WO2020141468A1 - Method and system for detecting position of a target area in a target subject - Google Patents

Method and system for detecting position of a target area in a target subject Download PDF

Info

Publication number
WO2020141468A1
WO2020141468A1 PCT/IB2020/050006 IB2020050006W WO2020141468A1 WO 2020141468 A1 WO2020141468 A1 WO 2020141468A1 IB 2020050006 W IB2020050006 W IB 2020050006W WO 2020141468 A1 WO2020141468 A1 WO 2020141468A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
captured image
parameters
parameter
captured
Prior art date
Application number
PCT/IB2020/050006
Other languages
French (fr)
Inventor
Mahabaleswara Ram BHATT
Shailendra Rao NALIGE
Shyam Vasudeva RAO
Original Assignee
Ehe Innovations Private Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ehe Innovations Private Limited filed Critical Ehe Innovations Private Limited
Publication of WO2020141468A1 publication Critical patent/WO2020141468A1/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/20Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • G06T7/73Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/74Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74Image or video pattern matching; Proximity measures in feature spaces
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/766Arrangements for image or video recognition or understanding using pattern recognition or machine learning using regression, e.g. by projecting features on hyperplanes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/695Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/20Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
    • A61B2034/2046Tracking techniques
    • A61B2034/2065Tracking using image or pattern recognition
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03Recognition of patterns in medical or anatomical images

