228] Still referring to FIG. 11, in an embodiment, voxels 1112 may be arranged in a regular pattern along three axis 816a-b, each pointing a distinct direction. In a non-limiting example, voxels 1112 may be arranged along x, y, and z axes, wherein such arrange may facilitate efficient manipulation and rendering of the 3D object 1104. In some cases, spatial features 820a- c such as, without limitation, edges, surfaces, textures, and any other spatial features as described above with reference to FIG. 8, may be extracted from 3D VOR 1100 by analyzing the relationships and patterns between neighboring voxels.

[0229] It is to be noted that any one or more of the aspects and embodiments described herein may be conveniently implemented using one or more machines (e.g., one or more computing devices that are utilized as a user computing device for an electronic document, one or more server devices, such as a document server, etc.) programmed according to the teachings of the present specification, as will be apparent to those of ordinary skill in the computer art. Appropriate software coding can readily be prepared by skilled programmers based on the teachings of the present disclosure, as will be apparent to those of ordinary skill in the software art. Aspects and implementations discussed above employing software and / or software modules may also include appropriate hardware for assisting in the implementation of the machine executable instructions of the software and / or software module.

[0230] Such software may be a computer program product that employs a machine-readable storage medium. A machine-readable storage medium may be any medium that is capable of storing and / or encoding a sequence of instructions for execution by a machine (e.g., a computing device) and that causes the machine to perform any one of the methodologies and / or embodiments described herein. Examples of a machine-readable storage medium include, but are not limited to, a magnetic disk, an optical disc (e.g, CD, CD-R, DVD, DVD-R, etc.), a magnetooptical disk, a read-only memory “ROM” device, a random access memory “RAM” device, a magnetic card, an optical card, a solid-state memory device, an EPROM, an EEPROM, and any combinations thereof. A machine-readable medium, as used herein, is intended to include a single medium as well as a collection of physically separate media, such as, for example, a collection of compact discs or one or more hard disk drives in combination with a computer memory. As used herein, a machine-readable storage medium does not include transitory forms of signal transmission.

[0231] Such software may also include information (e.g., data) carried as a data signal on a data carrier, such as a carrier wave. For example, machine-executable information may be included as a data-carrying signal embodied in a data carrier in which the signal encodes a sequence of instruction, or portion thereof, for execution by a machine (e.g., a computing device) and any related information (e.g., data structures and data) that causes the machine to perform any one of the methodologies and / or embodiments described herein.

[0232] Examples of a computing device include, but are not limited to, an electronic book reading device, a computer workstation, a terminal computer, a server computer, a handheld device (e.g., a tablet computer, a smartphone, etc.), a web appliance, a network router, a network switch, a network bridge, any machine capable of executing a sequence of instructions that specify an action to be taken by that machine, and any combinations thereof. In one example, a computing device may include and / or be included in a kiosk.

[0233] FIG. 12 shows a diagrammatic representation of one embodiment of a computing device in the exemplary form of a computer system 1200 within which a set of instructions for causing a control system to perform any one or more of the aspects and / or methodologies of the present disclosure may be executed. It is also contemplated that multiple computing devices may be utilized to implement a specially configured set of instructions for causing one or more of the devices to perform any one or more of the aspects and / or methodologies of the present disclosure. Computer system 1200 includes a processor 1204 and memory 1208 that communicate with each other, and with other components, via a bus 1212. Bus 1212 may include any of several types of bus structures including, but not limited to, memory bus, memory controller, a peripheral bus, a local bus, and any combinations thereof, using any of a variety of bus architectures.

[0234] Processor 1204 may include any suitable processor, such as without limitation a processor incorporating logical circuitry for performing arithmetic and logical operations, such as an arithmetic and logic unit (ALU), which may be regulated with a state machine and directed by operational inputs from memory and / or sensors; processor 1204 may be organized according to Von Neumann and / or Harvard architecture as a non-limiting example. Processor 1204 may include, incorporate, and / or be incorporated in, without limitation, a microcontroller, microprocessor, digital signal processor (DSP), Field Programmable Gate Array (FPGA), Complex Programmable Logic Device (CPLD), Graphical Processing Unit (GPU), general purpose GPU, Tensor Processing Unit (TPU), analog or mixed signal processor, Trusted Platform Module (TPM), a floating point unit (FPU), and / or system on a chip (SoC).

