Device and method for assisted anatomical localization

Abstract

Device and method of assisted anatomical localization The invention relates to a method and a device (2) of anatomical localization.The method comprises: obtaining an image (IM1) of a subject (UR2); detecting at least two first anatomical points (PT1) associated with the subject's skull; determining a first value (V1) representing a reference distance (D1) separating the first two anatomical points; detecting two second anatomical points (PT2) associated with the subject's shoulders; locating a sternal midline equidistant from the second anatomical points (PT2), the sternal midline intersecting at a point of intersection a segment defined by the second anatomical points (PT2); and determining coordinates (C1) of at least one target anatomical point (PT3) by calculating second distances (D2x, D2y) of said at least one target anatomical point with respect to the point of intersection, the second distances being determined from the first value (V1). Figure for the abstract: Fig. 1.

Description

Title of the invention: Device and method for assisted anatomical localization technical field

[0001] The invention relates to the field of anatomical point localization and more particularly to methods and devices enabling assisted localization of a subject's anatomical points, for purposes such as auscultation or applications in the fields of medicine, telemedicine, or wellness. Prior art

[0002] Traditionally, auscultation requires a thorough knowledge of human physiology in order to correctly identify the areas to be auscultated. Such examinations require specific technical skills, generally reserved for physicians.

[0003] Current solutions do allow for the localization of auscultation points on the human body, but only approximately, with the quality of results varying depending on the subjects examined and the conditions of use. Practitioners are not always able to precisely identify the position of the auscultation points, which can compromise the accuracy and reliability of the auscultation.

[0004] Some current systems, particularly digital stethoscopes, offer the ability to capture bodily sounds but often lack precision, especially in non-clinical environments, for example, in noisy intervention settings that are not conducive to listening and concentration. The use of this type of equipment presents a particular challenge for non-specialized users who are generally not qualified to locate the anatomical area(s) to be auscultated.

[0005] One of the main problems with current solutions is therefore the need for specialized training to perform a correct auscultation, which limits the use of these devices by non-specialized emergency responders. Without adequate training, it is difficult for these responders to precisely locate the auscultation points, which can lead to examination or diagnostic errors, or to ineffective listening to bodily sounds.

[0006] Furthermore, obtaining an accurate diagnosis under less than optimal or noisy conditions is difficult without appropriate equipment. Conventional devices are not always designed to operate effectively in noisy environments, which can further complicate the task of practitioners and emergency responders.

[0007] Moreover, current solutions generally lack ergonomics, making their use impractical, especially in emergency situations or noisy environments.

[0008] The lack of adequate technological support to guide auscultation in patients, particularly in emergency settings, therefore constitutes a major obstacle to patient care. In these situations, healthcare professionals often have to perform rapid and accurate examinations without the ideal conditions of a clinical setting. The inability to quickly and accurately locate auscultation sites can have adverse consequences, potentially leading to inaccurate observations and therefore incorrect diagnoses or medical findings, which can be detrimental to the patient being examined, for example, in cases where the patient requires medical attention. These difficulties can, in particular, lead to a decline in the quality of care provided to the patient in question.

[0009] In short, current techniques do not meet the needs of healthcare and wellness professionals, and more generally, of any practitioner tasked with examining or interacting with specific anatomical points on a human or animal body. These techniques suffer in particular from shortcomings in terms of precision, reliability, and ease of use, especially in emergency situations where speed and accuracy are crucial and / or when practitioners are not specialists in this type of procedure. Description of the invention

[0010] One of the objects of the present invention is to solve at least one of the problems or deficiencies of the technological background described above.

[0011] Another object of the present invention is to enable precise, reliable and rapid localization of anatomical points of a subject, for example of an individual or an animal.

[0012] Another object of the present invention is to assist, or guide, in an efficient and ergonomic manner a practitioner in the localization of anatomical points of a subject.

[0013] To this end, a first aspect of the present invention relates to a method, also called an assisted anatomical localization method, implemented by a processing device, said method comprising: a) obtaining an image of a subject; b) detection, in said image, of at least two initial anatomical points associated with the subject's skull; c) determination of a first representative value of a reference distance separating said at least two first anatomical points; d) detection, in said image, of two second anatomical points associated with the subject's shoulders; e) location of a sternal midline equidistant from the second anatomical points, the sternal midline intersecting at a point of intersection a segment defined by the second anatomical points; and f) determination of coordinates of at least one target anatomical point by calculating second distances of said at least one target anatomical point with respect to the point of intersection, at least one of the second distances being determined from the first value.

[0014] The method according to the invention may include other features which may be taken separately or in combination, in particular among the following embodiments which are presented by way of illustration only and may be combined or associated unless otherwise stipulated.

[0015] According to a particular embodiment, obtaining the image a) comprises at least one of the following: - image acquisition using a camera embedded in the processing device; - receiving the image from a remote server; and - image extraction from memory embedded in the processing device.

[0016] According to a particular embodiment, the first anatomical points detected in b) include at least one of the following: - anatomical points associated with the subject's eyes, visible in the image; and - anatomical points associated with the subject's ears visible in the image.

[0017] According to a particular embodiment, the method comprises: - searching for candidate anatomical points in the image; and - selection, from among the candidate anatomical points detected in the image, of the first anatomical points used in c).

[0018] According to a particular embodiment, during determination c), the first value is defined as a function of a number of pixels, in the image obtained in a), separating said at least two first anatomical points.

[0019] According to a particular embodiment, during detection d), the second anatomical points are associated with the subject's shoulder points.

[0020] According to a particular embodiment, the sternal midline forms a perpendicular bisector of the segment defined by the second anatomical points.

[0021] According to a particular embodiment, during the determination f), one of the second distances is determined by applying a proportionality coefficient to the first value.

[0022] According to a particular embodiment, determination f) comprises: - determination of a second distance D2x along a first axis of an anatomical reference frame associated with the subject visible in the image such that D2x = Kx; and - determination of a second distance D2y along a second axis of the anatomical reference frame such that D2y = Ky * VI, where Kx is a constant and Ky is said proportionality coefficient.