Definitions

  • the present subject matter is generally related to image processing and analysis and more particularly, but not exclusively, to method and system for detecting position of a target area in a target subject.
  • Precise object detection is important during surgery because the surgery equipment or a treating device has to operate defective portion only and not any other portion.
  • the imaging and computer vision based techniques have been utilized to aid in this aspect. In fact, the imaging based techniques are used for both diagnostic and guiding aids for surgical activities.
  • the analyzed images are used as guiding information, it is highly challenging for reproducing exactly same position and orientation for camera and object during instant image capture. Additionally, illumination conditions on the object, would also matter. Because the co-ordinates for organ (object) and camera (image frame) are independent in world coordinates. Also, it is challenging for real time tracking and guiding of surgical instrument to access the precise position of region in the organ where the surgery has to be carried out with minimal invasiveness based on the earlier diagnosed spot information.
  • object coordinate system there would not be any definite relation between object coordinate system and camera coordinates system unless they are rigidly tied together.
  • the object and camera conditions would differ because both are mutually independent.
  • the world co-ordinate system would give varying translation, rotation and scaling unless the instant image acquisition conditions made exactly same. Assuring the above said condition is very hard practically because object and camera are independent degrees of freedom respective translation, rotation and scaling.
  • the method comprises capturing an image of the target area using an image sensor. Thereafter, the method comprises converting, by an image analysis and processing system associated with the image sensor, the captured image from Cartesian coordinate system to spherical coordinate system. The method comprises obtaining regression parameters for captured image using a predefined regression technique on the spherical coordinate system. The captured image is transformed based on the regression parameters. The method comprises detecting the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject.
  • the present disclosure discloses an image analysis and a processing system for detecting position of a target area in a target subject.
  • the image analysis and processing system comprises a processor and a memory communicatively coupled to the processor, wherein the memory stores the processor-executable instructions, which, on execution, causes the processor to capture an image of the target area using an image sensor associated with the image analysis and a processing system.
  • the processor converts the captured image from Cartesian coordinate system to spherical coordinate system. Thereafter, the processor obtains regression parameters for captured image using a predefined regression technique on the spherical coordinate system.
  • the processor transforms the captured image based on the regression parameters.
  • the processor detects the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject.
  • the system comprises an image position controller associated with the image sensor for repositioning the image sensor, wherein the image sensor is repositioned to capture the image when there is a mismatch between the captured image and the stored image.
  • FIG. 1 shows an exemplary environment for detecting position of a target area in a target subjectin accordance with some embodiments of the present disclosure
  • FIG. 2 shows a block diagram illustrating an exemplary fordetecting position of a target area in a target subjectin accordance with some embodiments of the present disclosure
  • FIG. 3A illustrates translation, rotation and scaling with respect to two independent coordinate systems in accordance with some embodiments of the present disclosure
  • FIG.3B-3C illustrates process of image formation in accordance with some embodiments of the present disclosure
  • FIG.4, FIG.5A, FIG.5B illustrates process of partitioning the image in accordance with some embodiments of the present disclosure
  • FIG.6A illustrates a method for identifying coarse translated position of the image patch in the captured image frame in accordance with some embodiments of the present disclosure
  • FIG.6B illustrates process for tracking and detection of location of the object in accordance with some embodiments of the present disclosure
  • FIG.7A-7B shows exemplary sampled points of stored image and captured image in cartesian coordinate and spherical co-ordinate system in accordance with some embodiments of the present disclosure
  • FIG.7C shows exemplary regression ellipse fit for partial intersected region of the captured image and the stored image in accordance with some embodiments of the present disclosure
  • FIG.7D shows exemplary intersection region where captured image is inside stored image in accordance with some embodiments of the present disclosure
  • FIG.7E shows geometrical interpretation of using regression based scale and rotation parameters for matching and identification for partial region intersection scenario in accordance with some embodiments of the present disclosure
  • FIG.7F shows geometrical interpretation of using regression based scale and rotation parameters for matching and identification for full region intersection scenario in accordance with some embodiments of the present disclosure
  • FIG.7G shows estimation process for ellipse fit parameters in accordance with some embodiments of the present disclosure
  • FIG.8A, FIG.8B, FIG.8C illustrates the process of identification of scale parameter and rotation parameter for the target image and the stored imagein accordance with some embodiments of the present disclosure
  • FIG.9 shows a flowchart illustrating a method for identifying intrinsic image parameters required for rotation and scalingin accordance with some embodiments of the present disclosure.
  • FIG.10 illustrates a process of acquiring image corresponding to the target image when the object undergoes unknown translated positionin accordance with some embodiments of the present disclosure.
  • the present disclosure relates to method and system for detecting position of a target area in a target subject.
  • the method comprises capturing an image of the target area using an image sensor. Thereafter, the method comprises converting, by an image analysis and processing system associated with the image sensor, the captured image from cartesian coordinate system to spherical coordinate system.
  • the method comprises obtaining regression parameters for captured image using a predefined regression technique on the spherical coordinate system.
  • the captured image is transformed based on the regression parameters.
  • the regression parameters comprises scale parameter, rotation parameter and translation parameter.
  • the system obtains stored image from a database associated with the system the stored image is obtained during diagnosis of the subject.
  • the stored image is converted from cartesian to spherical coordinates and the regression parameters are obtained.
  • the orientation and lengths of major axis and minor axis, area of ellipse are also obtained.
  • the system also computes orientation and lengths of major axis and minor axis, area of ellipse of the captured image.
  • the system finds angle between orientation of the captured image and the stored image either using major axis or minor axis and also calculates ratio between areas or major axis or minor axis between the captured image with the area of stored image which are set as scale parameters.
  • the system applies spatial transformation to the co-ordinates of the stored image.
  • the system checks for match between the stored images with the transformed image. If matches, the system detects that the image is identified or else the system continues the process by repeating image capturing by repositioning image till the image match is obtained.
  • target object/area corresponds to a part of tissue or part of organ which is identified as defective during diagnosis of a subject.
  • the subject may be a human.
  • the environment is described using the biomedical application but should not be construed as limiting to this application only.
  • the present disclosure is applicable for any other application in which detection of target area is required for performing one or more functions.
  • the environment comprises an image sensor 102 such as camera configured to illuminate specific area referred as target area of a target subject using illumination rays 103 and irradiant rays 104 to capture image of the target subject. Further, the image sensor 102 is associated with camera position and motion controller 105 and camera image acquisition controller 106.
  • the captured image is provided to image analysis and processing system 107 (alternately referred as system).
  • the system 107 converts the captured image from cartesian coordinate system to spherical coordinate system. Thereafter, the system 107 obtains regression parameters for the captured image using a predefined regression technique on the spherical coordinate system.