[0235] Memory 1208 may include various components (e.g., machine-readable media) including, but not limited to, a random-access memory component, a read only component, and any combinations thereof. In one example, a basic input / output system 1216 (BIOS), including basic routines that help to transfer information between elements within computer system 1200, such as during start-up, may be stored in memory 1208. Memory 1208 may also include (e.g., stored on one or more machine-readable media) instructions (e.g., software) 1220 embodying any one or more of the aspects and / or methodologies of the present disclosure. In another example, memory 1208 may further include any number of program modules including, but not limited to, an operating system, one or more application programs, other program modules, program data, and any combinations thereof.

[0236] Computer system 1200 may also include a storage device 1224. Examples of a storage device (e.g., storage device 1224) include, but are not limited to, a hard disk drive, a magnetic disk drive, an optical disc drive in combination with an optical medium, a solid-state memory device, and any combinations thereof. Storage device 1224 may be connected to bus 1212 by an appropriate interface (not shown). Example interfaces include, but are not limited to, SCSI, advanced technology attachment (ATA), serial ATA, universal serial bus (USB), IEEE 1394 (FIREWIRE), and any combinations thereof. In one example, storage device 1224 (or one or more components thereof) may be removably interfaced with computer system 1200 (e.g., via an external port connector (not shown)). Particularly, storage device 1224 and an associated machine-readable medium 1228 may provide nonvolatile and / or volatile storage of machine- readable instructions, data structures, program modules, and / or other data for computer system 1200. In one example, software 1220 may reside, completely or partially, within machine- readable medium 1228. In another example, software 1220 may reside, completely or partially, within processor 1204.

[0237] Computer system 1200 may also include an input device 1232. In one example, a user of computer system 1200 may enter commands and / or other information into computer system 1200 via input device 1232. Examples of an input device 1232 include, but are not limited to, an alpha-numeric input device (e.g., a keyboard), a pointing device, a joystick, a gamepad, an audio input device (e. , a microphone, a voice response system, etc.), a cursor control device (e.g., a mouse), a touchpad, an optical scanner, a video capture device (e.g., a still camera, a video camera), a touchscreen, and any combinations thereof. Input device 1232 may be interfaced to bus 1212 via any of a variety of interfaces (not shown) including, but not limited to, a serial interface, a parallel interface, a game port, a USB interface, a FIREWIRE interface, a direct interface to bus 1212, and any combinations thereof. Input device 1232 may include a touch screen interface that may be a part of or separate from display 1236, discussed further below. Input device 1232 may be utilized as a user selection device for selecting one or more graphical representations in a graphical interface as described above.

[0238] A user may also input commands and / or other information to computer system 1200 via storage device 1224 (e.g., a removable disk drive, a flash drive, etc.) and / or network interface device 1240. A network interface device, such as network interface device 1240, may be utilized for connecting computer system 1200 to one or more of a variety of networks, such as network 1244, and one or more remote devices 1248 connected thereto. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone / voice provider (e.g., a mobile communications provider data and / or voice network), a direct connection between two computing devices, and any combinations thereof. A network, such as network 1244, may employ a wired and / or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software 1220, etc.) may be communicated to and / or from computer system 1200 via network interface device 1240.

[0239] Computer system 1200 may further include a video display adapter 1252 for communicating a displayable image to a display device, such as display device 1236. Examples of a display device include, but are not limited to, a liquid crystal display (LCD), a cathode ray tube (CRT), a plasma display, a light emitting diode (LED) display, and any combinations thereof. Display adapter 1252 and display device 1236 may be utilized in combination with processor 1204 to provide graphical representations of aspects of the present disclosure. In addition to a display device, computer system 1200 may include one or more other peripheral output devices including, but not limited to, an audio speaker, a printer, and any combinations thereof. Such peripheral output devices may be connected to bus 1212 via a peripheral interface 1256. Examples of a peripheral interface include, but are not limited to, a serial port, a USB connection, a FIREWIRE connection, a parallel connection, and any combinations thereof.

[0240] The foregoing has been a detailed description of illustrative embodiments of the invention. Various modifications and additions can be made without departing from the spirit and scope of this invention. Features of each of the various embodiments described above may be combined with features of other described embodiments as appropriate in order to provide a multiplicity of feature combinations in associated new embodiments. Furthermore, while the foregoing describes a number of separate embodiments, what has been described herein is merely illustrative of the application of the principles of the present invention. Additionally, although particular methods herein may be illustrated and / or described as being performed in a specific order, the ordering is highly variable within ordinary skill to achieve methods and apparatuses according to the present disclosure. Accordingly, this description is meant to be taken only by way of example, and not to otherwise limit the scope of this invention. Exemplary embodiments have been disclosed above and illustrated in the accompanying drawings. It will be understood by those skilled in the art that various changes, omissions and additions may be made to that which is specifically disclosed herein without departing from the spirit and scope of the present invention.