[0023] According to a particular embodiment, the method comprises: - display of an image, called second image, according to an image reference, said second image being determined from the image (IM1) obtained in a); in which determination f) comprises: - determination of initial coordinates (XI, Y1) of said at least one target anatomical point in the anatomical reference frame, said initial coordinates (XI, Y1) being defined as a function of the pair (D2x, D2y); and - conversion of the first coordinates (XI, Yl), as second coordinates (X2, Y2), in the image reference frame (RF2) associated with said second image.

[0024] According to a particular embodiment, determination f) comprises: - conversion of the second coordinates (X2, Y2) as third coordinates (X3, Y3) in the display reference associated with a graphical representation of said image.

[0025] According to a particular embodiment, the process comprises: e) providing instructions to guide a user to said at least one target anatomical point.

[0026] According to a particular embodiment, the instruction provided in e) includes at least one of the following: - a visual instruction; - an audible instruction; and - a vibration control.

[0027] According to a particular embodiment, the supply e) comprises: - display of the image and at least one visual indicator, superimposed on the displayed image, said at least one visual indicator being representative of a target position associated with said at least one target anatomical point.

[0028] According to a particular embodiment, the first value represents a number of pixels separating said at least two first anatomical points, in which the supply e) comprises: - estimation of a dimension ratio between the number of pixels defined by the first value and a reference value; and - adaptation of at least one dimension of the visual indicator according to said dimension ratio.

[0029] In a particular embodiment, the different steps of the process according to the first aspect are determined by computer program instructions.

[0030] Accordingly, the invention according to a second aspect relates to a computer program on an information medium (or recording medium), this program being capable of being implemented in a device, called a processing device, or more generally in a computer, this program comprising instructions adapted to the implementation of the steps of a process according to the first aspect of the invention.

[0031] This program may use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0032] The invention according to a third aspect relates to an information medium (or recording medium) readable by a computer, and comprising instructions for a computer program according to the second aspect of the invention.

[0033] The information medium can be any entity or device capable of storing the program. For example, the medium can include a storage means, such as rewritable non-volatile memory or ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a hard disk drive.

[0034] On the other hand, the information medium can be a transmissible medium such as an electrical or optical signal, which can be transmitted via an electrical or optical cable, by radio, or by other means. The program according to the invention can, in particular, be downloaded onto an Internet-type network.

[0035] Alternatively, the information carrier may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the process in question.

[0036] The invention according to a fourth aspect relates to a device, also called a processing device or localization device, configured to implement the method of the first aspect of the invention. In particular, the fourth aspect of the invention relates to a processing device comprising: - a data acquisition module configured to obtain an image of a subject; - a first detection module configured to detect, in said image, at least two initial anatomical points associated with the subject's skull; - a first determination module configured to determine a first representative value of a reference distance (Dl) separating said at least two first anatomical points; - a second detection module configured to detect, in said image, two second anatomical points associated with the subject's shoulders; - a localization module configured to locate a sternal midline equidistant from the second anatomical points, the sternal midline intersecting at an intersection point a segment defined by the second anatomical points; and - a second determination module configured to determine coordinates of at least one target anatomical point by calculating second distances of said at least one target anatomical point with respect to the intersection point, the second distances being determined from the first value.

[0037] It should be noted that the various embodiments mentioned above (as well as those described below) in relation to the process according to the first aspect of the invention and the associated advantages apply in a similar way to the device according to the fourth aspect of the invention.

[0038] For each step of the process according to the first aspect, the device according to the fourth aspect may include a corresponding module configured to carry out said step.

[0039] According to one embodiment, the invention is implemented by means of software and / or hardware components. In this context, the term "module" may refer in this document to a software component, a hardware component, or a set of hardware and software components.

[0040] A software component corresponds to one or more computer programs, one or more subroutines of a program, or more generally to any element of a program or software capable of implementing a function or set of functions, as described below for the module concerned. Such a software component can be executed by a data processor of a physical entity (terminal, server, etc.) and is capable of accessing the hardware resources of that physical entity (memory, storage media, communication bus, input / output electronic cards, user interfaces, etc.).

[0041] Similarly, a hardware component corresponds to any element of a hardware assembly capable of implementing a function or a set of functions, as described below for the module concerned. It may be a programmable hardware component or one with an integrated processor for executing software, for example an integrated circuit, a smart card, a memory card, an electronic card for executing firmware, etc. Brief description of the drawings

[0042] Other features and advantages of the present invention will become apparent from the description of the particular and non-limiting embodiments of the present invention below, with reference to the attached Figures 1 to 5, in which:

[0043] [Fig. 1] The [Fig. 1] schematically represents a processing device and more generally a localization system, according to at least one particular embodiment of the invention;

[0044] [Fig.2] Fig.2 schematically represents the processing device of Fig.1 according to at least one particular embodiment of the invention;

[0045] [Fig.3] The [Fig.3] schematically represents, in the form of a diagram, the steps of an assisted anatomical localization process according to particular embodiments of the invention;

[0046] [Fig.4] The [Fig.4] schematically represents the implementation of steps of the assisted anatomical localization process of the [Fig.3], according to at least one particular embodiment of the invention;

[0047] [Fig.5] The [Fig.5] schematically represents the implementation of steps of the assisted anatomical localization process of the [Fig.3], according to at least one particular embodiment of the invention;

[0048] [Fig. 6] [Fig. 6] schematically represents the implementation of steps in the anatomically assisted localization process of [Fig. 3], according to at least one particular embodiment of the invention; and

[0049] [Fig.7] The [Fig.7] schematically represents the implementation of steps of the assisted anatomical localization process of the [Fig.3], according to at least one particular embodiment of the invention. Description of the implementation methods

[0050] Examples of implementations of the invention will now be described in the following with joint reference to Figures 1-8. Unless otherwise indicated, common or similar elements in several figures bear the same reference symbols and have identical or similar characteristics, so that these common elements are generally not described again for the sake of simplicity.

[0051] The terms "first(s)" (or first(s)), "second(s)", etc.) are used in this document by arbitrary convention to allow identification and distinction of different elements (such as operations, threshold values, etc.) implemented in the embodiments described below.

[0052] In the following description, the terms "upper / lower", "right / left", "top / bottom", "front / back" and, in general, terms designating a spatial reference, refer to the position in which the elements appear on the attached figures in accordance with their context of use.