  • the regression parameters comprises scale parameter, rotation parameter and translation parameter. Further, the image analysis and processing system 107 transforms the captured image based on the regression parameters. In an embodiment, the transformation is based on the following steps.
  • the system 107 identifies angle parameter based on difference between angle of orientation of the target area in the captured image and angle of orientation of the target area in the stored image using at least one of major axis and the minor axis of an image plane associated with angle coordinate of the image sensor associated with the captured image and angle coordinate of the image sensor associated with the stored image.
  • the method further comprises identifying scale parameter based on difference between area of the stored image based on position parameter related to the stored image and area of the captured image based on position parameter related to the captured image.
  • the angle parameter and the scale parameter is stored as first intrinsic parameters. Based on the first intrinsic parameters the system 107 performs at least one of resizing the captured image and rotating the captured image.
  • the process of resizing the captured image and rotating the captured image involves upscale and downscale process performed using a predefined deep learning technique.
  • the system 107 receives stored image 108 from a database associated with the system 107.
  • the stored image 108 is obtained during diagnosis of the subject.
  • the system 107 checks for match between the stored image 108 and the captured image.
  • the control is shifted to a treating device 113 which is associated with position and motion controller 110 configured to position accurately the treating device 113 based on the detected position and laser is switched on using switching controller 112 with a pre-estimated delay 111 after positioning on the target object 101.
  • the control under mismatch condition, is shifted to the camera image acquisition controller 106 and camera position and motion controller 105 for repositioning the camera to capture the image of the target object 101 again.
  • the system 107 also obtains translation point of the image by performing the following steps. At first, the system 107 identifies translation position parameters of the captured image using a predefined technique. Thereafter, the system 107 translates the stored image 108 based on translation position parameters of the captured image. Further, the system 107 detects a match between the captured image and the stored image 108. Upon detecting the match, the translation point of the image is identified as the second intrinsic parameter. In an embodiment, first intrinsic parameters and the second intrinsic parameter are converted into extrinsic parameters using predefined navigation techniques to reposition the image sensor 102. The extrinsic parameters are provided to the camera position and motion controller 105associated with the image sensor 102 for repositioning the image sensor 102. The image sensor 102 is repositioned to capture the image when there is a mismatch between the captured image and the stored image 108.
  • FIG. 2 shows a block diagram illustrating an exemplary fordetecting position of a target area in a target subjectin accordance with some embodiments of the present disclosure.
  • the block diagram comprise three main subsystems, namely camera subsystem 201, computing subsystem 207for generating various control signals and image/video frames processing, and treating device subsystem 215.
  • the camera subsystem 201 is capable of taking control parameters from the computing subsystem 207 for camera position and motion controller 202, image/video acquisition 204. subsequently, the acquired image frame is stored and transferred to the computing sub system 207.
  • the computing subsystem 207 is configured to generate camera extrinsic control signal/parameters 210 and treating device control signal 216 based on the processes described in detail in the below description using image/video frames 206. This provides the intrinsic target image features 211 for image matching condition based decision making purpose at block 214. If there is a match, the control is passed on to the treating device subsystem for positioning the treating device. If there is no match, the control is passed on to the camera subsystem to capture the image again. Further the computing subsystem 207 comprises a target image of the target subject that is acquired during diagnosis process of the target subject.
  • the treating device subsystem 215 is configured to control the treating device 223 using the treating device controller 218 which is driven by extrinsic control parameter based on image processing activities at the computing subsystem 207. While controlling the treating device 223 upon precise positioning, the camera viewing conditions are maintained at fixed pre determined position as guided by control methods as indicated by signal 201 and 222.
  • FIG. 3 A illustrates two independent coordinate systems having (x c ,y c ,z c ) axes for object coordinate system 303 and (xo,yo,zo ) axes for camera coordinate systems 302.
  • the object coordinate system 303 and the camera coordinate system302 are relatively located in world coordinate system 301 having (xw,y w ,z w ) axes.
  • (Tow, Row, Sow ) which represents translation, rotation and scaling intrinsic parameters of the target object 114 in object coordinate system 303 with respect to world coordinate system 301.
  • (Tcw,Rcw,Scw ) represents translation, rotation and scaling intrinsic parameters in camera coordinate system 302 with respect to world coordinate system 301.
  • (Tco,Rco,Sco) represents translation, rotation and scaling intrinsic parameters between camera and object coordinate systems.
  • FIG.3B illustrates process of image formation in accordance with some embodiments of the present disclosure.
  • FIG.3B shows a graph illustrating image capturing of a target object 101 having three different coordinate systems such as object coordinate system, camera coordinate system and world coordinate system.
  • FIG.3B shows the captured image frame having camera field view 306, wherein image patch 307 corresponds to the target object 304, when optical axis of image sensor 305 is assumed to be in perpendicular to image plane in (x- y) space plane.
  • FIG.3C illustrates another embodiment of image capturing with a field of view of the camera which is EFGH 309, wherein image patch 308 corresponds to the target object 304 by having optical axis 310 orientation of the image sensor (also referred as camera sensor) 305 perpendicular to the image plane EFGH.
  • the image plane EFGH has degrees of freedom for rotations around both x-axis and z-axis.
  • FIG.4 illustrates process of partitioning the image 309 captured in the field of view of the camera.
  • the point S is identified as (m,n) in the captured image 309.
  • the sub image 613, 614, 615 and 616 are represented respectively as below:
  • a window 330 is selected as shown in FIG.5A which is not in arbitrary way or a window is selected either from any corner of the image frame.
  • any sliding window based object detection starts form left most corner of the image frame and sliding operation of the window.
  • the initial coordinate point (m,n) of window sliding for scanning is selected on the basis of imaging axis which intersects the image plane shown as point S in FIG.3C, where centre point of window typical sizes are 5x5, 7x7, etc. Based on this, the image is split into four sub images. The splitting will yield in various images such as 313, 314, 315, 316 as shown in FIG.5B.
  • FIG.6 A illustrates a method for identifying coarse translated position of the image patch in the captured image frame 304 corresponding to the target object 308 [as shown in FIG.3C]
  • the process involves exploitation of either sliding window based or as an example Viola and Jones technique which is suitable for a multicore computing process and fast computation.
  • the process starts with identifying four partitioned sub-image in the manner described in FIG. 4 and described as image arrays in an appropriate way as mentioned above asQi,Qn,Qm and Win.
  • the present disclosure illustrates the process of splitting the image frame 601 into four sub images 602, 603, 604 and 605 with an initial point as (m,n), which is determined by the intersection of optical axis of camera and the image frame 601.
  • the four integral process sums as a stream processes concurrently to exploit four core processing and facilitates in quick processing.
  • the sub images are matched with template image 606 (stored image 108) stored with initial window selected image patch.
  • the Viola and Jones object detection process 607 is implemented in four streams. If the match condition is identified, then the stream processing is stopped and then either of centre or centroid of the image patch is identified and the co-ordinates are stored appropriately as coarse translated position of the captured image at block 607. At block 608, the centre coordinates are converted to camera position coordinates.
