Pre-trained diffusion model for downstream medical vision tasks

Abstract

Aspects as described herein provide a computer-implemented method, system, and computer program product that use a pre-trained diffusion model for downstream medical vision tasks. Accessing a decoder stage of the pre-trained diffusion model can provide features that exhibit object localization based on frames of a surgical video. The features can be used to perform one or more medical vision tasks.

Description

PRE-TRAINED DIFFUSION MODEL FOR DOWNSTREAM MEDICAL VISIONTASKSCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of U.S. Provisional Patent Application Serial No. 63 / 645,373, filed May 10, 2024, the entire content of which is incorporated herein by referenceBACKGROUND

[0002] The present disclosure relates in general to computing technology and relates more particularly to computing technology using a pre-trained diffusion model for downstream medical vision tasks.

[0003] Computer-assisted systems, particularly computer-assisted surgery systems (CASs), rely on video data digitally captured during a surgery. Such video data can be stored and / or streamed. In some cases, the video data can be used to augment a person’s physical sensing, perception, and reaction capabilities. For example, such systems can effectively provide the information corresponding to an expanded field of vision, both temporal and spatial, that enables a person to adjust current and future actions based on the part of an environment not included in his or her physical field of view. Alternatively, or in addition, the video data can be stored and / or transmitted for several purposes, such as archival, training, post-surgery analysis, and / or patient consultation.SUMMARY

[0004] Aspects of the disclosure can include a computer-implemented method, system and computer program product that use a pre-trained diffusion model for downstream medical vision tasks.

[0005] The above features and advantages, and other features and advantages, of the disclosure are readily apparent from the following detailed description when taken in connection with the accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The specifics of the exclusive rights described herein are particularly pointed out and distinctly claimed in the claims at the conclusion of the specification. The foregoing and other features and advantages of the aspects of the disclosure are apparent from the following detailed description taken in conjunction with the accompanying drawings in which:

[0007] FIG. 1 depicts a computer-assisted surgery (CAS) system according to one or more aspects;

[0008] FIG. 2 depicts a surgical procedure system according to one or more aspects;

[0009] FIG. 3 depicts a system for analyzing video and data according to one or more aspects;

[0010] FIG. 4 depicts a diffusion feature extraction process according to one or more aspects;

[0011] FIG. 5 depicts a diffusion feature extraction pipeline according to one or more aspects;

[0012] FIG. 6 depicts a rendering of features from a decoder at different levels according to one or more aspects;

[0013] FIG. 7 depicts a mask prediction algorithm according to one or more aspects;

[0014] FIG. 8 depicts an example of key-point tracking according to one or more aspects;

[0015] FIG. 9 depicts an example of segmentation according to one or more aspects;

[0016] FIG. 10 depicts a flowchart of a process according to one or more aspects; and

[0017] FIG. 11 depicts a block diagram of a computer system according to one or more aspects.

[0018] The diagrams depicted herein are illustrative. There can be many variations to the diagrams and / or the operations described herein without departing from the spirit of the described aspects. For instance, the actions can be performed in a differing order, or actions can be added, deleted, or modified. Also, the term “coupled” and variations thereof describe having a communications path between two elements and do not imply a direct connection between the elements with no intervening elements / connections between them. All of these variations are considered a part of the specification.DETAILED DESCRIPTION

[0019] Exemplary aspects of the technical solutions described herein include systems and methods using a pre-trained diffusion model for downstream medical vision tasks. In some aspects, the pre-trained diffusion model can be used for segmentation of video, for instance, to predict segmentation of frames of surgical video. In other aspects, the pre-trained diffusion model can be used for key -point tracking of features in frames of surgical video.

[0020] Training separate neural models for different downstream vision tasks is computationally expensive and typically demands different manually annotated labels. This is even more difficult in the context of medical imaging, where there is data scarcity, and medical expertise is typically needed for accurate annotation. According to aspects as further described herein, a pre-trained text-to-image diffusion model (DM) can be used as a single backbone feature extractor for one or more downstream medical vision tasks without further training / fine-tuning. Text-to-image DMs (e.g., Stable Diffusion - (SD)) can be used for high-resolution image generation, editing, inpainting, and other such tasks. Prior to producing generative image content, internal representations can be produced within an SD model, which can be used for non- generative downstream tasks. Although typically used for image generation, internal feature maps of a denoiser in an SD model can contain highly localized semantic information, that is robust to pose and geometry. To this end, up-sampling block feature maps of a pre-trained SD model can be used as backbone features for key-point tracking and segmentation in medical images. This can yield competitive results without training / fine-tuning and can be used as a single pre-trained foundation model with taskspecific adaptation.

[0021] As one example, a video frame of a surgical video can be passed to a pre-trained version of an SD model to extract internal feature maps of an up-sampling block, which act as a backbone feature. For key-point tracking, given a source point on the first frame, a pixel-wise cosine can be computed for similarity between the source point SD model feature and features of subsequent frames for correspondence. For segmentation, a segmentation mask of the first frame can be used to create a spatial neighborhood and extract features. Similarity can be calculated between the features and features of subsequent frames to generate segmentation masks. In some aspects, the results maybe directly used as outputs on a display or provided as a starting point for subsequent analysis and processing, such as input to another machine learning model or as estimates for further adjustment by a human annotator.

[0022] Aspects can be used, for example, for unsupervised semantic segmentation in laparoscopic cholecystectomy (LC) surgical videos, targeting localization of critical anatomical structures and instruments. The approach, as described herein, addresses the challenges of pixel-level annotations and label inconsistencies inherent in existing datasets. Leveraging the inherent object localization capabilities of pre-trained text-to- image diffusion models, representative features can be extracted from surgical frames without training or fine-tuning. These features can exhibit superior object localization and consistent semantics across different decoder levels and temporal frames. The segmentation framework can utilize these features, along with cross-frame interactions via an affinity matrix to ensure temporal continuity in the segmentation process.

[0023] Surgical video analysis can provide surgeons with useful analytics and insights on minimally invasive computer-assisted surgeries as well as other types of medical procedures. Semantic segmentation deals with pixel-level classification of surgical frames, localizing important instruments and anatomies present in the scene. Semantic segmentation can aid computer-assisted intervention (CAI) in three ways, (i) Preoperative surgeon training can prepare surgeons in correctly identifying and segregating critical organ structure and surgery-planning, (ii) Intra-operative guidance can provide real-time feedback to surgeons during operation, helping them establish Critical View of Safety (CVS) to minimize the risk of unwanted organ injury, (iii) Post-operative case study can equip medical practitioners with analytics and insights about failure and success cases.

[0024] As one example, a training-free framework can be provided for video semantic segmentation of Laparoscopic Cholecystectomy (LC) surgery, that deals with surgical extraction of gallbladder. The accurate identification of certain anatomies likes cystic duct and artery is critical to avoid the risk of bile duct injury, a serious injury associated with severe post-operative morbidity, lower long-term survival and negative impact on a patient’s quality of life. Surgical video semantic segmentation can help avoid these scenarios by providing accurate localization of crucial instruments and anatomies.

