DESC:FORM 2
The Patent Act 1970
(39 of 1970)
&
The Patent Rules, 2005

COMPLETE SPECIFICATION
(SEE SECTION 10 AND RULE 13)

TITLE OF THE INVENTION

“Systems and methods for determining a path through the colon in virtual colonography”

APPLICANTS:
Name Nationality Address
SAMSUNG R&D Institute India - Bangalore Private Limited India # 2870, Orion Building, Bagmane Constellation Business Park, Outer Ring Road, Doddanekundi Circle, Marathahalli Post, Bangalore-560037, India

The following specification particularly describes and ascertains the nature of this invention and the manner in which it is to be performed:-

FIELD OF INVENTION
Embodiments herein relate to the field of medical diagnostics and more particularly to medical diagnostics for performing virtual colonoscopy.

BACKGROUND OF INVENTION
In virtual colonography, a centerline is crucial for endoluminal navigation from the rectum to the cecum, as the clinical need is a smooth centered path from the rectum to the cecum, for interactive navigation along the colonic lumen. The primary challenge is breakages or blockages in the colon, due to fecal residue, abnormalities, poor insufflation and inadequate electronic cleansing. Insufficient distention or insufflation with CO2 can result in partially or completely collapsed colon regions, leading to a scanned image with a fragmented colon. In statistics gathered from 50 subjects, it has been found that a segmented colon has 1 to 10 fragments with an average of 2.4 and also that 40% of volumes have collapsed colon segments in a clinical database. A secondary challenge is the presence of extra-colonic (air filled) regions such as the stomach, small intestine, appendix and lungs that must not be included as part of the centerline.
Several approaches have been proposed to address segmentation of a fragmented colon. These include anatomy based colon segmentation algorithms. A current solution proposes a centerline chaining based colon segmentation method with a plurality of heuristic rules and parameters; for example, when to stop region growing, number of seed points and how many slides apart to look for the next seed. However the essential drawbacks of rule-based methods still hold good for this solution and this solution is only valid for well-distended or slightly collapsed colon cases.
Other solutions perform an automatic and exhaustive search on seed allocation based on 2-D image analysis. Their seed placement method looks at various criteria, which includes size and shape filters. Their underlying assumption is that any isolated colon fragment would have at least one segment that is well distended, and seeds are identified in the intersections of slices and well- distended colon areas. Their shape filtering looks at various measures (size, principal diameters with different metrics for the small bowel, rectum).
Another solution proposes a hybrid segmentation of the colon tissue, based on region growing, extraction of an air and opacified fluid pocket tree, followed by fuzzy connectivity, hole filling and level-set guided segmentation. Another approach adopts a machine learning approach to separate colon segments from small intestine, stomach and other extra-colonic parts using a binary classifier followed by sequentially daisy chaining components between the rectum and cecum. Another approach uses a graph inference scheme for colon segmentation linking using semantic information derived from a multi-class classifier.
Several approaches have been used for centerline extraction: topology preserving thinning, a modified Dijkstra approach with relocation of the centerline points based on distance to boundary and so on. Level set methods have been used to extract the minimal path from a source to target based on fast marching using the distance map and gradient descent on the arrival map. These have been modified to detect and correct loops, resulting from two segments of the colonic wall touching each other (due to inaccurate segmentation and/or fully distended colon segments) resulting in a loop in the centerline. Another approach uses two geodesic active contour level set formulations, one for the outer and one for the inner colon wall to account for loops and small under-distentions, however its computation times run into minutes and are too slow to be clinically usable.

OBJECT OF INVENTION
The principal object of the embodiments herein is to disclose methods and systems for providing a path from the rectum to the cecum in the colon, for interactive navigation along the colonic lumen for use in virtual colonoscopy, when the colon is fragmented and has undergone segmentation.
Another object of the embodiments herein is to disclose methods and systems for providing a path from the rectum to the cecum in the colon, for interactive navigation along the colonic lumen for use in virtual colonoscopy, by providing an energy minimizing geodesic that uses a set of strong/weak candidate segmentations, possibly fragmented, to extract the energy minimizing geodesic path, when the colon is fragmented and has undergone segmentation.

SUMMARY
Accordingly the invention provides a method for determining a path through a colon, the method comprising performing cleansing of scanned data to remove non-relevant regions by a processing engine; localizing points in the colon in the cleansed scanned data by the processing engine, wherein the localized points comprise a rectum and a cecum; creating a segmentation of the colon by initiating a region growing from the localized points by the processing engine; and extracting a path through the localized points by the processing engine, wherein the path punches through breakages in the segmentation and the path traverses from the rectum to the cecum.
Accordingly the invention provides an electronic device for determining a path through a colon, the electronic device configured to perform cleansing of scanned data to remove non-relevant regions; localize points in the colon in the cleansed scanned data, wherein the localized points comprise a rectum and a cecum; create a segmentation of the colon by initiating a region growing from the localized points; and extract a path through the localized points, wherein the path punches through breakages in the segmentation and the path traverses from the rectum to the cecum.
These and other aspects of the embodiments herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions, while indicating preferred embodiments and numerous specific details thereof, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the embodiments herein without departing from the spirit thereof, and the embodiments herein include all such modifications.