  • FIG.6B illustrates process for tracking and detection of location of the object.
  • camera position is controlled to be at some convenient starting location so that initial field of view is obtained and instant image capturing process 610 enables to acquire an image frame.
  • the captured image and the stored image 612 are used to find the coarse translation position of image patch corresponding to the target object using coarse translation matching process using the embodiment described in FIG.6 A.
  • the camera is positioned to view the target object 101 by providing appropriate position coordinate to camera position and motion controller 105 and capture next instant image frame that assures to have small translation to the target image.
  • the pixel position which is in Cartesian coordinates for the stored image is converted to spherical coordinates.
  • scale and rotation parameters are identified using regression convolution neural network.
  • the image is resized with suitable coordinates. If there is only change in angle parameter, then the image is rotated. However, if there is both scale change and angle change then the image is resized and rotated.
  • the target image is matched with captured image.
  • first intrinsic parameters i.e. scale parameter and the angle parameter that is in spherical coordinates are converted into suitable extrinsic camera parameter and then camera is positioned to view suitably and to continue with image frame acquisition process.
  • FIG.7A(i) shows both stored image 701 and captured image 702.
  • FIG. 7A(ii) represents spherical coordinate points corresponding to FIG.7A(i).
  • FIG.7B(i) is replica of FIG.7A(i) with difference of images 701 and 702 overlapped partially.
  • FIG.7B(ii) represents spherical coordinate representation system 709corresponding to FIG.7B(ii).
  • the straight lines 710 and 711 shown in FIG. 7B (ii) is regression line fit for the selected sample points represented in spherical coordinate system for stored image 70 land the captured image 702 respectively.
  • FIG.7C (i) and FIG7C (ii) are replica of 7B (i) and 7B (ii), with a difference of regression ellipse fits 712 and 713 with respective major axes as 714 and 716. Additionally, 715 and 717 are respective minor axes.
  • the parameter difference angle f 714 is obtained by using slopes parameters major axes.
  • FIG.7D (i) and FIG.7D (ii) are replica of FIG.7C (i) and FIG.7C (ii), respectively with difference that regions of stored image is fully overlapped with captured image.
  • FIG. 7E shows geometric image parameters.
  • FIG.7F is replica of FIG.7E with a difference of overlap regions of regression fit ellipses.
  • FIG.7G shows estimation process for ellipse fit parameters.
  • FIG.8A, FIG.8B, FIG.8C illustrates the process of identification of scale parameter and rotation parameter for the target image and the stored image.
  • the angle parameter is identified based on difference between angle of orientation of the target area in the captured image and angle of orientation of the target area in the stored image using at least one of major axis and the minor axis of an image plane associated with angle coordinate of the image sensor associated with the captured image and angle coordinate of the image sensor associated with the stored image.
  • the scale parameter is identified based on difference between area of the stored image based on position parameter related to the stored image and area of the captured image based on position parameter related to the captured image.
  • the angle parameter and the scale parameter are stored as first intrinsic parameters.
  • block 802 and 805 illustrates steps involved in the process of identifying the scale parameter and the rotation parameter for the target image and the stored image respectively.
  • the compositional spatial transformation network is formulated based on the scale parameter and the rotation parameter/angle parameter using technique described in FIG.8B.At 808, the system provides the scaled and rotated target image patch using the compositional spatial transformation network.
  • an image grid structure (series of intersection of horizontal and vertical lines, typically two dimensional positional point) is generated for image pixels in the image frame 810 by image grid generator 812.
  • a warped image grid structure is generated using the angle parameter cp.
  • a new image is composed at block 815 at each position of new grid structure obtained at 814 that serves as an input to the fast scaled convolution neural network to either up sample or down sample process at block 807 based on scaling parameters, s, as per FIG.8 A.
  • FIG.8C illustrates process of up scaling or downscaling of the image based on convolution neural network.
  • the stored image is either downscaled or upscaled depending on value of scale parameter.
  • scale, s is checked whether it is greater 1 or not. If s> 1, then flow computation is shifted to block 813, where image is set for down sampling with any down sample network to arrive at required image 811. If s ⁇ l, then flow is shifted block 815, where up sampling sub network is performed to arrive at required image 812.
  • FIG.9 shows a flowchart illustrating a method for identifying intrinsic image parameters required for rotation and scaling in accordance with some embodiments of the present disclosure.
  • the method involves performing image capturing after positioning the camera view using camera position and motion controller 106 by utilizing intrinsic image parameters to generate the camera extrinsic parameters.
  • the method involves buffering the coarse translated localized image.
  • the method comprises converting the captured image, which are typically in cartesian coordinates to spherical coordinate transformation.
  • the method comprises performing regression network to get the regression parameters either for linear function or ellipse fit for the captured image.
  • the method comprises performing conversion of cartesian coordinates to spherical image coordinates for stored image 903.
  • the method comprises obtaining regression parameters using regression network on the spherical coordinates from block 912 either for either linear fit or ellipse fit.
  • the method comprises utilizing the regression parameters obtained for stored image with corresponding similar regression parameters for the captured image.
  • the method comprises composing scaled and rotated template corresponding to stored image.
  • the method comprises checking for match between the images using block 915 and the output of buffered image at 905. If there is no match, the processing is repeated by initiating the block 902, otherwise the method proceeds to block 918.
  • controller of surgery process is initiated to control the treating device.
  • FIG.10 illustrates a process of acquiring image corresponding to the target image when the object undergoes unknown translated position in accordance with some embodiments of the present disclosure.
  • the camera is directed to set at an instant pose and acquire an image.
  • the captured image 1002 is used to find the object location that provides translated position with respect to camera imaging focus axis.
  • the captured image 1004 is extracted from the translated position and the image from the instant acquired position is provided for matching at block 1010.
  • the stored image 1006 (which is target image corresponding to tissue or organ acquired during diagnostic time) is used to find the rotation and scaling process.
  • the image and its intrinsic image parameters 1009 are matched with the captured image 1004.
  • the intrinsic image parameters are converted to extrinsic parameters using techniques which includes but not limited to robotic navigation algorithm. It indicates the repositioning of camera for next image acquisition to arrive at corrected precise image coordinates, for object location.
  • the process of instant image acquisition is guided by the extrinsic parametersl013 obtained at block 1012.
  • the translation position parameters 1005 are obtained at block 1003.
  • the image (indirectly object) position and allied required parameter is shifted to control the treating device for performing surgical activities.
  • the present disclosure provides a method and system for accurate detection of position of target area in a target subject with minimal invasiveness.
  • the method of present disclosure generates precise coordinates for the treating device using intrinsic and extrinsic parameters of the camera.
  • the present disclosure uses target image captured during diagnosis of the subject and the target image features are used for detecting accurate position of the target area in real time.
  • the present disclosure enables real time tracking and guiding of treating device to access precise position of target area in the target subject.
  • an embodiment means “one or more (but not all) embodiments of the invention(s)" unless expressly specified otherwise.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Surgery (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Signal Processing (AREA)
  • Robotics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Image Analysis (AREA)