[0025] Semantic segmentation of surgical videos is non-trivial. Firstly, the high annotation cost of pixel-level masks makes large-scale fully-supervised training of segmentation networks (with mask-photo pairs) unfeasible. Second, producing accurate annotations is even more difficult for surgical videos. Most of the existing datasets were annotated with a semi-automated segmentation pipeline, which can inherently introduce label inconsistencies to some extent. Consequently, fully supervised training might be erroneous in such scenarios.

[0026] Text-to-image diffusion models can provide high resolution image and video synthesis. Apart from image generation, DMs can also depict exceptional performance in semantic local editing, object detection, and correspondence learning, among different tasks. Internal representations can inherently encompass some form of object localization and grouping, even though not specifically targeted during training. The internal representations of pre-trained text-to-image latent diffusion models can be used to extract representative features from surgical frames that contain useful geometry and object-localization. These backbone features can be used for unsupervised semantic segmentation of minimally-invasive surgical videos without further training or finetuning. This can alleviate issues pertaining to pixel-level annotation cost and noisy ground truths.

[0027] Internal features of a pre-trained diffusion model can inherently hold superior object localization, making the pre-trained diffusion model suitable for semantic segmentation. These features can showcase different feature-granularity based on different decoder levels of the architecture. Moreover, semantics of these features can enforce temporal-consistency. The framework can include identifying a most optimal layer, timestep, and other network parameters tailored to the task of semantic segmentation. A segmentation module can utilize these features to perform semantic segmentation on an entire video of frames without the need of ground-truth masks. In the segmentation module, cross-frame interactions can be utilized in the temporal direction via an affinity matrix to maintain temporal-continuity.

[0028] Diffusion models can generate images by iterative elimination of noise from a 2D isotropic Gaussian noise image. Two reciprocal processes can be used - (i) forward and (ii) reverse diffusion. The forward process adds random Gaussian noise to a clean Vn ]IjAxw’x3 image0from a training dataset in an iterative manner (for t timesteps) tocreate a noisy image:here, e~N(0, 1), andis the noise schedule wherewith t~U(0,T).During the reverse process, a denoising UNet Ue can be trained (with an h objective), which estimates the input noise e ~ Ue (xt , t) from the noisy image xt at each t. Once trained, Ue can recover the original image from a noisy image. The inference procedure starts from a random 2D noise XT~N(0, I). A well-trained Ue can be employed iteratively (for T timesteps) to remove noise from each timestep progressively to get a cleaner image xt-i. This eventually yields one of the cleanest samples xo from the target distribution. The denoising process can also be “conditional” by governing Ue with an additional conditional signal c (e.g., text prompt, dense map, etc.). Accordingly, Ue (xt , t,c) denoises xt while being governed by c through cross-attention.

[0029] The training and inference procedure of a general diffusion model can be time- consuming as it operates on the full image resolution (i.e.,Contrarily, in a Latent Diffusion Model (i.e., Stable Diffusion - SD), denoising occurs on the encoder latent space, making it much faster and stable. In the first stage, SD can employ a pre-trained variational autoencoder (an encoder E( ) and a decoder D( ) in sequence).E( ) converts the input image into a latent representation. in the second stage, SD trains a UNet Ue, that performs denoising directly on the latent images. Ue can include, for instance, 12 encoding, 1 bottleneck, and 12 decoding blocks. Inside these encoding and decoding layers, there can be 4 down-sampling (Ul-4d ) and 4 up-sampling (Ul-4u ) layers respectively. A pre-trained language encoder T (•) can convert a textual prompt c into a token sequence, that governs Ue through cross-attention. Ue can be trained over an h objective:During testing, E( ) can be discarded and a noisy latent zt can be sampled directly as: zt~N(0, 1). Ue can remove noise from zt iteratively over T timesteps (conditioned on c) to yield a denoised latent imageThe final image can be generated as:

[0030] Apart from high-resolution text-to-image generation, SD can also delineate excellent performance in semantic local editing, object detection, and correspondence learning, etc. Internal representations can inherently encompass some form of objectlocalization and grouping, even though not specifically targeted during training. Internal feature maps of an SD decoder can contain highly localized hierarchical object groupings, where the semantics of this localized information are temporally-consistent.

[0031] Turning now to FIG. 1, an example computer-assisted system (CAS) system 100 is generally shown in accordance with one or more aspects. The CAS system 100 includes at least a computing system 102, a video recording system 104, and a surgical instrumentation system 106. As illustrated in FIG. 1, an actor 112 can be medical personnel that uses the CAS system 100 to perform a surgical procedure on a patient 110. Medical personnel can be a surgeon, assistant, nurse, administrator, or any other actor that interacts with the CAS system 100 in a surgical environment. The surgical procedure can be any type of surgery, such as but not limited to cataract surgery, laparoscopic cholecystectomy, endoscopic endonasal transsphenoidal approach (eTSA) to resection of pituitary adenomas, or any other surgical procedure. In other examples, actor 112 can be a technician, an administrator, an engineer, or any other such personnel that interacts with the CAS system 100. For example, actor 112 can record data from the CAS system 100, configure / update one or more attributes of the CAS system 100, review past performance of the CAS system 100, repair the CAS system 100, and / or the like including combinations and / or multiples thereof.

[0032] A surgical procedure can include multiple phases, and each phase can include one or more surgical actions. A “surgical action” can include an incision, a compression, a stapling, a clipping, a suturing, a cauterization, a sealing, or any other such actions performed to complete a phase in the surgical procedure. A “phase” represents a surgical event that is composed of a series of steps (e.g., closure). A “step” refers to the completion of a named surgical objective (e.g., hemostasis). During each step, certain surgical instruments 108 (e.g., forceps) are used to achieve a specific objective by performing one or more surgical actions. In addition, a particular anatomical structure of the patient may be the target of the surgical action(s).

[0033] The video recording system 104 includes one or more cameras 105, such as operating room cameras, endoscopic cameras, and / or the like including combinationsand / or multiples thereof. The cameras 105 capture video data of the surgical procedure being performed. The video recording system 104 includes one or more video capture devices that can include cameras 105 placed in the surgical room to capture events surrounding (i.e., outside) the patient being operated upon. The video recording system 104 further includes cameras 105 that are passed inside (e.g., endoscopic cameras) the patient 110 to capture endoscopic data. The endoscopic data provides video and images of the surgical procedure.