[0053] The present invention is based on processing an image of a subject to detect at least one target anatomical point from elements of the human body serving as a reference for said detection. This processing particularly involves the detection of at least two initial anatomical points associated with the subject's skull (or head) and at least two secondary anatomical points associated with the subject's shoulders. From physiological data deduced from these initial and secondary anatomical points, one can determine with precision, reliability, and speed the coordinates of at least one target anatomical point that one wishes to identify in or on a subject's body.

[0054] The invention, according to its various embodiments, thus implements a method of assisted anatomical localization, said method comprising: a) obtaining an image of a subject; b) detection, in said image, of at least two initial anatomical points associated with the subject's skull; c) determination of a first representative value of a reference distance (Dl) separating said at least two first anatomical points; d) detection, in said image, of two second anatomical points associated with the subject's shoulders; e) location of a sternal midline equidistant from the second anatomical points, the sternal midline intersecting at a point of intersection a segment defined by the second anatomical points; and f) determination of coordinates of at least one target anatomical point by calculating second distances of said at least one target anatomical point with respect to the point of intersection, the second distances being determined from the first value.

[0055] The process can be implemented by a device, hereafter referred to as a processing device (or localization device), this device comprising means for implementing the steps of the process of the invention.

[0056] Other aspects and advantages of the present invention will become apparent from the embodiments described below with reference to the drawings mentioned above.

[0057] In this document, examples of implementation of the invention are described in the context of locating anatomical points of an individual, that is to say, applied to a human body (this individual being, where applicable, living or dead). Alternatively, the invention can be applied analogously to an animal, in particular to a vertebrate animal (for example, a mammal).

[0058] In this document, an anatomical point refers to a specific and well-defined location on a human (or possibly animal) body, identifiable by physical or physiological landmarks. These points can be used as references for various medical or wellness procedures, including auscultation, for observations, examinations, manipulations, treatments, diagnoses, various interventions (surgical or otherwise), etc. They can serve as guides for healthcare or wellness professionals to locate underlying structures, such as organs, bones, and vessels. blood vessels allow for targeted and effective assessment and intervention. Precise localization of anatomical points ensures appropriate medical examinations and treatments, particularly in contexts where speed and accuracy are crucial.

[0059] Unless otherwise indicated, common or similar elements in several figures bear the same reference signs and have identical or similar characteristics, so that these common elements are generally not described again for the sake of simplicity.

[0060] Figure 1 schematically represents a processing device (also called a device) 2 intended for use by a practitioner (or user) UR1 to implement an assisted localization procedure (also called an anatomical localization procedure or localization procedure) in order to locate at least one target anatomical point PT3 of a subject, namely an individual UR2 in this example. For example, this individual UR2 could be a person requiring first aid, for example in an emergency situation, although other contexts of use are possible. The subject UR2 undergoing assisted anatomical localization could be in any position, for example in a supine position (prone, supine, or other).

[0061] To implement the assisted anatomical localization process, also referred to hereafter as the localization process, the processing device 2 includes in the example considered a processor 4, a user interface 6 and a non-volatile memory 8.

[0062] As illustrated in [Fig. 1], the device 2 is configured to obtain an image IM1 from an image acquisition device 10. It is assumed hereafter that this image acquisition device 10 is (or includes) a camera, although any other suitable device for capturing images or videos may be used. The device 2 and the camera 10 together form a system, also called a localization system, denoted SY1.

[0063] The camera 10 may be part of the device 2 or be external to it. For the sake of example, the processing device 2 is assumed to be a terminal, for example a mobile terminal such as a smartphone, tablet or other, comprising a camera 10 capable of acquiring one or more images IM1 of a subject UR2, for example in the form of at least one image, a sequence of images or video.

[0064] Device 2 shown in [Fig. 1] is only one example of an embodiment; other implementations are possible within the scope of the invention. Those skilled in the art will understand, in particular, that certain elements of Device 2 are described here only to facilitate understanding of the invention, as these elements are not required to implement the invention.

[0065] The memory 8 may comprise one or more memories, in particular a rewritable non-volatile memory and / or a read-only memory (ROM). The memory 8 may optionally also comprise volatile memory (RAM). This memory 8 constitutes a recording medium (or information storage medium) according to a particular embodiment, readable by the device 2, and on which a computer program PG1 according to a particular embodiment is stored. This computer program PG1 contains instructions for executing the steps of an assisted anatomical localization process according to a particular embodiment. The steps of this process are represented, in a particular embodiment of the invention, in [Fig. 3] described later.

[0066] The processor 4 is configured to execute the instructions of the computer program PG1 in order to perform steps of the assisted anatomical localization process. To this end, the processor 4 may include integrated memory, an input / output interface, and various circuits known to those skilled in the art. In particular, the processor 4 may use the memory 8 to perform the various operations and functions necessary for the operation of the device 2, including executing the computer program PG1 during the implementation of the anatomical localization process of the invention. At least a portion of the memory 8 may be part of the processor 4.

[0067] The memory 8 is capable of storing various data useful for implementing the localization method of the invention, in particular: an image IM1 in the form of image data, anatomical point data defining first and second anatomical points PT1 and PT2 respectively, a value VI representing a reference distance Dl, as well as the coordinates denoted Cl, C2 and C3 of at least one target anatomical point PT3 located by the localization method. The nature of this data and how it is used will become clearer in the description of the following embodiments with reference to the figures.

[0068] The user interface 6 may include any appropriate means enabling the UR1 practitioner to interact with his device 2.

[0069] According to one example, the output interface of the user interface 6 may include means of supplying information suitable for transmitting (or presenting) information to the practitioner UR1, in particular at least one instruction CM1 to guide the practitioner and help him locate at least one target anatomical point PT3.

[0070] The user interface 6 may further include an input interface allowing the practitioner UR1 to provide user instructions to the device 2, in particular to control the execution of the localization process according to the computer program PG1.

[0071] As illustrated in [Fig. 1] by way of a specific example, device 2 is capable of triggering the display, by a display device 14 (for example, a screen), of an image IM2 of all or part of the subject UR2, this image IM2 being obtained (or derived) by processing from the image IM1 acquired by the image acquisition device 10. It is assumed hereafter that the display device 14 is included in the user interface 6, although other examples where the display device 14 is external to device 2 are possible. It should be noted that embodiments in which no image IM2 is displayed are also possible.