Abstract

Disclosed herein is a method and system for detecting position of a target area in a target subject. The method comprises capturing an image of the target area using an image sensor. The captured image is converted from cartesian coordinate system to spherical coordinate system by an image analysis and processing system associated with the image sensor. The method further comprises obtaining regression parameters for the captured image using a predefined regression technique on the spherical coordinate system. Thereafter, the captured image is transformed based on the regression parameters. The method further comprises detecting the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject. The present disclosure provides a mechanism for accurate detection of position of a target area.

Description

METHOD AND SYSTEM FOR DETECTING POSITION OF A TARGET AREA IN A
TARGET SUBJECT
TECHNICAL FIELD
The present subject matter is generally related to image processing and analysis and more particularly, but not exclusively, to method and system for detecting position of a target area in a target subject.
BACKGROUND
Precise object detection is important during surgery because the surgery equipment or a treating device has to operate defective portion only and not any other portion. The imaging and computer vision based techniques have been utilized to aid in this aspect. In fact, the imaging based techniques are used for both diagnostic and guiding aids for surgical activities.
Though there are advantages of imaging based techniques, there still exist some lacunas. The content of images and the process of acquisition of images using camera are highly dependent on position, illumination environment, object to be imaged and camera condition while imaging. The tissues or organs are usually diagnosed at one time and set to surgical activities at some other time. Obviously, the posing of object to camera is very subjective and varies. At the same time camera coordinate conditions may also vary during diagnosing instant and surgical activities.
Though the analyzed images are used as guiding information, it is highly challenging for reproducing exactly same position and orientation for camera and object during instant image capture. Additionally, illumination conditions on the object, would also matter. Because the co-ordinates for organ (object) and camera (image frame) are independent in world coordinates. Also, it is challenging for real time tracking and guiding of surgical instrument to access the precise position of region in the organ where the surgery has to be carried out with minimal invasiveness based on the earlier diagnosed spot information.
Further, it is important to note that there would not be any definite relation between object coordinate system and camera coordinates system unless they are rigidly tied together. The object and camera conditions would differ because both are mutually independent. The world co-ordinate system would give varying translation, rotation and scaling unless the instant image acquisition conditions made exactly same. Assuring the above said condition is very hard practically because object and camera are independent degrees of freedom respective translation, rotation and scaling.
The information disclosed in this background of the disclosure section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
SUMMARY
Disclosed herein is a method of detecting position of a target area in a target subject. The method comprises capturing an image of the target area using an image sensor. Thereafter, the method comprises converting, by an image analysis and processing system associated with the image sensor, the captured image from Cartesian coordinate system to spherical coordinate system. The method comprises obtaining regression parameters for captured image using a predefined regression technique on the spherical coordinate system. The captured image is transformed based on the regression parameters. The method comprises detecting the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject.
Further, the present disclosure discloses an image analysis and a processing system for detecting position of a target area in a target subject. The image analysis and processing system comprises a processor and a memory communicatively coupled to the processor, wherein the memory stores the processor-executable instructions, which, on execution, causes the processor to capture an image of the target area using an image sensor associated with the image analysis and a processing system. The processor converts the captured image from Cartesian coordinate system to spherical coordinate system. Thereafter, the processor obtains regression parameters for captured image using a predefined regression technique on the spherical coordinate system. The processor transforms the captured image based on the regression parameters. Finally, the processor detects the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject. The system comprises an image position controller associated with the image sensor for repositioning the image sensor, wherein the image sensor is repositioned to capture the image when there is a mismatch between the captured image and the stored image. The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, explain the disclosed principles. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the figures to reference like features and components. Some embodiments of system and/or methods in accordance with embodiments of the present subject matter are now described, by way of example only, and regarding the accompanying figures, in which:
FIG. 1 shows an exemplary environment for detecting position of a target area in a target subjectin accordance with some embodiments of the present disclosure;
FIG. 2 shows a block diagram illustrating an exemplary fordetecting position of a target area in a target subjectin accordance with some embodiments of the present disclosure;
FIG. 3Aillustrates translation, rotation and scaling with respect to two independent coordinate systems in accordance with some embodiments of the present disclosure;
FIG.3B-3C illustrates process of image formation in accordance with some embodiments of the present disclosure;
FIG.4, FIG.5A, FIG.5B illustrates process of partitioning the image in accordance with some embodiments of the present disclosure;
FIG.6A illustrates a method for identifying coarse translated position of the image patch in the captured image frame in accordance with some embodiments of the present disclosure;
FIG.6B illustrates process for tracking and detection of location of the object in accordance with some embodiments of the present disclosure; FIG.7A-7B shows exemplary sampled points of stored image and captured image in cartesian coordinate and spherical co-ordinate system in accordance with some embodiments of the present disclosure;
FIG.7C shows exemplary regression ellipse fit for partial intersected region of the captured image and the stored image in accordance with some embodiments of the present disclosure;
FIG.7D shows exemplary intersection region where captured image is inside stored image in accordance with some embodiments of the present disclosure;
FIG.7E shows geometrical interpretation of using regression based scale and rotation parameters for matching and identification for partial region intersection scenario in accordance with some embodiments of the present disclosure;
FIG.7F shows geometrical interpretation of using regression based scale and rotation parameters for matching and identification for full region intersection scenario in accordance with some embodiments of the present disclosure;
FIG.7G shows estimation process for ellipse fit parameters in accordance with some embodiments of the present disclosure;
FIG.8A, FIG.8B, FIG.8C illustrates the process of identification of scale parameter and rotation parameter for the target image and the stored imagein accordance with some embodiments of the present disclosure;
FIG.9 shows a flowchart illustrating a method for identifying intrinsic image parameters required for rotation and scalingin accordance with some embodiments of the present disclosure; and
FIG.10 illustrates a process of acquiring image corresponding to the target image when the object undergoes unknown translated positionin accordance with some embodiments of the present disclosure.
It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, state transition diagrams, pseudo code, and the like represent various processes which may be substantially represented in computer readable medium and executed by a computer or processor, whether such computer or processor is explicitly shown.
DETAILED DESCRIPTION
In the present document, the word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
While the disclosure is susceptible to various modifications and alternative forms, specific embodiment thereof has been shown by way of example in the drawings and will be described in detail below. It should be understood, however that it is not intended to limit the disclosure to the specific forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternative falling within the scope of the disclosure.
The terms“comprises”,“comprising”,“includes”,“including” or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a setup, device, or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a system or apparatus proceeded by“comprises... a” does not, without more constraints, preclude the existence of other elements or additional elements in the system or method.
The present disclosure relates to method and system for detecting position of a target area in a target subject. The method comprises capturing an image of the target area using an image sensor. Thereafter, the method comprises converting, by an image analysis and processing system associated with the image sensor, the captured image from cartesian coordinate system to spherical coordinate system. The method comprises obtaining regression parameters for captured image using a predefined regression technique on the spherical coordinate system. The captured image is transformed based on the regression parameters. The regression parameters comprises scale parameter, rotation parameter and translation parameter. The system obtains stored image from a database associated with the system the stored image is obtained during diagnosis of the subject. The stored image is converted from cartesian to spherical coordinates and the regression parameters are obtained. Also, the orientation and lengths of major axis and minor axis, area of ellipse are also obtained. The system also computes orientation and lengths of major axis and minor axis, area of ellipse of the captured image. The system finds angle between orientation of the captured image and the stored image either using major axis or minor axis and also calculates ratio between areas or major axis or minor axis between the captured image with the area of stored image which are set as scale parameters. Further, the system applies spatial transformation to the co-ordinates of the stored image. The system checks for match between the stored images with the transformed image. If matches, the system detects that the image is identified or else the system continues the process by repeating image capturing by repositioning image till the image match is obtained.