[0034] The computing system 102 includes one or more memory devices, one or more processors, a user interface device, among other components. All or a portion of the computing system 102 shown in FIG. 1 can be implemented for example, by all or a portion of computer system 800 of FIG. 11. Computing system 102 can execute one or more computer-executable instructions. The execution of the instructions facilitates the computing system 102 to perform one or more methods, including those described herein. The computing system 102 can communicate with other computing systems via a wired and / or a wireless network. In one or more examples, the computing system 102 includes one or more trained machine learning models that can detect and / or predict features of / from the surgical procedure that is being performed or has been performed earlier. Features can include structures, such as anatomical structures, surgical instruments 108 in the captured video of the surgical procedure. Features can further include events, such as phases and / or actions in the surgical procedure. Features that are detected can further include the actor 112 and / or patient 110. Based on the detection, the computing system 102, in one or more examples, can provide recommendations for subsequent actions to be taken by the actor 112. Alternatively, or in addition, the computing system 102 can provide one or more reports based on the detections. The detections by the machine learning models can be performed in an autonomous or semi-autonomous manner.

[0035] The machine learning models can include artificial neural networks, such as deep neural networks, convolutional neural networks, recurrent neural networks, vision transformers, encoders, decoders, or any other type of machine learning model. The machine learning models can be trained in a supervised, unsupervised, or hybrid manner. The machine learning models can be trained to perform detection and / or prediction using one or more types of data acquired by the CAS system 100. Forexample, the machine learning models can use the video data captured via the video recording system 104. Alternatively, or in addition, the machine learning models use the surgical instrumentation data from the surgical instrumentation system 106. In yet other examples, the machine learning models use a combination of video data and surgical instrumentation data.

[0036] Additionally, in some examples, the machine learning models can also use audio data captured during the surgical procedure. The audio data can include sounds emitted by the surgical instrumentation system 106 while activating one or more surgical instruments 108. Alternatively, or in addition, the audio data can include voice commands, snippets, or dialog from one or more actors 112. The audio data can further include sounds made by the surgical instruments 108 during their use.

[0037] In one or more examples, the machine learning models can detect surgical actions, surgical phases, anatomical structures, surgical instruments, and various other features from the data associated with a surgical procedure. The detection can be performed in real-time in some examples. Alternatively, or in addition, the computing system 102 analyzes the surgical data, i.e., the various types of data captured during the surgical procedure, in an offline manner (e.g., post-surgery). In one or more examples, the machine learning models detect surgical phases based on detecting some of the features, such as the anatomical structure, surgical instruments, and / or the like including combinations and / or multiples thereof.

[0038] A data collection system 150 can be employed to store the surgical data, including the video(s) captured during the surgical procedures. The data collection system 150 includes one or more storage devices 152. The data collection system 150 can be a local storage system, a cloud-based storage system, or a combination thereof. Further, the data collection system 150 can use any type of cloud-based storage architecture, for example, public cloud, private cloud, hybrid cloud, and / or the like including combinations and / or multiples thereof. In some examples, the data collection system can use a distributed storage, i.e., the storage devices 152 are located at different geographic locations. The storage devices 152 can include any type of electronic data storage media used for recording machine-readable data, such as semiconductor-based, magnetic-based, optical-based storage media, and / or the like including combinations and / or multiples thereof. For example, the data storage media can include flash-basedsolid-state drives (SSDs), magnetic-based hard disk drives, magnetic tape, optical discs, and / or the like including combinations and / or multiples thereof.

[0039] In one or more examples, the data collection system 150 can be part of the video recording system 104, or vice-versa. In some examples, the data collection system 150, the video recording system 104, and the computing system 102, can communicate with each other via a communication network, which can be wired, wireless, or a combination thereof. The communication between the systems can include the transfer of data (e.g., video data, instrumentation data, and / or the like including combinations and / or multiples thereof), data manipulation commands (e.g., browse, copy, paste, move, delete, create, compress, and / or the like including combinations and / or multiples thereof), data manipulation results, and / or the like including combinations and / or multiples thereof. In one or more examples, the computing system 102 can manipulate the data already stored / being stored in the data collection system 150 based on outputs from the one or more machine learning models (e.g., phase detection, anatomical structure detection, surgical tool detection, and / or the like including combinations and / or multiples thereof). Alternatively, or in addition, the computing system 102 can manipulate the data already stored / being stored in the data collection system 150 based on information from the surgical instrumentation system 106.

[0040] In one or more examples, the video captured by the video recording system 104 is stored on the data collection system 150. In some examples, the computing system 102 curates parts of the video data being stored on the data collection system 150. In some examples, the computing system 102 filters the video captured by the video recording system 104 before it is stored on the data collection system 150. Alternatively, or in addition, the computing system 102 filters the video captured by the video recording system 104 after it is stored on the data collection system 150.

[0041] Turning now to FIG. 2, a surgical procedure system 200 is generally shown according to one or more aspects. The example of FIG. 2 depicts a surgical procedure support system 202 that can include or may be coupled to the CAS system 100 of FIG. 1. The surgical procedure support system 202 can acquire image or video data using one or more cameras 204. The surgical procedure support system 202 can also interface with one or more sensors 206 and / or one or more effectors 208. The sensors 206 may be associated with surgical support equipment and / or patient monitoring. The effectors208 can be robotic components or other equipment controllable through the surgical procedure support system 202. The surgical procedure support system 202 can also interact with one or more user interfaces 210, such as various input and / or output devices. The surgical procedure support system 202 can store, access, and / or update surgical data 214 associated with a training dataset and / or live data as a surgical procedure is being performed on patient 110 of FIG. 1. The surgical procedure support system 202 can store, access, and / or update surgical objectives 216 to assist in training and guidance for one or more surgical procedures. User configurations 218 can track and store user preferences.

[0042] Turning now to FIG. 3, a system 300 for analyzing video and data is generally shown according to one or more aspects. In accordance with aspects, the video and data is captured from video recording system 104 of FIG. 1. The analysis can result in predicting features that include surgical phases and structures (e.g., instruments, anatomical structures, and / or the like including combinations and / or multiples thereof) in the video data using machine learning. System 300 can be the computing system 102 of FIG. 1, or a part thereof in one or more examples. System 300 uses data streams in the surgical data to identify procedural states according to some aspects.

[0043] System 300 includes a data reception system 305 that collects surgical data, including the video data and surgical instrumentation data. The data reception system 305 can include one or more devices (e.g., one or more user devices and / or servers) located within and / or associated with a surgical operating room and / or control center. The data reception system 305 can receive surgical data in real-time, i.e., as the surgical procedure is being performed. Alternatively, or in addition, the data reception system 305 can receive or access surgical data in an offline manner, for example, by accessing data that is stored in the data collection system 150 of FIG. 1.