[0072] The image IM2 thus displayed may be identical or different from the image IM1 captured by the camera 10, as the case may be. The displayed image IM2 may, for example, be a version of the acquired image IM1 but in a different format or resolution than that of the acquired image IM1. An image conversion may, for example, be performed by the device 2 if the format of the obtained image IM1 does not correspond to the format of the image IM2 that one wishes to display on the screen 14, as described below in specific examples.

[0073] As shown in [Fig.2] according to a particular embodiment, the processor 2 controlled by the computer program PG1 here implements a number of modules, namely: an acquisition module MD2, a first detection module MD4, a first determination module MD6, a second detection module MD8, a localization module MD 10, a second determination module MD 12, and possibly also a supply module (or guidance module) MD 14.

[0074] More specifically, the MD2 acquisition module is configured to obtain an IM1 image of a UR2 subject (or person).

[0075] The first detection module MD4 is configured to detect, in the IM1 image, at least two first anatomical points PT1 associated with the skull of subject UR2.

[0076] The first determination module MD6 is configured to determine a first value VI representative of a reference distance DI separating said at least two first anatomical points PT1.

[0077] The second detection module MD8 is configured to detect, in the IM1 image, two (or at least two) second anatomical points PT2 associated with the subject's shoulders.

[0078] The MD10 localization module is configured to locate a sternal midline L1 equidistant from the second anatomical points PT2, this sternal midline L1 intersecting at an intersection point noted PN1 a segment SGI defined by the second anatomical points PT2.

[0079] The second determination module MD 12 is configured to determine Cl coordinates of at least one target anatomical point PT3 by calculating second distances D2x and D2y of said at least one anatomical target point PT3 vis-à-vis the intersection point PN1, these second distances D2x and D2y being determined from the first value V1 determined by the first determination module MD6.

[0080] Where appropriate, the MD 14 supply module is configured to provide (or generate, or render) a CM1 instruction to guide a user, namely the UR1 practitioner in this example, to said at least one anatomical target point PT3.

[0081] In a particular embodiment, the modules M2 to MD14 are controlled or implemented by the processor 4 by executing the computer program PG1, the latter being able to take the form of a software application or any other suitable form.

[0082] The configuration and operation of the MD2-MD14 modules of device 2 will be shown more precisely in the embodiment examples described below with reference to the figures. Note that the MD2-MD14 modules as shown in [Fig. 2] represent only one non-limiting implementation example of the invention.

[0083] Generally, for each step of the assisted anatomical localization process of the invention, the device 2 may include a corresponding module configured to perform said step.

[0084] As illustrated in Figures 3 to 7 according to particular embodiments, the steps of the assisted anatomical localization process of the invention implemented by the processing device 2, and more generally by the SY2 system, as previously described with reference to Figures 1-2, are now described. To this end, the device 2 executes the instructions of the computer program PG1 to implement the steps S2-S14 of the anatomical localization process.

[0085] In what follows, it is assumed that a practitioner UR1, for example, an emergency responder (a firefighter, paramedic, or other), uses the device 2, here taking the form of a smartphone or tablet, for example, to locate at least one target anatomical point PT3 of an individual UR2 requiring examination, for example, for first aid purposes. To do this, the practitioner UR1 positions the device 2 and triggers the camera 10 so as to acquire an image IM1 of all or part of the subject URL. It is assumed, for the sake of example, that the camera 10 is integrated into the device 2.

[0086] Thus, during a step S2 of acquisition (Figures 3-4), the device 2 obtains the image(s) IM1 acquired by the camera 10. This image IM1 is stored, as image data, in the memory 8 for further processing ([Fig. 1]). A plurality of images IM1 can thus be obtained during this step S2, for example in the form of a sequence of images or a video, these images IM1 then being processed in a manner analogous to that described below for the image IM1 under consideration.

[0087] Note that the IM1 image can be obtained (S2, [Fig.3]) in various ways by the device 2, for example by receiving the IM1 image from an entity external to the device 2 such as a remote server (not shown), or by extracting (or reading) the IM1 image from an embedded memory (for example memory 8) in the processing device 2.

[0088] As shown in [Fig. 4], the device 2 can cause the display device 14 to display an image IM2 obtained from the acquired image IML. The image IM2 may be identical to or different from the image IM1, particularly in terms of resolution and / or format. For example, it is assumed that the image IM2 is obtained by processing the image IM1, this processing including, for example, coordinate and / or resolution conversions from the image IM1 to the image IM2, as described in more detail later in specific examples.

[0089] The acquired IM1 image and the displayed IM2 image include all or part of the body of the individual UR2 undergoing assisted anatomical localization. In particular, it is considered here that the IM1 and IM2 images include all or part of the skull 20 (or head) and shoulders 26 of the subject UR2 ([Fig.4]).

[0090] During a detection step S4, the device 2 detects, in (or from) the IM1 image, at least two first anatomical points PT1 associated with the skull 20 of the subject UR2. These anatomical points PT1 are preferably chosen so as to vary minimally according to the morphology of the individuals. For example, it is assumed hereafter that the device 2 detects (S4) two first anatomical points PT1 associated with the eyes 22 of the subject visible in the IML image. Note, however, that it is possible to use other anatomical points as first anatomical points PT1, alternatively or cumulatively to the two aforementioned anatomical points of the eyes 22.

[0091] According to one example, the first anatomical points PT1 detected in S4 include at least one of the following: - anatomical points associated with the eyes 22 of subject UR2 visible in (or contained within) image IM1; and - anatomical points associated with the ears 24 of subject UR2 visible in (or contained within) the IML image

[0092] The use of the eyes 22 and / or the ears 24 as first anatomical points PT1 is advantageous in that these anatomical parts have the particularity of being arranged at a distance that varies proportionally with the human skeletal structure. This distance is relatively independent of the subject's morphology UR2. In particular, it has been observed that there is generally little adipose tissue in these regions of the human body, so the detection of these The points are not disturbed by such tissues, which allows for accurate and reliable detection.