In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration of embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.
The environment is described by considering a biomedical application as an example, wherein target object/area corresponds to a part of tissue or part of organ which is identified as defective during diagnosis of a subject. As an example, the subject may be a human. The environment is described using the biomedical application but should not be construed as limiting to this application only. The present disclosure is applicable for any other application in which detection of target area is required for performing one or more functions.
The environment comprises an image sensor 102 such as camera configured to illuminate specific area referred as target area of a target subject using illumination rays 103 and irradiant rays 104 to capture image of the target subject. Further, the image sensor 102 is associated with camera position and motion controller 105 and camera image acquisition controller 106. The captured image is provided to image analysis and processing system 107 (alternately referred as system). The system 107 converts the captured image from cartesian coordinate system to spherical coordinate system. Thereafter, the system 107 obtains regression parameters for the captured image using a predefined regression technique on the spherical coordinate system. The regression parameters comprises scale parameter, rotation parameter and translation parameter. Further, the image analysis and processing system 107 transforms the captured image based on the regression parameters. In an embodiment, the transformation is based on the following steps.
At first, the system 107 identifies angle parameter based on difference between angle of orientation of the target area in the captured image and angle of orientation of the target area in the stored image using at least one of major axis and the minor axis of an image plane associated with angle coordinate of the image sensor associated with the captured image and angle coordinate of the image sensor associated with the stored image. The method further comprises identifying scale parameter based on difference between area of the stored image based on position parameter related to the stored image and area of the captured image based on position parameter related to the captured image. The angle parameter and the scale parameter is stored as first intrinsic parameters. Based on the first intrinsic parameters the system 107 performs at least one of resizing the captured image and rotating the captured image. In an embodiment, the process of resizing the captured image and rotating the captured image involves upscale and downscale process performed using a predefined deep learning technique.
In an embodiment, the system 107 receives stored image 108 from a database associated with the system 107. The stored image 108 is obtained during diagnosis of the subject. At 109, the system 107 checks for match between the stored image 108 and the captured image. Under match condition, the control is shifted to a treating device 113 which is associated with position and motion controller 110 configured to position accurately the treating device 113 based on the detected position and laser is switched on using switching controller 112 with a pre-estimated delay 111 after positioning on the target object 101. In an embodiment, under mismatch condition, the control is shifted to the camera image acquisition controller 106 and camera position and motion controller 105 for repositioning the camera to capture the image of the target object 101 again.
In an embodiment, the system 107 also obtains translation point of the image by performing the following steps. At first, the system 107 identifies translation position parameters of the captured image using a predefined technique. Thereafter, the system 107 translates the stored image 108 based on translation position parameters of the captured image. Further, the system 107 detects a match between the captured image and the stored image 108. Upon detecting the match, the translation point of the image is identified as the second intrinsic parameter. In an embodiment, first intrinsic parameters and the second intrinsic parameter are converted into extrinsic parameters using predefined navigation techniques to reposition the image sensor 102. The extrinsic parameters are provided to the camera position and motion controller 105associated with the image sensor 102 for repositioning the image sensor 102. The image sensor 102 is repositioned to capture the image when there is a mismatch between the captured image and the stored image 108.
FIG. 2shows a block diagram illustrating an exemplary fordetecting position of a target area in a target subjectin accordance with some embodiments of the present disclosure. As shown in Fig.2, the block diagram comprise three main subsystems, namely camera subsystem 201, computing subsystem 207for generating various control signals and image/video frames processing, and treating device subsystem 215.
The camera subsystem 201 is capable of taking control parameters from the computing subsystem 207 for camera position and motion controller 202, image/video acquisition 204. subsequently, the acquired image frame is stored and transferred to the computing sub system 207.
The computing subsystem 207 is configured to generate camera extrinsic control signal/parameters 210 and treating device control signal 216 based on the processes described in detail in the below description using image/video frames 206. This provides the intrinsic target image features 211 for image matching condition based decision making purpose at block 214. If there is a match, the control is passed on to the treating device subsystem for positioning the treating device. If there is no match, the control is passed on to the camera subsystem to capture the image again. Further the computing subsystem 207 comprises a target image of the target subject that is acquired during diagnosis process of the target subject.
The treating device subsystem 215 is configured to control the treating device 223 using the treating device controller 218 which is driven by extrinsic control parameter based on image processing activities at the computing subsystem 207. While controlling the treating device 223 upon precise positioning, the camera viewing conditions are maintained at fixed pre determined position as guided by control methods as indicated by signal 201 and 222.
FIG. 3 A illustrates two independent coordinate systems having (xc,yc,zc) axes for object coordinate system 303 and (xo,yo,zo ) axes for camera coordinate systems 302. The object coordinate system 303 and the camera coordinate system302are relatively located in world coordinate system 301 having (xw,yw,zw) axes. (Tow, Row, Sow ) which represents translation, rotation and scaling intrinsic parameters of the target object 114 in object coordinate system 303 with respect to world coordinate system 301. Similarly, (Tcw,Rcw,Scw ) represents translation, rotation and scaling intrinsic parameters in camera coordinate system 302 with respect to world coordinate system 301. Further, (Tco,Rco,Sco) represents translation, rotation and scaling intrinsic parameters between camera and object coordinate systems.
FIG.3B illustrates process of image formation in accordance with some embodiments of the present disclosure.FIG.3B shows a graph illustrating image capturing of a target object 101 having three different coordinate systems such as object coordinate system, camera coordinate system and world coordinate system. FIG.3B shows the captured image frame having camera field view 306, wherein image patch 307 corresponds to the target object 304, when optical axis of image sensor 305 is assumed to be in perpendicular to image plane in (x- y) space plane. Similarly, FIG.3C illustrates another embodiment of image capturing with a field of view of the camera which is EFGH 309, wherein image patch 308 corresponds to the target object 304 by having optical axis 310 orientation of the image sensor (also referred as camera sensor) 305 perpendicular to the image plane EFGH. The image plane EFGH has degrees of freedom for rotations around both x-axis and z-axis.
FIG.4 illustrates process of partitioning the image 309 captured in the field of view of the camera. The point S is identified as (m,n) in the captured image 309. The sub image 613, 614, 615 and 616are represented respectively as below:
Figure imgf000011_0001
In an embodiment, a window 330 is selected as shown in FIG.5A which is not in arbitrary way or a window is selected either from any corner of the image frame. Typically, any sliding window based object detection starts form left most corner of the image frame and sliding operation of the window. The initial coordinate point (m,n) of window sliding for scanning is selected on the basis of imaging axis which intersects the image plane shown as point S in FIG.3C, where centre point of window typical sizes are 5x5, 7x7, etc. Based on this, the image is split into four sub images. The splitting will yield in various images such as 313, 314, 315, 316 as shown in FIG.5B.
FIG.6 A illustrates a method for identifying coarse translated position of the image patch in the captured image frame 304 corresponding to the target object 308 [as shown in FIG.3C] The process involves exploitation of either sliding window based or as an example Viola and Jones technique which is suitable for a multicore computing process and fast computation. The process starts with identifying four partitioned sub-image in the manner described in FIG. 4 and described as image arrays in an appropriate way as mentioned above asQi,Qn,Qm and Win.
The present disclosure illustrates the process of splitting the image frame 601 into four sub images 602, 603, 604 and 605 with an initial point as (m,n), which is determined by the intersection of optical axis of camera and the image frame 601. The four integral process sums as a stream processes concurrently to exploit four core processing and facilitates in quick processing.
At block 605, the sub images are matched with template image 606 (stored image 108) stored with initial window selected image patch. The Viola and Jones object detection process 607 is implemented in four streams. If the match condition is identified, then the stream processing is stopped and then either of centre or centroid of the image patch is identified and the co-ordinates are stored appropriately as coarse translated position of the captured image at block 607. At block 608, the centre coordinates are converted to camera position coordinates.
FIG.6B illustrates process for tracking and detection of location of the object. At block 608, initially, camera position is controlled to be at some convenient starting location so that initial field of view is obtained and instant image capturing process 610 enables to acquire an image frame. At block 612, the captured image and the stored image 612 are used to find the coarse translation position of image patch corresponding to the target object using coarse translation matching process using the embodiment described in FIG.6 A.
At block 609 the camera is positioned to view the target object 101 by providing appropriate position coordinate to camera position and motion controller 105 and capture next instant image frame that assures to have small translation to the target image.