[0044] System 300 further includes a machine learning processing system 310 that processes the surgical data using one or more machine learning models to identify one or more features, such as surgical phase, instrument, anatomical structure, and / or the like including combinations and / or multiples thereof, in the surgical data. It will be appreciated that machine learning processing system 310 can include one or more devices (e.g., one or more servers), each of which can be configured to include part or all of one or more of the depicted components of the machine learning processingsystem 310. In some instances, a part or all of the machine learning processing system 310 is cloud-based and / or remote from an operating room and / or physical location corresponding to a part or all of data reception system 305. It will be appreciated that several components of the machine learning processing system 310 are depicted and described herein. However, the components are just one example structure of the machine learning processing system 310, and that in other examples, the machine learning processing system 310 can be structured using a different combination of the components. Such variations in the combination of the components are encompassed by the technical solutions described herein.

[0045] The machine learning processing system 310 can include a machine learning training system 325, which can be a separate device (e.g., server) that stores its output as one or more trained machine learning models 330. The machine learning models 330 are accessible by a machine learning execution system 340. The machine learning execution system 340 can be separate from the machine learning training system 325 in some examples. In other words, in some aspects, devices that “train” the models are separate from devices that “infer,” i.e., perform real-time processing of surgical data using the trained machine learning models 330.

[0046] Machine learning processing system 310, in some examples, further includes a data generator 315 to generate simulated surgical data, such as a set of synthetic images and / or synthetic video, in combination with real image and video data from the video recording system 104, to generate trained machine learning models 330. Data generator 315 can access (read / write) a data store 320 to record data, including multiple images and / or multiple videos. The images and / or videos can include images and / or videos collected during one or more procedures (e.g., one or more surgical procedures). For example, the images and / or video may have been collected by a user device worn by the actor 112 of FIG. 1 (e.g., surgeon, surgical nurse, anesthesiologist, and / or the like including combinations and / or multiples thereol) during the surgery, a non-wearable imaging device located within an operating room, an endoscopic camera inserted inside the patient 110 of FIG. 1, and / or the like including combinations and / or multiples thereof. The data store 320 is separate from the data collection system 150 of FIG. 1 in some examples. In other examples, the data store 320 is part of the data collection system 150.

[0047] Each of the images and / or videos recorded in the data store 320 for performing training (e.g., generating the machine learning models 330) can be defined as a base image and can be associated with other data that characterizes an associated procedure and / or rendering specifications. For example, the other data can identify a type of procedure, a location of a procedure, one or more people involved in performing the procedure, surgical objectives, and / or an outcome of the procedure. Alternatively, or in addition, the other data can indicate a stage of the procedure with which the image or video corresponds, rendering specification with which the image or video corresponds and / or a type of imaging device that captured the image or video (e.g., and / or, if the device is a wearable device, a role of a particular person wearing the device, and / or the like including combinations and / or multiples thereof). Further, the other data can include image-segmentation data that identifies and / or characterizes one or more objects (e.g., tools, anatomical objects, and / or the like including combinations and / or multiples thereof) that are depicted in the image or video. The characterization can indicate the position, orientation, or pose of the object in the image. For example, the characterization can indicate a set of pixels that correspond to the object and / or a state of the object resulting from a past or current user handling. Localization can be performed using a variety of techniques for identifying objects in one or more coordinate systems.

[0048] The machine learning training system 325 uses the recorded data in the data store 320, which can include the simulated surgical data (e.g., set of synthetic images and / or synthetic video) and / or actual surgical data to generate the trained machine learning models 330. The trained machine learning models 330 can be defined based on a type of model and a set of hyperparameters (e.g., defined based on input from a client device). The trained machine learning models 330 can be configured based on a set of parameters that can be dynamically defined based on (e.g., continuous or repeated) training (i.e., learning, parameter tuning). Machine learning training system 325 can use one or more optimization algorithms to define the set of parameters to minimize or maximize one or more loss functions. The set of (learned) parameters can be stored as part of the trained machine learning models 330 using a specific data structure for a particular trained machine learning model of the trained machinelearning models 330. The data structure can also include one or more non-leamable variables (e.g., hyperparameters and / or model definitions).

[0049] In aspects, the trained machine learning models 330 can include one or more imported models 335 that are pre-trained, such as a pre-trained diffusion model. Models trained by the machine learning training system 325 can interact with the imported models 335 as part of a larger-scale system or for purposes of model performance testing, for example.

[0050] Machine learning execution system 340 can access the data structure(s) of the trained machine learning models 330 and accordingly configure the trained machine learning models 330 for inference (e.g., prediction, classification, and / or the like including combinations and / or multiples thereof). The trained machine learning models 330 can include, for example, a fully convolutional network adaptation, an adversarial network model, an encoder, a decoder, or other types of machine learning models. The type of the trained machine learning models 330 can be indicated in the corresponding data structures. The trained machine learning models 330 can be configured in accordance with one or more hyperparameters and the set of learned parameters.

[0051] The trained machine learning models 330, during execution, receive, as input, surgical data to be processed and subsequently generate one or more inferences according to the training. For example, the video data captured by the video recording system 104 of FIG. 1 can include data streams (e.g., an array of intensity, depth, and / or red-green-blue (RGB) values) for a single image or for each of a set of frames (e.g., including multiple images or an image with sequencing data) representing a temporal window of fixed or variable length in a video. The video data that is captured by the video recording system 104 can be received by the data reception system 305, which can include one or more devices located within an operating room where the surgical procedure is being performed. Alternatively, the data reception system 305 can include devices that are located remotely, to which the captured video data is streamed live during the performance of the surgical procedure. Alternatively, or in addition, the data reception system 305 accesses the data in an offline manner from the data collection system 150 or from any other data source (e.g., local or remote storage device).

[0052] The data reception system 305 can process the video and / or data received. The processing can include decoding when a video stream is received in an encoded format such that data for a sequence of images can be extracted and processed. The data reception system 305 can also process other types of data included in the input surgical data. For example, the surgical data can include additional data streams, such as audio data, RFID data, textual data, measurements from one or more surgical instruments / sensors, and / or the like including combinations and / or multiples thereof, that can represent stimuli / procedural states from the operating room. The data reception system 305 synchronizes the different inputs from the different devices / sensors before inputting them in the machine learning processing system 310.

[0053] The trained machine learning models 330, once trained, can analyze the input surgical data, and in one or more aspects, predict and / or characterize features (e.g., structures) included in the video data included with the surgical data. The video data can include sequential images and / or encoded video data (e.g., using digital video file / stream formats and / or codecs, such as MP4, MOV, AVI, WEBM, AVCHD, OGG, and / or the like including combinations and / or multiples thereof). The prediction and / or characterization of the features can include segmenting the video data or predicting the localization of the structures with a probabilistic heatmap. In some instances, the one or more trained machine learning models 330 include or are associated with a preprocessing or augmentation (e.g., intensity normalization, resizing, cropping, and / or the like including combinations and / or multiples thereof) that is performed prior to segmenting the video data. An output of the one or more trained machine learning models 330 can include image-segmentation or probabilistic heatmap data that indicates which (if any) of a defined set of structures are predicted within the video data, a location and / or position and / or pose of the structure(s) within the video data, and / or state of the structure(s). The location can be a set of coordinates in an image / frame in the video data. For example, the coordinates can provide a bounding box. The coordinates can provide boundaries that surround the structure(s) being predicted. The trained machine learning models 330, in one or more examples, are trained to perform higher-level predictions and tracking, such as predicting a phase of a surgical procedure and tracking one or more surgical instruments used in the surgical procedure.