Claims

WHAT IS CLAIMED IS:

1. An apparatus for object pose estimation in a medical image, the apparatus comprising: at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segment the plurality of echo depth maps; determine a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps; generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generate a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum comprises: training a pose estimation model using pose estimation training data, wherein the pose estimation training data comprises exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

2. The apparatus of claim 1, wherein receiving the plurality of sets of echo data comprises receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

3. The apparatus of claim 1, wherein the plurality of sets of echo data is related to a cadaveric organ with a catheter inserted into it.

4. The apparatus of claim 1, wherein segmenting the plurality of echo depth maps comprises: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

5. The apparatus of claim 1, wherein determining the depth datum comprises: training a depth model using depth training data, wherein the depth training datacomprises exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

6. The apparatus of claim 1, wherein the pose estimation model comprises a regression model and the pose estimation training data comprises synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

7. The apparatus of claim 1, wherein: the pose datum comprises a 6D pose datum; and generating the pose datum comprises generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

8. The apparatus of claim 7, further comprising: a catheter, wherein the catheter comprises: the object of interest comprising a distinct shape; and an electrode comprising the second object of interest.

9. The apparatus of claim 1, wherein the memory contains instructions further configuring the at least a processor to: generate a 3D model as a function of the 3D point cloud; and generate a user interface including the 3D model; and display the user interface including the 3D model on a display device.

10. The apparatus of claim 1, wherein the memory contains instructions further configuring the at least a processor to generate an anatomical datum as a function of the pose datum, wherein the anatomical datum comprises a dimension datum of the object of interest.

11. The apparatus of claim 1, wherein the pose datum comprises a five degree (5D) pose datum.

12. A method for object pose estimation in a medical image, the method comprising: receiving, using at least a processor, a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segmenting, using the at least a processor, the plurality of echo depth maps; determining, using the at least a processor, a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps;generating, using the at least a processor, a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generating, using the at least a processor, a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum comprises: training a pose estimation model using pose estimation training data, wherein the pose estimation training data comprises exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

13. The method of claim 12, wherein receiving the plurality of sets of echo data comprises receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

14. The method of claim 12, wherein the plurality of sets of echo data is related to a cadaveric organ with an inserting device of a catheter inserted into it.

15. The method of claim 12, wherein segmenting the plurality of echo depth maps comprises: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

16. The method of claim 12, wherein determining the depth datum comprises: training a depth model using depth training data, wherein the depth training data comprises exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

17. The method of claim 12, wherein the pose estimation model comprises a regression model and the pose estimation training data comprises synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

18. The method of claim 12, wherein: the pose datum comprises a 6D pose datum; and generating the pose datum comprises generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

19. The method of claim 18, wherein:the object of interest of a catheter comprises a distinct shape; and an electrode of the catheter comprises the second object of interest.

20. The method of claim 12, further comprising: generating, by the at least a processor, a 3D model as a function of the 3D point cloud; and generating, by the at least a processor, a user interface including the 3D model; and displaying, by the at least a processor, the user interface including the 3D model on a display device.

21. The method of claim 12, further comprising: generating, using the at least a processor, an anatomical datum as a function of the pose datum, wherein the anatomical datum comprises a dimension datum of the object of interest.

22. The method of claim 12, wherein the pose datum comprises a five degree (5D) pose datum.

23. An apparatus for object pose estimation in a medical image, the apparatus comprising: one or more ultrasound imaging systems located on a surface of a subject; an object of interest configured to be placed within the subject; at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive a plurality of sets of echo data from the one or more ultrasound imaging systems, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segment the plurality of echo depth maps; determine a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps; generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generate a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum comprises: training a pose estimation model using pose estimation training data,wherein the pose estimation training data comprises exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

24. The apparatus of claim 23, wherein receiving the plurality of sets of echo data comprises receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

25. The apparatus of claim 23, wherein the plurality of sets of echo data is related to a cadaveric organ with a catheter inserted into it.