[0093] As shown in [Fig. 5], the first anatomical points PT1 detected in S4 correspond, for example, to the two outer extremities of the eyes 22 of individual UR2. The use of the outer extremities of the eyes advantageously ensures precise and reliable detection of the first anatomical points PT1 regardless of the state of the eyelids, including when the eyes 24 are partially or totally closed, thus optimizing detection under all circumstances. Furthermore, such detection is possible regardless of the morphology of the eyes 22, including for almond-shaped or non-slanted eyes.

[0094] Similarly, the origin of the ears 24 can be used as first anatomical points PT1 insofar as these anatomical parts are generally devoid of fat and therefore allow reliable and precise detection.

[0095] S4 detection can be performed using a pre-trained artificial intelligence (AI) model (or algorithm) to recognize the anatomical points sought based on a training dataset. This AI model can be part of the PG1 computer program.

[0096] It should be noted that the presence or absence of eyes 22 in the IM1 image obtained in S2 depends in particular on the position of the subject UR2 relative to the camera 10. According to a specific example, during the detection step S4, the device 2 (for example, its artificial intelligence algorithm) searches for candidate anatomical points in the IM1 image and selects, from among the candidate anatomical points detected in the IM1 image, the first anatomical points PT1 to be used in the subsequent steps of the process. This search can be performed by analyzing the image using the aforementioned AI algorithm.

[0097] In a particular example, candidate anatomical points can be searched for in order of priority within the acquired IM1 image. In one embodiment, the device searches for the eyes 22 of subject UR2 and, if detected, selects as the first anatomical points PT1 the anatomical points associated respectively with the two eyes 22 thus detected. If, on the other hand, the eyes 22 cannot be detected (for example, because subject UR2 is lying prone), the device 2 searches for the ears 24 of subject UR2 and uses as the first anatomical points PT1 the anatomical points associated respectively with the two ears 24 thus detected. This ensures good accuracy and reliability of detection regardless of the position or morphology of subject UR2.According to one embodiment, device 2 only searches for eyes 22 if it has not previously managed to detect both eyes 22 of subject UR1, which ensures. good detection quality while minimizing the resources required, particularly in terms of capacity and processing time.

[0098] For the sake of simplicity in the description of the invention, it is hereafter considered that only one IM1 image is obtained during this step S2, although it is possible to obtain a plurality of IM1 images, for example in the form of a sequence of images or a video, these IM1 images then being processed in a manner analogous to that described below for the IM1 image considered.

[0099] During a determination step S6 (Figures 3 and 5), the device 2 determines a first value VI representing a reference distance DI separating the first two anatomical points PT1 associated in this example with the eyes 22 of subject UR2. This first value V1 is, for example, defined as a function of the number of pixels in the image IM1 obtained in S2 separating the first two anatomical points PT1. By counting the number of pixels between the two anatomical points PT1, one can, in particular, evaluate the size of the skeletal structure of subject UR2 as visualized by the camera 10. As indicated below, the distance DI between the eyes 22 (for example, between the outer extremities of the eyes in this example) can advantageously serve as a reliable reference for determining target anatomical points in the human body.

[0100] During a detection step S8 (Figures 3 and 6), the device 2 detects, in the IM1 image, two (or at least two) second anatomical points PT2 associated with the shoulders 26 of the subject UR2. To do this, the device 2 can, for example, use the aforementioned AI model to search for characteristics (shapes, dimensions, etc.) specific to shoulders in the human body.

[0101] According to a particular example, the second anatomical points PT2 detected at S8 are associated with the subject's shoulder points, although other parts of the shoulders can be used as a reference. Generally, the device 2 can be configured to detect, as second anatomical points, bony and / or muscular features of the shoulders, such as the upper bony points of the shoulders 26 and / or external muscular points of the shoulders 26.

[0102] During a localization step S10 (Figures 3 and 6), the device 2 locates (or identifies) a sternal midline L1 equidistant from the two second anatomical points PT2, the sternal midline L1 intersecting (or cutting) at an intersection point PN1 a segment SGI defined by the second anatomical points PT2 associated respectively with the two shoulders 26. In other words, the segment SGI terminates at its extremities by the two anatomical points PT2.

[0103] As illustrated in [Fig. 6], this sternal midline L1 forms the perpendicular bisector of the segment SGI defined by the second anatomical points PT2. This line L1 is here parallel to the Y-axis of the anatomical reference frame RFI associated with the subject's body UR2 while that the SGI segment extends in a direction parallel to the X axis of said anatomical reference frame RFI.

[0104] It has been observed that, despite the variability in the morphologies of people, the sternal midline L1 generally agrees with the position of the spine of subject UR2.

[0105] During this localization step S10, the device 2 can in particular determine and store the coordinates of the intersection point PN1, for example in an anatomical reference frame RFI associated with the subject's body UR2.

[0106] During a determination step S12 (Figures 3 and 6), the device 2 determines the coordinates, denoted Cl, of at least one anatomical target point PT3 of the subject UR2. For the sake of simplicity, it is hereafter considered that only one anatomical target point PT3 is determined during the process, although it is possible to determine and use a plurality of such anatomical target points in a manner analogous to that described in this disclosure for the anatomical target point PT3 under consideration. The anatomical target point PT3 thus identified may correspond to various anatomical points of the human body, as appropriate, for example, an anatomical point associated with a lung or the heart of the subject UR2, as described later in specific examples.

[0107] In this example, the coordinates Cl of the target anatomical point PT3 are considered to be defined according to the anatomical reference frame RFI associated with the subject UR2 as positioned in the image IM1. These coordinates Cl can therefore be expressed as a pair of positions (XI, Y2) along the X axis of the abscissas and Y axis of the ordinates of the anatomical reference frame RFI, respectively.

[0108] The determination S12 of the Cl coordinates can be carried out by calculating second distances, denoted D2x and D2y ([Fig. 6]), from the target anatomical point PT3 with respect to the intersection point PN1 along the X and Y axes of the RFI anatomical reference frame, respectively. These second distances Dx and Dy (or at least one of them) can be determined from the first value V1, which serves as the reference value.

[0109] In particular, one of these second distances, namely D2y in the following examples, is determined by applying a proportionality coefficient Ky to the first value VL. The other of these distances, namely D2x in this example, can be determined from, or be equal to, a constant (or predetermined value) denoted Kx.