[0054] While some techniques for predicting a surgical phase (“phase”) in the surgical procedure are described herein, it should be understood that any other technique for phase prediction can be used without affecting the aspects of the technical solutions described herein. In some examples, the machine learning processing system 310 includes a detector 350 that uses the trained machine learning models 330 to identify various items or states within the surgical procedure (“procedure”). The detector 350 can use a particular procedural tracking data structure 355 from a list of procedural tracking data structures. The detector 350 can select the procedural tracking data structure 355 based on the type of surgical procedure that is being performed. In one or more examples, the type of surgical procedure can be predetermined or input by actor 112. For instance, the procedural tracking data structure 355 can identify a set of potential phases that can correspond to a part of the specific type of procedure as “phase predictions”, where the detector 350 is a phase detector.

[0055] In some examples, the procedural tracking data structure 355 can be a graph that includes a set of nodes and a set of edges, with each node corresponding to a potential phase. The edges can provide directional connections between nodes that indicate (via the direction) an expected order during which the phases will be encountered throughout an iteration of the procedure. The procedural tracking data structure 355 may include one or more branching nodes that feed to multiple next nodes and / or can include one or more points of divergence and / or convergence between the nodes. In some instances, a phase indicates a procedural action (e.g., surgical action) that is being performed or has been performed and / or indicates a combination of actions that have been performed. In some instances, a phase relates to a biological state of a patient undergoing a surgical procedure. For example, the biological state can indicate a complication (e.g., blood clots, clogged arteries / veins, and / or the like including combinations and / or multiples thereof), pre-condition (e.g., lesions, polyps, and / or the like including combinations and / or multiples thereof). In some examples, the trained machine learning models 330 are trained to detect an “abnormal condition,” such as hemorrhaging, arrhythmias, blood vessel abnormality, and / or the like including combinations and / or multiples thereof.

[0056] Each node within the procedural tracking data structure 355 can identify one or more characteristics of the phase corresponding to that node. The characteristics caninclude visual characteristics. In some instances, the node identifies one or more tools that are typically in use or available for use (e.g., on a tool tray) during the phase. The node also identifies one or more roles of people who are typically performing a surgical task, a typical type of movement (e.g., of a hand or tool), and / or the like including combinations and / or multiples thereof. Thus, detector 350 can use the segmented data generated by machine learning execution system 340 that indicates the presence and / or characteristics of particular objects within a field of view to identify an estimated node to which the real image data corresponds. Identification of the node (i.e. , phase) can further be based upon previously detected phases for a given procedural iteration and / or other detected input (e.g., verbal audio data that includes person-to-person requests or comments, explicit identifications of a current or past phase, information requests, and / or the like including combinations and / or multiples thereof).

[0057] The detector 350 can output predictions, such as a phase prediction associated with a portion of the video data that is analyzed by the machine learning processing system 310. The phase prediction is associated with the portion of the video data by identifying a start time and an end time of the portion of the video that is analyzed by the machine learning execution system 340. The phase prediction that is output can include segments of the video where each segment corresponds to and includes an identity of a surgical phase as detected by the detector 350 based on the output of the machine learning execution system 340. Further, the phase prediction, in one or more examples, can include additional data dimensions, such as, but not limited to, identities of the structures (e.g., instrument, anatomy, and / or the like including combinations and / or multiples thereof) that are identified by the machine learning execution system 340 in the portion of the video that is analyzed. The phase prediction can also include a confidence score of the prediction. Other examples can include various other types of information in the phase prediction that is output. Further, other types of outputs of the detector 350 can include state information or other information used to generate audio output, visual output, and / or commands. For instance, the output can trigger an alert, an augmented visualization, identify a predicted current condition, identify a predicted future condition, command control of equipment, and / or result in other such data / commands being transmitted to a support system component, e.g., through surgical procedure support system 202 of FIG. 2.

[0058] The above paragraphs discuss phase prediction. It is to be appreciated that aspects of the presently disclosed system and do not require phase prediction, but the aspects could be applied to a range of models and applications.

[0059] It should be noted that although some of the drawings depict endoscopic videos being analyzed, the technical solutions described herein can be applied to analyze video and image data captured by cameras that are not endoscopic (i.e., cameras external to the patient’s body) when performing open surgeries (i.e., not laparoscopic surgeries). For example, the video and image data can be captured by cameras that are mounted on one or more personnel in the operating room (e.g., surgeon). Alternatively, or in addition, the cameras can be mounted on surgical instruments, walls, or other locations in the operating room. Alternatively, or in addition, the video can be images captured by other imaging modalities, such as ultrasound.

[0060] Turning now to FIG. 4, a diffusion feature extraction process 400 is depicted according to one or more aspects. The diffusion feature extraction process 400 can provide a series of frames of surgical video to perform diffusion feature extraction using a pre-trained diffusion model, where extracted features can be used for various computer vision tasks such as key-point tracking and segmentation. A tracking module can use features extracted by diffusion feature extraction to identify one or more keypoints in frames of surgical video. The key -points can be output to a display for visual identification of the key-points and / or tracked in data structures for other uses, for instance, by another machine learning model as part of a larger scale and / or subsequent process. An example output of the tracking module is depicted in a sequence of keypoints 600 tracked in FIG. 8. A segmentation model can use features extracted by diffusion feature extraction to identify one or more segmentation predictions as a predicted mask for frames of surgical video. For instance, the segmentation can group anatomical structures, surgical instruments, and other such regions. An example output of the segmentation module is depicted in a sequence of segmentations 650 tracked in FIG. 9.

[0061] FIG. 5 depicts a diffusion feature extraction pipeline 450 according to one or more aspects. The diffusion feature extraction pipeline 450 receives a surgical video frame and converts the surgical video frame into a latent representation. Noise diffusion is performed to produce a noisy latent that is iteratively denoised over timesteps to yield a denoised latent image after passing through a multi-level decoder. Rather than generating a final image, internal features can be extracted from one or more levels of the decoder. An example result of a rendering of features from the decoder at different levels is illustrated in FIG. 6 as images 500, with frames at different times (e.g., t-5, t, and t+5) and corresponding extracted features from different decoder levels (e.g., levels 1, 2, 3, 4). A null-prompt (i.e. , “”) can be used as input for image generation. A Principal Component Analysis (PCA) can be performed on the extracted features with the top-3 principal components in images 500. Notably, the images 500 illustrate (i) object grouping: instrument and anatomy features in the PCA maps are highly localized and distinct; (ii) feature-granularity: features extracted from different decoder levels depict different feature-granularity. Features shift from coarse (i.e., low- frequency) to fine-grained (i.e., high-frequency) with increasing decoder levels; (iii) temporal-consistency: SD internal features are temporally-consistent (i.e., same anatomy / instrument is represented by same color across temporal direction). This resolves two primary bottlenecks of surgical video segmentation - i.e., temporalconsistency and object-localization without GT masks.