26. The apparatus of claim 23, wherein segmenting the plurality of echo depth maps comprises: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

27. The apparatus of claim 23, wherein determining the depth datum comprises: training a depth model using depth training data, wherein the depth training data comprises exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

28. The apparatus of claim 23, wherein the pose estimation model comprises a regression model and the pose estimation training data comprises synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

29. The apparatus of claim 23, wherein: the pose datum comprises a 6D pose datum; and generating the pose datum comprises generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

30. The apparatus of claim 29, further comprising: a catheter, wherein the catheter comprises: the object of interest comprising a distinct shape; and an electrode comprising the second object of interest.

31. The apparatus of claim 23, wherein the memory contains instructions further configuringthe at least a processor to: generate a 3D model as a function of the 3D point cloud; generate a user interface including the 3D model; and display the user interface including the 3D model on a display device.

32. The apparatus of claim 23, wherein the memory contains instructions further configuring the at least a processor to generate an anatomical datum as a function of the pose datum, wherein the anatomical datum comprises a dimension datum of the object of interest.

33. The apparatus of claim 23, wherein the pose datum comprises a five degree (5D) pose datum.

34. A method for object pose estimation in a medical image, the method comprising: placing, one or more ultrasound imaging systems on a surface of a subject; positioning an object of interest configured to be within the subject; receiving, using at least a processor and from the one or more ultrasound imaging systems, a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps; segmenting, using the at least a processor, the plurality of echo depth maps; determining, using the at least a processor, a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps; generating, using the at least a processor, a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum; and generating, using the at least a processor, a pose datum of the object of interest as a function of the 3D point cloud, wherein generating the pose datum comprises: training a pose estimation model using pose estimation training data, wherein the pose estimation training data comprises exemplary 3D point clouds correlated to exemplary pose datums; and generating the pose datum using the trained pose estimation model.

35. The method of claim 34, wherein receiving the plurality of sets of echo data comprises receiving the plurality of sets of echo data from a plurality of echo transducers, wherein each of the plurality of echo transducers is located at a different angle.

36. The method of claim 34, wherein the plurality of sets of echo data is related to a cadaveric organ with an inserting device of a catheter inserted into it.

37. The method of claim 34, wherein segmenting the plurality of echo depth maps comprises: extracting object features of the object of interest from the plurality of echo depth maps; and segmenting the plurality of echo depth maps as a function of the object features.

38. The method of claim 34, wherein determining the depth datum comprises: training a depth model using depth training data, wherein the depth training data comprises exemplary segmented echo depth maps correlated to exemplary depth datums; and determining the depth datum using the trained depth model.

39. The method of claim 34, wherein the pose estimation model comprises a regression model and the pose estimation training data comprises synthetic 3D point clouds generated from a computer aided design (CAD) model of the object of interest.

40. The method of claim 34, wherein: the pose datum comprises a 6D pose datum; and generating the pose datum comprises generating the 6D pose datum related to a second object of interest relative to the object of interest as a function of a rigidity constraint between the object and the second object of interest.

41. The method of claim 40, wherein: the object of interest of a catheter comprises a distinct shape; and an electrode of the catheter comprises the second object of interest.

42. The method of claim 34, further comprising: generating, by the at least a processor, a 3D model as a function of the 3D point cloud; generating, by the at least a processor, a user interface including the 3D model; and displaying, by the at least a processor, the user interface including the 3D model on a display device.

43. The method of claim 34, further comprising: generating, using the at least a processor, an anatomical datum as a function of the pose datum, wherein the anatomical datum comprises a dimension datum of the object of interest.

44. The method of claim 34, wherein the pose datum comprises a five degree (5D) pose datum.

45. An apparatus for object pose estimation in a medical image, the apparatus comprising: one or more ultrasound imaging systems located on a surface of a subject; an object of interest configured to be placed within the subject; at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive a plurality of sets of echo data from the one or more ultrasound imaging systems, wherein the plurality of sets of echo data are configured for generation of a plurality of echo depth maps; generate a three dimensional (3D) point cloud related to the object of interest as a function of the plurality of sets of echo data; and generate a pose datum of the object of interest as a function of the 3D point cloud using a pose estimation model.

46. The apparatus of claim 45, wherein the one or more ultrasound imaging systems comprises: a first ultrasound imaging system located at a first position on the surface of the subject; and a second ultrasound imaging system located at a second position on the surface of the subject.

47. The apparatus of claim 45, wherein a first set of echo data of the plurality of sets of echo data and a second set of echo data of the plurality of sets of echo data comprise differing views of the object of interest.