At block 613, the pixel position which is in Cartesian coordinates for the stored image is converted to spherical coordinates.
At block 614, scale and rotation parameters are identified using regression convolution neural network.
At block 615, if there is only change in the scale parameter, then the image is resized with suitable coordinates. If there is only change in angle parameter, then the image is rotated. However, if there is both scale change and angle change then the image is resized and rotated.
At block 616, the target image is matched with captured image.
At block 619, if the captured image matches with the target image then it is declared that the object is identified. Otherwise at block 618, first intrinsic parameters i.e. scale parameter and the angle parameter that is in spherical coordinates are converted into suitable extrinsic camera parameter and then camera is positioned to view suitably and to continue with image frame acquisition process.
FIG.7A(i) shows both stored image 701 and captured image 702. C_ref(xr,yr ) 704 and Cacq (x¾ya) 703, are centres or centroids, respectively. Furthermore, the sample points set shown
705 and 707 are obtained either by uniform or non-uniform sampling of the boundary curves
706 and 708, respectively for stored image 701 and captured image 702. FIG. 7A(ii) represents spherical coordinate points corresponding to FIG.7A(i).
FIG.7B(i) is replica of FIG.7A(i) with difference of images 701 and 702 overlapped partially.
FIG.7B(ii) represents spherical coordinate representation system 709corresponding to FIG.7B(ii). The straight lines 710 and 711 shown in FIG. 7B (ii) is regression line fit for the selected sample points represented in spherical coordinate system for stored image 70 land the captured image 702 respectively. FIG.7C (i) and FIG7C (ii) are replica of 7B (i) and 7B (ii), with a difference of regression ellipse fits 712 and 713 with respective major axes as 714 and 716. Additionally, 715 and 717 are respective minor axes. The parameter difference angle f 714 is obtained by using slopes parameters major axes.
Further, FIG.7D (i) and FIG.7D (ii) are replica of FIG.7C (i) and FIG.7C (ii), respectively with difference that regions of stored image is fully overlapped with captured image.
FIG. 7E shows geometric image parameters. FIG.7F is replica of FIG.7E with a difference of overlap regions of regression fit ellipses. FIG.7G shows estimation process for ellipse fit parameters.
FIG.8A, FIG.8B, FIG.8C illustrates the process of identification of scale parameter and rotation parameter for the target image and the stored image. The angle parameter is identified based on difference between angle of orientation of the target area in the captured image and angle of orientation of the target area in the stored image using at least one of major axis and the minor axis of an image plane associated with angle coordinate of the image sensor associated with the captured image and angle coordinate of the image sensor associated with the stored image. The scale parameter is identified based on difference between area of the stored image based on position parameter related to the stored image and area of the captured image based on position parameter related to the captured image. The angle parameter and the scale parameter are stored as first intrinsic parameters. As shown in FIG.8 A, block 802 and 805 illustrates steps involved in the process of identifying the scale parameter and the rotation parameter for the target image and the stored image respectively. At block 807, the compositional spatial transformation network is formulated based on the scale parameter and the rotation parameter/angle parameter using technique described in FIG.8B.At 808, the system provides the scaled and rotated target image patch using the compositional spatial transformation network.
As shown in FIG, 8B, at block 812, an image grid structure (series of intersection of horizontal and vertical lines, typically two dimensional positional point) is generated for image pixels in the image frame 810 by image grid generator 812. At block 814, a warped image grid structure is generated using the angle parameter cp. Then, a new image is composed at block 815 at each position of new grid structure obtained at 814 that serves as an input to the fast scaled convolution neural network to either up sample or down sample process at block 807 based on scaling parameters, s, as per FIG.8 A.
FIG.8C illustrates process of up scaling or downscaling of the image based on convolution neural network. The stored image is either downscaled or upscaled depending on value of scale parameter.
At block 812, scale, s is checked whether it is greater 1 or not. If s> 1, then flow computation is shifted to block 813, where image is set for down sampling with any down sample network to arrive at required image 811.If s<l, then flow is shifted block 815, where up sampling sub network is performed to arrive at required image 812.
FIG.9 shows a flowchart illustrating a method for identifying intrinsic image parameters required for rotation and scaling in accordance with some embodiments of the present disclosure.
At block 901, the method involves performing image capturing after positioning the camera view using camera position and motion controller 106 by utilizing intrinsic image parameters to generate the camera extrinsic parameters.
At block 904, the method involves buffering the coarse translated localized image.
At 908, the method comprises converting the captured image, which are typically in cartesian coordinates to spherical coordinate transformation.
At block 908, the method comprises performing regression network to get the regression parameters either for linear function or ellipse fit for the captured image.
On the other hand, at block 912, the method comprises performing conversion of cartesian coordinates to spherical image coordinates for stored image 903.
At block 913, the method comprises obtaining regression parameters using regression network on the spherical coordinates from block 912 either for either linear fit or ellipse fit.
At block 914, the method comprises utilizing the regression parameters obtained for stored image with corresponding similar regression parameters for the captured image. At block 915, the method comprises composing scaled and rotated template corresponding to stored image.
At block 906, the method comprises checking for match between the images using block 915 and the output of buffered image at 905. If there is no match, the processing is repeated by initiating the block 902, otherwise the method proceeds to block 918.
At block 918, controller of surgery process is initiated to control the treating device.
FIG.10 illustrates a process of acquiring image corresponding to the target image when the object undergoes unknown translated position in accordance with some embodiments of the present disclosure.
At block 1001, the camera is directed to set at an instant pose and acquire an image. At block 1003, the captured image 1002 is used to find the object location that provides translated position with respect to camera imaging focus axis. The captured image 1004 is extracted from the translated position and the image from the instant acquired position is provided for matching at block 1010.
At block 1008, the stored image 1006 (which is target image corresponding to tissue or organ acquired during diagnostic time) is used to find the rotation and scaling process.
At block 1010, the image and its intrinsic image parameters 1009 are matched with the captured image 1004.
When there is no match, the intrinsic image parameters are converted to extrinsic parameters using techniques which includes but not limited to robotic navigation algorithm. It indicates the repositioning of camera for next image acquisition to arrive at corrected precise image coordinates, for object location.
The process of instant image acquisition is guided by the extrinsic parametersl013 obtained at block 1012. The translation position parameters 1005 are obtained at block 1003. At block 1010, on matching condition the image (indirectly object) position and allied required parameter is shifted to control the treating device for performing surgical activities.
Advantages of the embodiment of the present disclosure are illustrated herein. In an embodiment, the present disclosure provides a method and system for accurate detection of position of target area in a target subject with minimal invasiveness.
In an embodiment, the method of present disclosure generates precise coordinates for the treating device using intrinsic and extrinsic parameters of the camera.
In an embodiment, the present disclosure uses target image captured during diagnosis of the subject and the target image features are used for detecting accurate position of the target area in real time.
In an embodiment, the present disclosure enables real time tracking and guiding of treating device to access precise position of target area in the target subject.
The terms "an embodiment", "embodiment", "embodiments", "the embodiment", "the embodiments", "one or more embodiments", "some embodiments", and "one embodiment" mean "one or more (but not all) embodiments of the invention(s)" unless expressly specified otherwise.
The terms "including", "comprising",“having” and variations thereof mean "including but not limited to", unless expressly specified otherwise. The enumerated listing of items does not imply that any or all the items are mutually exclusive, unless expressly specified otherwise.
The terms "a", "an" and "the" mean "one or more", unless expressly specified otherwise.
A description of an embodiment with several components in communication with each other does not imply that all such components are required. On the contrary, a variety of optional components are described to illustrate the wide variety of possible embodiments of the invention.
When a single device or article is described herein, it will be clear that more than one device/article (whether they cooperate) may be used in place of a single device/article. Similarly, where more than one device or article is described herein (whether they cooperate), it will be clear that a single device/article may be used in place of the more than one device or article or a different number of devices/articles may be used instead of the shown number of devices or programs. The functionality and/or the features of a device may be alternatively embodied by one or more other devices which are not explicitly described as having such functionality/features. Thus, other embodiments of the invention need not include the device itself.
Finally, the language used in the specification has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. It is therefore intended that the scope of the invention be limited not by this detailed description, but rather by any claims that issue on an application based here on. Accordingly, the embodiments of the present invention are intended to be illustrative, but not limiting, of the scope of the invention, which is set forth in the following claims.
While various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the following claims.
Referral Numerals:
Figure imgf000019_0001