[0062] Given a first frame-mask pair {ii, mi} from a surgical video containing N frames, the process as described herein can segment the rest of the N-l frames to generate corresponding masks {m2, m3, r , • • • ,m\ }. Leveraging the information-rich interim representation of text-to-image SD model for training free semantic segmentation, interplay between subsequent video frames can be used with an affinity matrix, and features from n previous frames can be incorporated to ensure temporalconsistency. Efficient Utilization of the SD internal features can depend on a few application-specific hyperparameters. The extracted internal feature maps of SD vary significantly for different diffusion timesteps and architecture levels.

[0063] Given an image-prompt pair {x,c}, a latent representation ofA ccan be generated as. Given the timestep t, Gaussian noise can be added on zo using a noise scheduler to transform it to the tthstep noisy latent zt . Given zt, t, and prompt embedding T (c), these can be passed to a denoising Unet U( ) and the internal features extracted from the UNet decodersUn. For example, in SD,iraAxwx3 I p ; r • ' - ! 2K:i an input frame of x of size * would produce featuresc"fromand ttf • . . . . . . . . . respectively. These features can be used as backbone features for performing video semantic segmentation. For datasets that do not contain paired text prompts, null-prompts (i.e., “ ”) can be used in place of c. Diffusion feature extraction procedure is illustrated in FIG. 5.

[0064] FIG. 7 depicts a mask prediction algorithm 550 according to one or more aspects. Utilizing the extracted SD features for video semantic segmentation can involve two aspects — (i) selecting the most optimal SD feature representative enough for segmenting seemingly intricate instruments and anatomies present in surgical videos, and (ii) maintaining temporal-consistency between the segmented frames while doing so. In addressing the former, the best set of parameters can be determined, for example, thorough an ablative study. In addressing the latter, cross-frame interactions can be considered via an affinity matrix while involving multi-frame features during mask generation to maintain temporal-continuity. Mask generation from SD features can be similar attention-based approaches, for example, using key, query, and value. Given the first frame-mask pair {ii, mi}, analogically it can be assumed that the first frame ground truth mask mi and the next frame mask (i.e., m2 in this case) can be comparable to Key (K) and Query (Q) of attention respectively. Whereas, the affinity between SD features of frame 1 and 2 can be considered as Value (V). Consequently, self-similarity between the first frame ground truth mask (K) and affinity matrix (V) can yield the segmentation mask of the y2 immediate next frame (Q). Extracting an SD UNet feature (fromd) of all N frames of a video. Next, a spatial neighborhood mask (with a window n) can be created via a SpatialMask(-) function to restrict the feature-affinity matrix within a local spatial region. In practice, for every valid coordinate within the window n, SpatialMask(-) can make a corresponding mask entry in N to be 1. This can be similar to local attention. Starting from i2, for each of the N frames, the affinity matrix A can be calculated between the current frame feature and its immediate predecessoras:- i 1 -' y Where, the temperature hyperparameter T is set to be 0.2. Next, the neighborhood-restricted feature-affinity matrix AN can be generated as: AN=(A N). At the i* frame mask (starting from i=2 onwards) can be reached by multiplying AN with the (i- l)thframe mask2. Eventually, the final 1thsegmentation mask can be generated by applying argmax on mi. This is illustrated by the mask prediction algorithm 550 of FIG. 7.

[0065] Instead of calculating the 1thmask mi from the affinity with its immediate predecessor (i.e., mi-i), the predicted masks at each step can be accumulated in a queue and to utilize the previously predicted masks to compute the affinity. Consequently, the value of mi is influenced by all preceding masks. This mechanism ensures that a limited history of segmentation masks from past frames is considered, thereby maintaining a level of temporal-consistency.

[0066] According to another aspect of the disclosure, a computer program product includes a memory device having computer executable instructions stored thereon, which when executed by one or more processors cause the one or more processors to perform a plurality of operations including.

[0067] Different levels of a decoder can produce coarser or finer granularity. For example, a four-level decoder can be tested at each level to strike a balance between accurate object localization and fine-grained features. Testing various time step combinations can be used to determine an optimal task-specific adaption of pre-trained SD models. A balance point for a specific task type can be determined to avoid distortion. Varying numbers of past predictions can be tested to preserve temporal consistency while avoiding temporal distortion. Neighborhood mask window size can be varied to determine a window size that pairs well with an affinity matrix. Where multiple versions of a pre-trained SD model exist, it can be useful to test more than one version to select a best performing version for a targeted task type.

[0068] By harnessing the latent potential of pre-trained text-to-image diffusion models and integrating cross-frame interactions, methods disclosed herein can achieve improved results surpassing similar unsupervised baselines. This can offer a cost- effective solution to the challenges of pixel-level annotation as well as enhancing surgical guidance and postoperative analysis. Diffusion features may be used for othersurgical video analysis tasks such as phase-recognition, key-point-tracking, depthestimation, and other such tasks.

[0069] Turning now to FIG. 10, a flowchart of a method 700 for using a pre-trained diffusion model for downstream medical vision tasks is generally shown in accordance with one or more aspects. All or a portion of method 700 can be a computer-implemented method that is implemented, for example, by all or a portion of CAS system 100 of FIG. 1 and / or computer system 800 of FIG. 11.

[0070] At block 702, video data including a sequence of frames of a surgical video is received.

[0071] At block 704, the sequence of frames of the surgical video is passed to a pretrained diffusion model including a decoder.

[0072] At block 706, an intermediate level of the decoder is accessed to extract features.

[0073] At block 708, the features are provided to one or more medical vision tasks to characterize content of the surgical video.

[0074] According to aspects, the one or more medical vision tasks can include segmentation.

[0075] According to aspects, the one or more medical vision tasks can include keypoint tracking.

[0076] According to aspects, the one or more medical vision tasks can include phase recognition.

[0077] According to aspects, the one or more medical vision tasks can include depth estimation.

[0078] According to aspects, the method can include maintaining temporal continuity for multi-frame features using an affinity matrix.

[0079] According to aspects, the method can include where the pre-trained diffusion model includes noise diffusion that produces a noisy latent which is iteratively denoised over time steps through multiple levels of the decoder.

[0080] According to aspects, the method can include where the pre-trained diffusion model is trained with non-medical data.

[0081] According to aspects, the method can include where the pre-trained diffusion model is trained with medical data.