48. The apparatus of claim 45, wherein the at least a processor is further configured to segment the plurality of echo depth maps to generate a plurality of segmented echo depth maps.

49. The apparatus of claim 48, wherein segmenting the plurality of echo depth maps comprises: extracting the plurality of echo depth maps as a function of the plurality of sets of echo data; identifying a spatial expanse of the object of interest as a function of at least an object feature; andsegmenting the plurality of echo depth maps as a function of the spatial expanse.

50. The apparatus of claim 48, wherein the at least a processor is further configured to determine a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps.

51. The apparatus of claim 48, wherein the at least a processor is further configured to determine a depth datum using a depth model, wherein: the depth model comprises a convolutional neural network (CNN); and the at least a processor is further configured to use the depth model to predict the depth datum at each pixel of the plurality of segmented echo depth maps.

52. The apparatus of claim 45, wherein generating the 3D point cloud comprises aggregating each 3D point of a plurality of 3D points of the object of interest, wherein each 3D point of the plurality of 3D points is generated by converting a 2D pixel coordinate of a segmented echo depth map into a 3D coordinate by adding a depth datum as a z-value.

53. The apparatus of claim 45, wherein the at least a processor is further configured to generate a 3D model as a function of the 3D point cloud, wherein generating the 3D model comprises applying at least a 3D reconstruction algorithm to the 3D point cloud.

54. The apparatus of claim 45, wherein generating the pose datum comprises determining a pose of a sub-part of the object of interest, wherein: the sub-part has a fixed spatial relationship to a plurality of electrodes on a catheter; and determining the pose of a sub-part of the object of interest comprises calculating a pose of the plurality of electrodes as a function of the pose of the sub-part of the object of interest and a rigidity constraint between the sub-part of the object of interest and the plurality of electrodes.

55. A method for object pose estimation in a medical image, the method comprising: receiving, by at least a processor, a plurality of sets of echo data from one or more ultrasound imaging systems located on a surface of a subject, wherein the plurality of sets of echo data are configured for generation of a plurality of echo depth maps; generating, using the at least a processor, a three dimensional (3D) point cloud related to the object of interest as a function of the plurality of sets of echo data; and generating, using the at least a processor, a pose datum of the object of interest as afunction of the 3D point cloud using a pose estimation model.

56. The method of claim 55, wherein the one or more ultrasound imaging systems comprises: a first ultrasound imaging system located at a first position on the surface of the subject; and a second ultrasound imaging system located at a second position on the surface of the subject.

57. The method of claim 55, wherein a first set of echo data of the plurality of sets of echo data and a second set of echo data of the plurality of sets of echo data comprise differing views of the object of interest.

58. The method of claim 55, further comprising segmenting the plurality of echo depth maps to generate a plurality of segmented echo depth maps.

59. The method of claim 58, wherein segmenting the plurality of echo depth maps comprises: extracting the plurality of echo depth maps as a function of the plurality of sets of echo data; identifying a spatial expanse of the object of interest as a function of at least an object feature; and segmenting the plurality of echo depth maps as a function of the spatial expanse.

60. The method of claim 58, further comprising determining a depth datum related to pixels of the object of interest as a function of the plurality of segmented echo depth maps.

61. The method of claim 58, further comprising determining a depth datum using a depth model, wherein: the depth model comprises a convolutional neural network (CNN); and determining a depth datum further comprises using the depth model to predict the depth datum at each pixel of the plurality of segmented echo depth maps.

62. The method of claim 55, wherein generating the 3D point cloud comprises aggregating each 3D point of a plurality of 3D points of the object of interest, wherein each 3D point of the plurality of 3D points is generated by converting a 2D pixel coordinate of a segmented echo depth map into a 3D coordinate by adding a depth datum as a z-value.

63. The method of claim 55, further comprising generating a 3D model as a function of the 3D point cloud, wherein generating the 3D model comprises applying at least a 3D reconstruction algorithm to the 3D point cloud.

64. The method of claim 55, wherein generating the pose datum comprises determining a pose of a sub-part of the object of interest, wherein: the sub-part has a fixed spatial relationship to a plurality of electrodes on a catheter; and determining the pose of a sub-part of the object of interest comprises calculating a pose of the plurality of electrodes as a function of the pose of the sub-part of the object of interest and a rigidity constraint between the sub-part of the object of interest and the plurality of electrodes.

Images