[0110] More precisely, in the example considered, the distance D2x is determined at S12 such that D2x = Kx, where Kx is a constant (or predetermined value). The distance D2y is further determined (S12) as a function of the value VI previously determined at S6, that is, in this example, as a function of the value VI representative of the distance between the first two anatomical points PT1 associated with the eyes 22 of subject UR2. A proportionality coefficient Ky is applied to this constant Ky to deduce the distance D2y. The intersection point PN1 thus serves as a reference position to determine the position of the target anatomical point PT3 along the Y-axis of the anatomical reference frame RFI. To do this, device 2 can use as a reference the coordinates, in the anatomical reference frame RFI, of the intersection point PN1 located at S10.

[0111] The coordinates Cl of the anatomical target point PT3 can then be defined (S12) by the pair (XI, Y1) which is a function of the distances D2x and D2y respectively. According to a particular example, the coordinates Cl (XI, Y1) are such that XI = D2x and Y1 = D2y.

[0112] Note that the constant Kx and the proportionality coefficient Ky, associated respectively with the X and Y directions of the RFI anatomical reference frame, can be adapted by a person skilled in the art as appropriate, depending in particular on the type of anatomical target point PT3 sought.

[0113] According to a particular example, during the determination S12, the device 2 performs the following determinations: - determination of the second distance D2x along the first X axis of the anatomical reference frame RFI associated with the subject UR2 visible in (or contained within) the image IM1 such that Dix = Kx; and - determination of the second distance D2y along the second axis Y of the anatomical reference frame RFI such that D2y = Ky • VI, where Kx is a constant (or a positive value) and Ky is a proportionality coefficient.

[0114] As indicated above, D2x is here equal to the constant (or predetermined value, or coefficient) Kx, the first value VI not being used in this example to determine the abscissa coordinates (along X) of the target anatomical point PT3. Indeed, insofar as the reference position is used the point of intersection PN1 itself deduced from the gap between the two second anatomical points PT2 (at the level of the shoulders), it has been observed that it is not necessary to take into account the first value V1 to accurately determine the distance D2x along the abscissas in the subject's RFI anatomical frame of reference.On the other hand, the first value VI is taken into account for the calculation of the position of the target anatomical point PT3 along the Y ordinates (distance D2y) because the reference distance DI provides information on the subject's skeletal growth factor UR2 and therefore constitutes a reliable value for determining the position of the target anatomical point PT3 along the Y axis of the ordinates.

[0115] As illustrated by way of example in [Fig. 6], to detect the pulmonary focus of subject UR2 as the target anatomical point PT3, the values ​​of the constant Kx and the proportionality coefficient Ky such that Kx is equal (or approximately equal) to 0.7 and Ky is equal (or approximately equal) to 0.1. Thus, D2x = 0.7 and D2y = 0.1 * VI.

[0116] According to another example, it is possible to detect the mitral focus of subject UR2 as the target anatomical point PT3. In this case, the values ​​of the constant Kx and the proportionality coefficient Ky can be fixed so that Kx is equal to (or substantially equal to) 0.9 and Ky is equal to (or substantially equal to) 2.8. Thus, D2x = 0.9 and D2y = 2.8 * VI.

[0117] The above values ​​of the constant Kx and the proportionality coefficient Ky are, however, provided for illustrative purposes only. The values ​​of Kx and Ky can, in particular, be adapted according to the AI ​​model, and more broadly the operating system, used for the detection of the anatomical points PT1 and PT2.

[0118] According to a particular example, the coefficient Ky is different from 0 (Ky 0), for example to allow the localization of any anatomical target point PT3 other than on the ordinate position corresponding to the subject's spine UR2.

[0119] According to a particular example, the value of the coefficient Ky is adapted by the person skilled in the art, and may for example be different from 1 (Ky 1).

[0120] Once the coordinates Cl of the target anatomical point(s) PT3 have been determined, these coordinates can be used by device 2 in various ways depending on the use case. In particular, device 2 can inform practitioner UR1 of the position of the target anatomical point(s) PT3, although other uses of the result obtained in S12 are possible.

[0121] According to a particular example, device 2 triggers the display, by display device 14 (for example, a screen), of an image IM2 (called the second image) of all or part of the subject UR2, this second image IM2 being obtained from the image IM1 obtained in S2. This display can be performed as soon as the image IM1 is obtained S2, preferably in parallel with the execution of steps S4-S12, so that the practitioner can have visual feedback of the subject UR2 during the execution of the procedure. This display can, for example, be performed in real time in order, in particular, to help the practitioner UR1 to correctly position the camera 10 with respect to the subject UR2.

[0122] As already indicated, the second image IM2 thus displayed may be identical or different from the initial image IM1 captured by the camera 10, as the case may be. The displayed image IM2 may, for example, be a version of the acquired image IM1 having a format or resolution different from that of the acquired image IM1. An image conversion may, for example, be performed by the device 2 if the format of the obtained image IM1 does not correspond to the format of the image IM2 that one wishes to display on the screen 14, as described below.

[0123] According to a particular example, during the determination step S12, the device 2 determines the first coordinates Cl(XI, Yl) of the target anatomical point PT3 in the anatomical reference frame RFI, these first coordinates Cl(XI, Yl) being defined as a function of the pair (D2x, D2y). These first coordinates C1(X1, Yl) are then converted into second coordinates C2(X2, Y2) in an image reference frame RF2 associated this time with the second image IM2 displayed ([Fig. 7]). Such a conversion of the coordinates may be necessary to take into account the fact that the anatomical reference frame RFI in which the Cl coordinates of the target anatomical point PT1 were initially determined is different from the image reference frame RF2, i.e., the reference frame in which the image IM2 is displayed by the display device 14.This change in reference frame results in particular from the fact that the displayed image IM2 may have a different size (in terms of pixels) and / or resolution than the captured image IM1. For example, images IM1 and IM2 may have different DPI (dots per inch) values, corresponding to different levels of image quality.