[0082] According to aspects, the method can include where the pre-trained diffusion model is trained with a combination of non-medical data and medical data.

[0083] According to aspects, the method can include where an output of the one or more medical vision tasks provides one or more estimated annotations that are adjustable through a user interface.

[0084] According to aspects, a system can be configured to perform any of the aspects as described herein.

[0085] According to aspects, a computer program product can include a memory device having computer executable instructions stored thereon, which when executed by one or more processors cause the one or more processors to perform a plurality of operations comprising any of the aspects as described herein.

[0086] Turning now to FIG. 11, a computer system 800 is generally shown in accordance with an aspect. The computer system 800 can be an electronic computer framework comprising and / or employing any number and combination of computing devices and networks utilizing various communication technologies, as described herein. The computer system 800 can be easily scalable, extensible, and modular, with the ability to change to different services or reconfigure some features independently of others. The computer system 800 may be, for example, a server, desktop computer, laptop computer, tablet computer, or smartphone. In some examples, computer system 800 may be a cloud computing node. Computer system 800 may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. Computer system 800 may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media, including memory storage devices.

[0087] As shown in FIG. 11, the computer system 800 has one or more central processing units (CPU(s)) 801a, 801b, 801c, etc. (collectively or generically referred to as processor(s) 801). The processors 801 can be a single-core processor, multi-core processor, computing cluster, or any number of other configurations. The processors801 can be any type of circuitry capable of executing instructions. The processors 801, also referred to as processing circuits, are coupled via a system bus 802 to a system memory 803 and various other components. The system memory 803 can include one or more memory devices, such as read-only memory (ROM) 804 and a random-access memory (RAM) 805. The ROM 804 is coupled to the system bus 802 and may include a basic input / output system (BIOS), which controls certain basic functions of the computer system 800. The RAM is read-write memory coupled to the system bus 802 for use by the processors 801. The system memory 803 provides temporary memory space for operations of said instructions during operation. The system memory 803 can include random access memory (RAM), read-only memory, flash memory, or any other suitable memory systems.

[0088] The computer system 800 comprises an input / output (I / O) adapter 806 and a communications adapter 807 coupled to the system bus 802. The I / O adapter 806 may be a small computer system interface (SCSI) adapter that communicates with a hard disk 808 and / or any other similar component. The I / O adapter 806 and the hard disk 808 are collectively referred to herein as a mass storage 810.

[0089] Software 811 for execution on the computer system 800 may be stored in the mass storage 810. The mass storage 810 is an example of a tangible storage medium readable by the processors 801, where the software 811 is stored as instructions for execution by the processors 801 to cause the computer system 800 to operate, such as is described hereinbelow with respect to the various Figures. Examples of computer program product and the execution of such instruction is discussed herein in more detail. The communications adapter 807 interconnects the system bus 802 with a network 812, which may be an outside network, enabling the computer system 800 to communicate with other such systems. In one aspect, a portion of the system memory 803 and the mass storage 810 collectively store an operating system, which may be any appropriate operating system to coordinate the functions of the various components shown in FIG. 11.

[0090] Additional input / output devices are shown as connected to the system bus 802 via a display adapter 815 and an interface adapter 816. In one aspect, the adapters 806, 807, 815, and 816 may be connected to one or more I / O buses that are connected to the system bus 802 via an intermediate bus bridge (not shown). A display 819 (e.g., ascreen or a display monitor) is connected to the system bus 802 by a display adapter 815, which may include a graphics controller to improve the performance of graphicsintensive applications and a video controller. A keyboard, a mouse, a touchscreen, one or more buttons, a speaker, etc., can be interconnected to the system bus 802 via the interface adapter 816, which may include, for example, a Super I / O chip integrating multiple device adapters into a single integrated circuit. Suitable I / O buses for connecting peripheral devices such as hard disk controllers, network adapters, and graphics adapters typically include common protocols, such as the Peripheral Component Interconnect (PCI). Thus, as configured in FIG. 11, the computer system 800 includes processing capability in the form of the processors 801, and storage capability including the system memory 803 and the mass storage 810, input means such as the buttons, touchscreen, and output capability including the speaker 823 and the display 819.

[0091] In some aspects, the communications adapter 807 can transmit data using any suitable interface or protocol, such as the internet small computer system interface, among others. The network 812 may be a cellular network, a radio network, a wide area network (WAN), a local area network (LAN), or the Internet, among others. An external computing device may connect to the computer system 800 through the network 812. In some examples, an external computing device may be an external web server or a cloud computing node.

[0092] It is to be understood that the block diagram of FIG. 11 is not intended to indicate that the computer system 800 is to include all of the components shown in FIG. 11. Rather, the computer system 800 can include any appropriate fewer or additional components not illustrated in FIG. 11 (e.g., additional memory components, embedded controllers, modules, additional network interfaces, etc.). Further, the aspects described herein with respect to computer system 800 may be implemented with any appropriate logic, wherein the logic, as referred to herein, can include any suitable hardware (e.g., a processor, an embedded controller, or an application-specific integrated circuit, among others), software (e.g., an application, among others), firmware, or any suitable combination of hardware, software, and firmware, in various aspects. Various aspects can be combined to include two or more of the aspects described herein.

[0093] Aspects disclosed herein may be a system, a method, and / or a computer program product at any possible technical detail level of integration. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to carry out various aspects.

[0094] The computer-readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer- readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer-readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device, such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer-readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0095] Computer-readable program instructions described herein can be downloaded to respective computing / processing devices from a computer-readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network, and / or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium within the respective computing / processing device.

[0096] Computer-readable program instructions for carrying out operations of the present disclosure may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source-code or object code written in any combination of one or more programming languages, including an object-oriented programming language, such as Smalltalk, C++, high-level languages such as Python, or the like, and procedural programming languages, such as the “C” programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user’s computer, partly on the user’s computer, as a stand-alone software package, partly on the user’s computer and partly on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user’s computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some aspects, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PL A) may execute the computer-readable program instruction by utilizing state information of the computer-readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0097] Aspects are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to aspects of the disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer- readable program instructions.

[0098] These computer-readable program instructions may be provided to a processor of a computer system, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readablestorage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a particular manner, such that the computer-readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function / act specified in the flowchart and / or block diagram block or blocks.

[0099] The computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer-implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0100] The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various aspects. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the Figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustration, and combinations of blocks in the block diagrams and / or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0101] The descriptions of the various aspects have been presented for purposes of illustration but are not intended to be exhaustive or limited to the aspects disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described aspects. The terminology used herein was chosen to best explain the principles of the aspects, the practicalapplication or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the aspects described herein.