Claims

CLAIMS:
1. A method of detecting position of a target area in a target subject, the method comprising:
capturing an image of the target area using an image sensor; converting, by an image analysis and processing system associated with the image sensor, the captured image from cartesian coordinate system to spherical coordinate system; obtaining, by the image analysis and processing system, regression parameters for captured image using a predefined regression technique on the spherical coordinate system; transforming, by the image analysis and processing system, the captured image based on the regression parameters; and detecting, by an image analysis and processing system, the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject.
2. The method as claimed in claim 1, wherein the regression parameters comprises scale parameter and rotation parameter and translation parameter.
3. The method as claimed in claim 1, wherein the transforming the captured image comprises:
identifying angle parameter based on difference between angle of orientation of the target area in the captured image and angle of orientation of the target area in the stored image using at least one of major axis and the minor axis of an image plane associated with angle coordinate of the image sensor associated with the captured image and angle coordinate of the image sensor associated with the stored image; identifying scale parameter based on difference between area of the stored image based on position parameter related to the stored image and area of the captured image based on position parameter related to the captured image; storing the angle parameter and the scale parameter as first intrinsic parameters; and performing at least one of resizing the captured image and rotating the captured image based on the first intrinsic parameters.
4. The method as claimed in claim 3, wherein the at least one of resizing the captured image and rotating the captured image involves upscale and downscale process performed using a predefined deep learning technique.
5. The method as claimed in claim 1, wherein translation point of the image is identified by performing: identifying translation position parameters of the captured image using a predefined technique; translating the stored image based on translation position parameters of the captured image; and detecting a match between the captured image and the stored image wherein upon detecting the match, the translation point of the image is identified as the second intrinsic parameter.
6. The method as claimed in claims 3 and 5, wherein the first intrinsic parameters and the second intrinsic parameter are converted into extrinsic parameters using predefined navigation techniques to reposition the image sensor.
7. The method as claimed in claim 6 further comprises providing the extrinsic parameters to an image controller associated with the image sensor for repositioning the image sensor, wherein the image sensor is repositioned to capture the image when there is a mismatch between the captured image and the stored image.
8. An image analysis and a processing system for detecting position of a target area in a target subject, the image analysis and processing system comprising: a processor; and a memory communicatively coupled to the processor, wherein the memory stores the processor-executable instructions, which, on execution, causes the processor to: receive an image of the target area captured using an image sensor associated with the image analysis and a processing system; convert the captured image from cartesian coordinate system to spherical coordinate system; obtain regression parameters for captured image using a predefined regression technique on the spherical coordinate system; transform the captured image based on the regression parameters; detect the position of the target area in the target subject upon identifying a match between a stored image and the transformed image, wherein the stored image is obtained during diagnosis of the subject; and an image position controller associated with the image sensor for repositioning the image sensor, wherein the image sensor is repositioned to capture the image when there is a mismatch between the captured image and the stored image.
9. The system as claimed in claim 8, wherein the processor provides position of the target area in the target subject to a treating device associated with the system.
10. The system as claimed in claim 8, wherein the regression parameters comprises scale parameter and rotation parameter and translation parameter.
11. The system as claimed in claim 8, wherein the processor transforms the captured image by performing steps of:
identifying angle parameter based on difference between angle of orientation of the target area in the captured image and angle of orientation of the target area in the stored image using at least one of major axis and the minor axis of an image plane associated with angle coordinate of the image sensor associated with the captured image and angle coordinate of the image sensor associated with the stored image; identifying scale parameter based on difference between area of the stored image based on position parameter related to the stored image and area of the captured image based on position parameter related to the captured image; storing the angle parameter and the scale parameter as first intrinsic parameters; and performing at least one of resizing the captured image and rotating the captured image based on the first intrinsic parameters.
12. The system as claimed in claim 11, wherein the processor performs at least one of resizing the captured image and rotating the captured image based on upscale and downscale process performed using a predefined deep learning technique.
13. The system as claimed in claim 8, wherein the processor identifies translation point of the image by performing: identifying translation position parameters of the captured image using a predefined technique; translating the stored image based on translation position parameters of the captured image; and detecting a match between the captured image and the stored image wherein upon detecting the match, the translation point of the image is identified as the second intrinsic parameter.
14. The system as claimed in claims 11 and 13, wherein the processor converts the first intrinsic parameters and the second intrinsic parameter into extrinsic parameters using predefined navigation techniques to reposition the image sensor.
15. The system as claimed in claim 14, wherein the processor provides the extrinsic parameters to an image position controller associated with the image sensor for repositioning the image sensor, wherein the image sensor is repositioned to capture the image when there is a mismatch between the captured image and the stored image.
PCT/IB2020/050006 2019-01-03 2020-01-02 Method and system for detecting position of a target area in a target subject WO2020141468A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IN201941000335 2019-01-03
IN201941000335 2019-01-03

Publications (1)

Publication Number Publication Date
WO2020141468A1 true WO2020141468A1 (en) 2020-07-09

Family

ID=71406600

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2020/050006 WO2020141468A1 (en) 2019-01-03 2020-01-02 Method and system for detecting position of a target area in a target subject

Country Status (1)

Country Link
WO (1) WO2020141468A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112232132A (en) * 2020-09-18 2021-01-15 北京理工大学 Target identification and positioning method fusing navigation information
CN112861952A (en) * 2021-01-29 2021-05-28 云南电网有限责任公司电力科学研究院 Partial discharge image matching deep learning method
CN114581867A (en) * 2022-03-01 2022-06-03 亿咖通(湖北)技术有限公司 Object detection method, device, storage medium and program product
US11631219B2 (en) * 2019-03-21 2023-04-18 Squaremind Method for generating a three-dimensional working surface of a human body, system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015192117A1 (en) * 2014-06-14 2015-12-17 Magic Leap, Inc. Methods and systems for creating virtual and augmented reality
EP3373834A1 (en) * 2015-11-12 2018-09-19 Intuitive Surgical Operations Inc. Surgical system with training or assist functions

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015192117A1 (en) * 2014-06-14 2015-12-17 Magic Leap, Inc. Methods and systems for creating virtual and augmented reality
EP3373834A1 (en) * 2015-11-12 2018-09-19 Intuitive Surgical Operations Inc. Surgical system with training or assist functions

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11631219B2 (en) * 2019-03-21 2023-04-18 Squaremind Method for generating a three-dimensional working surface of a human body, system
CN112232132A (en) * 2020-09-18 2021-01-15 北京理工大学 Target identification and positioning method fusing navigation information
CN112861952A (en) * 2021-01-29 2021-05-28 云南电网有限责任公司电力科学研究院 Partial discharge image matching deep learning method
CN112861952B (en) * 2021-01-29 2023-04-28 云南电网有限责任公司电力科学研究院 Partial discharge image matching deep learning method
CN114581867A (en) * 2022-03-01 2022-06-03 亿咖通(湖北)技术有限公司 Object detection method, device, storage medium and program product
CN114581867B (en) * 2022-03-01 2024-05-14 亿咖通(湖北)技术有限公司 Object detection method, device, storage medium, and program product

Similar Documents

Publication Publication Date Title
WO2020141468A1 (en) Method and system for detecting position of a target area in a target subject
CN109977770B (en) Automatic tracking shooting method, device, system and storage medium
US8724865B2 (en) Method, computer software, and system for tracking, stabilizing, and reporting motion between vertebrae
JP5538160B2 (en) Pupil detection device and pupil detection method
KR100292564B1 (en) Position detection system and method
US10430650B2 (en) Image processing system
JP2009169934A (en) System and method for recognizing deformable object
US6751361B1 (en) Method and apparatus for performing fixturing in a machine vision system
US10779793B1 (en) X-ray detector pose estimation in medical imaging
WO2017109854A1 (en) Learning image automatic sorting device, learning image automatic sorting method, and learning image automatic sorting program
CN110288661A (en) Location regulation method, device, computer equipment and the storage medium of operating lamp
CN110742631A (en) Imaging method and device for medical image
JP2023505749A (en) Apparatus for defining motion sequences in a generic model
JP2012185752A (en) Robot device, position/attitude detecting device, position/attitude detecting program, and position/attitude detecting method
US7400760B2 (en) Image processing apparatus
JP6452236B2 (en) Eyeball identification device and eyeball identification method
US9582892B2 (en) Radiation imaging apparatus, radiation imaging method, and program
JP6288770B2 (en) Face detection method, face detection system, and face detection program
JP7498404B2 (en) Apparatus, method and program for estimating three-dimensional posture of subject
CN104318580B (en) Pattern search method and device
CN112584041A (en) Image identification dynamic deviation rectifying method
JPH0896132A (en) Image processor and processing method
Freedman et al. A subset approach to contour tracking in clutter
JP2005071125A (en) Object detector, object detection method, object data selection program and object position detection program
US20240331446A1 (en) Systems and methods for automatic hand gesture recognition

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20736150

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20736150

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20736150

Country of ref document: EP

Kind code of ref document: A1