[0124] As illustrated by way of example in [Fig. 7], the subject's body UR2, and therefore its RFI reference frame, can in particular be misaligned and / or offset relative to the RF2 reference frame of the IM2 image as displayed on the screen. This conversion can, for example, take into account an angular offset θ ([Fig. 7]) between the RFI and RF2 reference frames. The coordinate conversion can therefore include a rotation of the coordinates (XI, YI) by the angle θ in order to determine the location of the target anatomical point PT3 in the displayed IM2 image. This angle θ is, for example, defined by the shoulder segment SGI relative to the X-axis of the RF2 image reference frame.

[0125] Thanks to this coordinate conversion, the target anatomical point PT3 can advantageously be displayed accurately on (or in association with) the IM2 image, regardless of the position of the subject's body UR2 relative to the camera 10, and therefore regardless of the position or orientation of the subject UR2 relative to the contours of the IM2 image.

[0126] According to a particular example, during the determination S12 ([Fig. 3]), the device 2 performs a second coordinate conversion: it converts the second coordinates C2 defined by the pair (X2, Y2) into third coordinates C3 defined by the pair (X3, Y3), still within the display reference frame RF2 associated with the displayed image IM2. Such a conversion can be useful if the resolution of the image IM1 initially obtained in S2 differs from the resolution of the image IM2 displayed on the screen. This conversion from C2 to C3 makes it possible to express the position of the target anatomical point PT3 in the image reference frame RF2, taking into account the difference in resolution, along the X and Y axes, between the reference frames RF1 and RF2. For example, the image acquired at S2 may have a given resolution (e.g., 640 x 480 pixels) that is lower than the resolution of the displayed IM2 image (e.g., 1920 x 1080 pixels). During this conversion, device 2 calculates ratios between the resolutions of images IM1 and IM2 along the X and Y directions, respectively. These ratios are then used to convert the C2 coordinates into C3 coordinates. This conversion, linked to the change in image resolution, prevents any positioning errors of the target anatomical point PT3, and potentially of the visual indicator, as explained below.

[0127] According to a particular example (Figures 1-6), during a supply step S14 following steps S2-S12, the device 2 provides (or renders) an instruction CM1 to guide the user UR2 to the target anatomical point PT3, said instruction CM1 being determined according to the coordinates Cl of the target anatomical point PT3. This instruction CM1 can take various forms and be presented in various ways as appropriate.

[0128] According to a particular example, the INI instruction provided in S14 includes at least one of: a visual instruction; an audible instruction; and a vibratory instruction.

[0129] According to a particular example, the device 2 displays the image IM2 (derived from the image IM1 as already indicated) as well as at least one visual indicator INI associated with the target anatomical point PT3. Such a visual indicator INI can, in particular, be displayed superimposed on the displayed image IM2, according to the augmented reality display mode. The visual indicator INI is then representative of a target position associated with the target anatomical point PT3.

[0130] The visual indicator INI may take any form, for example that of a target or any other graphic object indicating the target anatomical point PT3 or a target position associated with this target anatomical point PT3 (for example a target position concordant with, or close to, the anatomical point PT2 where the practitioner must perform an auscultation).

[0131] According to a particular example, following steps S2-S12 ([Fig. 3]), device 2 sends a machine instruction (or command) CM2 which is a function of the coordinates of the target anatomical point PT3. This CM2 instruction is used to control or operate a system (not shown), such as, for example, a massage system, a medical treatment system (injection system, medical booth, wellness equipment, non-invasive or non-remotely controlled treatment system, etc.), other examples being possible. In response to this machine instruction CM2, the system can then automatically perform a given operation, for example by physically interacting with the subject UR2, for example to perform mechanical cardiac massage, an injection, or any other action, depending on the located target anatomical point(s) PT3.

[0132] As previously stated, the first VI value determined in S6 ([Fig. 3]) can represent the number of pixels separating the first two anatomical points PT1. In a particular example, during the supply step S14, the device 2 estimates a dimensional ratio between the number of pixels defined by the first VI value and a reference value. The device 2 then adjusts at least one dimension (size or other) of the visual indicator INI according to the dimensional ratio thus estimated. This reference value, which can be adjusted as appropriate, is the theoretical value of the reference distance Dl. For example, if the reference distance Dl corresponds to the distance between the subject's eyes, this theoretical value can be set at approximately 60 mm, although other values ​​are possible depending, in particular, on the positions of the eyes used as the first anatomical points PT1.

[0133] The device 2 can thus advantageously adapt the size, and more generally the appearance, of the visual indicator INI displayed on the screen (for example in superposition with the displayed image IM2) according to the distance of the camera 10 to the focal plane, so that the size of the visual indicator remains adapted regardless of the position of the camera 10 with respect to the body of the subject UR2.

[0134] The present invention advantageously allows for the precise, reliable, and rapid localization of anatomical points on a subject, particularly a human, or even an animal. The invention offers an ergonomic solution enabling any user, regardless of their level of training in anatomy or auscultation, to easily identify an anatomical point on the subject, even in noisy or suboptimal environments. The device of the invention significantly reduces the risk of human error, which is particularly advantageous for safe and effective medical interventions, as well as in the field of well-being.

[0135] The invention makes it possible, in particular, to assist or guide a practitioner effectively and ergonomically in locating anatomical points on a subject, for example, for auscultation or other purposes. By providing, for example, clear visual cues and compensating for variations in the patient's position, a practitioner can quickly and accurately locate anatomical points, even in challenging environments. This can prove particularly useful in emergency situations, where conditions are not always ideal for auscultation. In this way, it is possible, for example, to perform synchronous or asynchronous auscultation, to collect, for example, the subject's physiological parameters, or even to assist in establishing a diagnosis.

[0136] This guidance can advantageously be achieved using an augmented reality system that automatically locates the auscultation focus(s) by means of a body recognition algorithm and anthropometric calculations, making the application accessible and accurate for non-specialist users.

[0137] The invention may, for example, enable a person with any level of training to effectively perform a cardiopulmonary auscultation, or any other appropriate auscultation, at one or more anatomical points on the subject's body. To this end, the invention may, in particular, provide guidance, for example, to indicate where to place an instrument, such as a medical instrument (stethoscope, etc.), on the subject's body.