[0102] Various aspects are described herein with reference to the related drawings. Alternative aspects can be devised without departing from the scope of this disclosure. Various connections and positional relationships (e.g., over, below, adjacent, etc.) are set forth between elements in the following description and in the drawings. These connections and / or positional relationships, unless specified otherwise, can be direct or indirect, and the present disclosure is not intended to be limiting in this respect. Accordingly, a coupling of entities can refer to either a direct or an indirect coupling, and a positional relationship between entities can be a direct or indirect positional relationship. Moreover, the various tasks and process steps described herein can be incorporated into a more comprehensive procedure or process having additional steps or functionality not described in detail herein.

[0103] The following definitions and abbreviations are to be used for the interpretation of the claims and the specification. As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having,” “contains,” or “containing,” or any other variation thereof are intended to cover a non-exclusive inclusion. For example, a composition, a mixture, process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but can include other elements not expressly listed or inherent to such composition, mixture, process, method, article, or apparatus.

[0104] Additionally, the term “exemplary” is used herein to mean “serving as an example, instance or illustration.” Any aspect or design described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects or designs. The terms “at least one” and “one or more” may be understood to include any integer number greater than or equal to one, i.e., one, two, three, four, etc. The terms “a plurality” may be understood to include any integer number greater than or equal to two, i.e., two, three, four, five, etc. The term “connection” may include both an indirect “connection” and a direct “connection.”

[0105] The terms “about,” “substantially,” “approximately,” and variations thereof are intended to include the degree of error associated with measurement of the particularquantity based upon the equipment available at the time of filing the application. For example, “about” can include a range of ± 8% or 5%, or 2% of a given value.

[0106] For the sake of brevity, conventional techniques related to making and using aspects may or may not be described in detail herein. In particular, various aspects of computing systems and specific computer programs to implement the various technical features described herein are well known. Accordingly, in the interest of brevity, many conventional implementation details are only mentioned briefly herein or are omitted entirely without providing the well-known system and / or process details.

[0107] It should be understood that various aspects disclosed herein may be combined in different combinations than the combinations specifically presented in the description and accompanying drawings. It should also be understood that, depending on the example, certain acts or events of any of the processes or methods described herein may be performed in a different sequence, may be added, merged, or left out altogether (e.g., all described acts or events may not be necessary to carry out the techniques). In addition, while certain aspects of this disclosure are described as being performed by a single module or unit for purposes of clarity, it should be understood that the techniques of this disclosure may be performed by a combination of units or modules associated with, for example, a medical device.

[0108] In one or more examples, the described techniques may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored as one or more instructions or code on a computer-readable medium and executed by a hardware-based processing unit. Computer-readable media may include non-transitory computer-readable media, which corresponds to a tangible medium, such as data storage media (e.g., RAM, ROM, EEPROM, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer).

[0109] Instructions may be executed by one or more processors, such as one or more digital signal processors (DSPs), graphics processing units (GPUs), microprocessors, application-specific integrated circuits (ASICs), field programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuitry. Accordingly, the term “processor” as used herein may refer to any of the foregoing structure or any otherphysical structure suitable for implementation of the described techniques. Also, the techniques could be fully implemented in one or more circuits or logic elements.

[0110] The following examples are illustrative of the techniques described herein.

[0111] Example 1. A computer-implemented method comprising: receiving, by a system that includes at least one processor, video data comprising a sequence of frames of a surgical video; passing the sequence of frames of the surgical video to a pre-trained diffusion model comprising a decoder; accessing an intermediate level of the decoder to extract features; and providing the features to one or more medical vision tasks to characterize content of the surgical video.

[0112] Example 2. The computer-implemented method of Example 1, wherein the one or more medical vision tasks comprise segmentation.

[0113] Example 3. The computer-implemented method of Example 1, wherein the one or more medical vision tasks comprise key-point tracking.

[0114] Example 4. The computer-implemented method of Example 1, wherein the one or more medical vision tasks comprise phase recognition.

[0115] Example 5. The computer-implemented method of Example 1, wherein the one or more medical vision tasks comprise depth estimation.

[0116] Example 6. The computer-implemented method of Example 1, further comprising: maintaining temporal continuity for multi-frame features using an affinity matrix.

[0117] Example 7. The computer-implemented method of Example 1, wherein the pretrained diffusion model comprises noise diffusion that produces a noisy latent which is iteratively denoised over time steps through multiple levels of the decoder.

[0118] Example 8. The computer-implemented method of Example 1 , wherein the pretrained diffusion model is trained with non-medical data.

[0119] Example 9. The computer-implemented method of Example 1, wherein the pretrained diffusion model is trained with medical data.

[0120] Example 10. The computer-implemented method of Example 1, wherein the pre-trained diffusion model is trained with a combination of non-medical data and medical data.

[0121] Example 11. The computer-implemented method of Example 1, wherein an output of the one or more medical vision tasks provides one or more estimated annotations that are adjustable through a user interface.

[0122] Example 12. A system configured to perform any of the aspects as described herein.

[0123] Example 13. A computer program product comprising a memory device having computer executable instructions stored thereon, which when executed by one or more processors cause the one or more processors to perform a plurality of operations comprising any of the aspects as described herein.

Claims

CLAIMSWhat is claimed is:

1. A computer-implemented method comprising: receiving, by a system that includes at least one processor, video data comprising a sequence of frames of a surgical video; passing the sequence of frames of the surgical video to a pre-trained diffusion model comprising a decoder; accessing an intermediate level of the decoder to extract features; and providing the features to one or more medical vision tasks to characterize content of the surgical video.

2. The computer-implemented method of claim 1, wherein the one or more medical vision tasks comprise segmentation.

3. The computer-implemented method of claim 1 or claim 2, wherein the one or more medical vision tasks comprise key-point tracking.

4. The computer-implemented method of any preceding claim, wherein the one or more medical vision tasks comprise phase recognition.

5. The computer-implemented method of any preceding claim, wherein the one or more medical vision tasks comprise depth estimation.

6. The computer-implemented method of any preceding claim, further comprising: maintaining temporal continuity for multi-frame features using an affinity matrix.

7. The computer-implemented method of any preceding claim, wherein the pre-trained diffusion model comprises noise diffusion that produces a noisy latent which is iteratively denoised over time steps through multiple levels of the decoder.

8. The computer-implemented method of any preceding claim, wherein the pre-trained diffusion model is trained with non-medical data.

9. The computer-implemented method of any preceding claim, wherein the pre-trained diffusion model is trained with medical data.

10. The computer-implemented method of any preceding claim, wherein the pre-trained diffusion model is trained with a combination of non-medical data and medical data.

11. The computer-implemented method of any preceding claim, wherein an output of the one or more medical vision tasks provides one or more estimated annotations that are adjustable through a user interface.

12. A system configured to perform any of the methods of any preceding claim.

13. A computer program product comprising a memory device having computer executable instructions stored thereon, which when executed by one or more processors cause the one or more processors to perform a plurality of operations comprising any one or more of the methods of any preceding claim.

Images