[0138] The ergonomics and ease of use of the device of the invention are major advantages. By simplifying the process of locating anatomical points, it reduces the cognitive load and potential errors of practitioners. This can improve not only the efficiency of medical procedures, but also the overall quality of patient care.

[0139] The concept of the invention is based in particular on the counterintuitive observation that, with few exceptions, the human skeletal structure maintains certain proportions, regardless of the individual's morphology and age. For example, while the eyes remain roughly the same size throughout a person's life, their skull grows, leading to a widening of the distance between the eyes as they grow. This widening increases proportionally to the skeletal structure of the human body. Thus, using predetermined proportionality coefficients Kx and Ky and a reference distance DI as previously described in specific examples, the position of a target anatomical point can be accurately estimated, taking into account the natural growth of an individual's skeleton with age.

[0140] A recognition algorithm can thus be implemented by the device of the invention to track and detect target anatomical points on a subject's body. An augmented reality module can also be advantageously used to overlay visual information onto an image of a subject in order, for example, to guide the placement of a stethoscope or any other instrument (for auscultation or otherwise), thereby enabling high-quality listening to cardiac and / or pulmonary sounds without prior training, particularly in physiology. The invention therefore offers a valuable tool for emergency interventions by non-specialized professionals.

[0141] As those skilled in the art will understand, all the embodiments and variations described above, some of which have been intentionally simplified for ease of explanation, are merely non-limiting examples of implementation of this disclosure. In particular, those skilled in the art may consider any adaptation or combination of the embodiments and variations described above to meet a specific need.

[0142] The present invention is therefore not limited to the embodiments described above but extends in particular to a localization method that would include secondary steps without departing from the scope of the present invention. The same would apply to a processing device, or more generally to a localization system, for implementing such a method.

Claims

Demands

1. A method for assisted anatomical localization, implemented by a processing device (2), said method comprising: a) obtaining (S2) an image (IM1) of a subject (UR2); b) detecting (S4), in said image, at least two first anatomical points (PT1) associated with the skull of the subject; c) determining (S6) a first value (VI) representative of a reference distance (Dl) separating said at least two first anatomical points; d) detecting (S8), in said image, two second anatomical points (PT2) associated with the shoulders (S26) of the subject; e) localizing (S10) a sternal midline (Ll) equidistant from the second anatomical points, the sternal midline intersecting at a point of intersection (PN1) a segment (SGI) defined by the second anatomical points;and f) determination (S 12) of coordinates (Cl) of at least one target anatomical point (PT3) by calculating second distances (D2x, D2y) of said at least one target anatomical point with respect to the point of intersection, at least one of the second distances being determined from the first value (VI).;

2. A method according to claim 1, wherein the first anatomical points detected in b) include at least one of: - anatomical points associated with the eyes (22) of the subject visible in the image; and - anatomical points associated with the ears (24) of the subject visible in the image.

3. A method according to claim 1 or 2, wherein the method comprises: - searching for candidate anatomical points in the image (IM1); and - selecting, from among the candidate anatomical points detected in the image, the first anatomical points (PT1) used in c).

4. A method according to any one of the preceding claims, wherein, during determination c), the first value (VI) is defined as a function of a number of pixels, in the image (IM1) obtained in a), separating said at least two first anatomical points (PT1).

5. A method according to any one of the preceding claims, wherein the sternal midline (Ll) forms a perpendicular bisector of the segment (SGI) defined by the second anatomical points.

6. A method according to any one of the preceding claims, wherein, during the determination f), one of the second distances (D2x, D2y) is determined by applying a proportionality coefficient (Ky) to the first value (VI).

7. Method according to claim 6, wherein determination f) comprises: - determination of a second distance D2x along a first axis of an anatomical reference frame (RFI) associated with the subject visible in the image such that D2x = Kx; and - determination of a second distance D2y along a second axis of the anatomical reference frame such that D2y = Ky * VI, where Kx is a constant and Ky is said proportionality coefficient.

8. A method according to any one of the preceding claims, wherein the method comprises: - displaying an image (IM2), referred to as the second image, according to an image reference frame (RF2), said second image being determined from the image (IM1) obtained in a); wherein the determination f) comprises: - determining first coordinates (XI, Y1), of said at least one anatomical target point (PT3) in the anatomical reference frame, said first coordinates (XI, Y1) being defined as a function of the pair (D2x, D2y); and - converting the first coordinates (XI, Y1), as second coordinates (X2, Y2), in the image reference frame (RF2) associated with said second image.

9. Method according to claim 8, wherein determination f) comprises: - conversion of the second coordinates (X2, Y2) as third coordinates (X3, Y3) in the display reference frame associated with a graphic representation of said image.

10. A method according to any one of the preceding claims, wherein the method comprises: e) providing instructions to guide a user to said at least one target anatomical point.

11. A method according to claim 10, wherein the instruction provided in e) includes at least one of: - a visual instruction; - an audible instruction; and - a vibration instruction.

12. Method according to claim 10 or 11, wherein the supply e) comprises: - display of the image and at least one visual indicator, superimposed on the displayed image, said at least one visual indicator being representative of a target position associated with said at least one anatomical target point.

13. A method according to claim 12, wherein the first value is representative of a number of pixels separating said at least two first anatomical points, wherein the supply e) comprises: - estimation of a dimension ratio between the number of pixels defined by the first value and a reference value; and - adaptation of at least one dimension of the visual indicator according to said dimension ratio.

14. Computer program (PG1) comprising instructions for carrying out the steps of a process according to any one of the preceding claims when said program is executed by a computer.

15. Processing device (2) comprising: - an acquisition module (MD2) configured to obtain an image of a subject; - a first detection module (MD4) configured to detect, in said image, at least two first anatomical points associated with the subject's skull; - a first determination module (MD6) configured to determine a first representative value of a reference distance (Dl) separating said at least two first anatomical points; - a second detection module (MD8) configured to detect, in said image, two second anatomical points associated with the subject's shoulders; - a localization module (MD10) configured to localize a sternal midline equidistant from the second points anatomical, the sternal midline intersecting at a point of intersection a segment defined by the second anatomical points; and - a second determination module (MD 12) configured to determine coordinates of at least one target anatomical point by calculating second distances of said at least one target anatomical point with respect to the point of intersection, the second distances being determined from the first value.

Images