BRIEF DESCRIPTION OF FIGURES
This invention is illustrated in the accompanying drawings, throughout which like reference letters indicate corresponding parts in the various figures. The embodiments herein will be better understood from the following description with reference to the drawings, in which:
FIG. 1 is a flowchart depicting the process of providing a path from the rectum to the cecum in the colon, according to embodiments as disclosed herein;
FIG. 2 depicts a system for scanning a colon of a user, processing the scanned data and storing the processed and scanned data, according to embodiments as disclosed herein;
FIG. 3 illustrates a plurality of components of the processing engine, according to embodiments as disclosed herein;
Fig 4 depicts regions that are included in the colon segmentation via daisy chaining for two datasets, according to embodiments as disclosed herein;
FIG. 5 depicts the process of path extraction from cecum to sphlenic flexure, according to embodiments as disclosed herein;
FIG. 6 depicts the process of path extraction from rectum to sphlenic flexure, according to embodiments as disclosed herein;
FIG. 7 shows the evolution of the front, on the same dataset, again from the rectum to the sphlenic flexure, according to embodiments as disclosed herein;
FIG. 8 depicts the dataset from the ACRIN CTC trial with a 55mm tumor in the rectum, effectively separating the rectum from the sigmoid and its extracted centerline, according to embodiments as disclosed herein;
FIG. 9 illustrates the effect of the geodesic repulsion field, according to embodiments as disclosed herein;
FIG. 10 illustrates the execution of front propagation(s) between each pair of consecutive points on the Speed Image, according to embodiments as disclosed herein; and
FIG. 11 illustrates collect centerline components from each marching joining together to get the complete centerline, according to embodiments as disclosed herein.

DETAILED DESCRIPTION OF INVENTION
The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein can be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.
The embodiments herein to disclose methods and systems for providing a path from the rectum to the cecum in the colon, for interactive navigation along the colonic lumen for use in virtual colonoscopy, when the colon is fragmented and has undergone segmentation. Referring now to the drawings, and more particularly to FIGS. 1 through 11, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments.
Embodiments herein extract centerlines through a set of points providing a path from the rectum to the cecum in the colon from segmental end points, using a cost function for handling breakages in the colon and a cost function enforcing order.
FIG. 1 is a flowchart depicting the process of providing a path from the rectum to the cecum in the colon. At step 101, electronic cleansing is performed of scanned data, wherein outer-air, the scanning table, bones, lungs, fecal matter, and other unrelated tissue and material are located and discarded using mathematical morphology based methods that take into account intensity, location, connectivity and size measures. At step 102, four points are automatically localized: the rectum, the cecum, the sphlenic and the hepatic flexures. The rectum can be localized using metrics such as location, size, intensity and connectivity to the outer body. The cecum is typically the fattest segment of the colon. The sphlenic and hepatic flexures can be localized based on their proximity R, L edges of the scan and the lung. At step 103, an initial segmentation is generated by region growing and heuristic approaches. At step 104, colon blobs that may have been missed as weaker candidate segmentations are linked using a daisy chaining procedure. At step 105, a minimal energy path through the ordered set of points is extracted using a front propagation. This propagation can be guided by forces such as a force given by the distance to the colon segmentation surface, a force derived from the intensity, a force from weaker candidate colon segmentations, and a geodesic repulsive force, where the other points exhibit an repelling force in their voronoi partition, the force proportional to the geodesic distance to the point. At step 106, a path extraction method is provided for the colon that is the energy minimized geodesic favouring centeredness, punching through gaps, traversing in so far as possible through lower intensity regions and ordered through the localized points. The various actions in method 100 may be performed in the order presented, in a different order or simultaneously. Further, in some embodiments, some actions listed in FIG. 1 may be omitted.
Given an ordered set of points, and a crude segmentation(s), embodiments herein find the "best" path that passes through these points, in so far as possible centered within the segmented chunks. For each consecutive pair in the ordered list, embodiments herein compute a minimal path. This minimal path is obtained by creating three fields: a strong potential field with the potential varying through the segmentation and given by the distance to the boundary, a weak potential field with the potential directly derived from the intensity and a repulsion field, where the other points exhibit an repelling polarity to the oncoming path and repel any advancement of the path in their geodesic vicinity. Each of the points in the set is given their own "zone of influence". The zones of influence form a voronoi partition of the segmentation given the set of points; cutoff distances being geodesic (any voxel in the segmentation belongs to the zone of influence of the point to whom it is geodesically closest; through the segmentation). The repulsion at any voxel within an opposing zone of influence is proportional to the geodesic distance from the epicenter of its zone of influence.
Embodiments herein strives to minimize the arc length of the path at the same time ensuring centeredness of the path, punching through gaps, traversing in so far as possible through lower intensity regions and ensuring that the path does not traverse enroute one of the other stops.
Embodiments herein enable extraction of a medial line from an ordered set of points ensuring Homotopic (topology preserving), centeredness, and smoothness (an absolute centeredness would cause the medial line to be sensitive to noise; i.e., boundary perturbations).
FIG. 2 depicts a system for scanning a colon of a user, processing the scanned data and storing the processed and scanned data. The system, as depicted comprises of a scanner 201, a processing engine 202 and a Picture Archiving and Communication System (PACS) 203. The scanner 201 can use at least one of a CT (X-ray computed tomography (X-ray CT)/ computerized axial tomography scan (CAT scan)) or a MRI (Magnetic Resonance Imaging) for scanning the abdominal/colon region of a patient. Data from the scanning done by the scanner 201 can be provided to the processing engine 202. On receiving the scanned data, the processing engine 202 can perform electronic cleansing, wherein outer-air, the scanning table, bones, lungs, fecal matter, and other unrelated tissue and material are located and discarded using mathematical morphology based methods that take into account intensity, location, connectivity and size measures. After electronically cleansing the data, the processing engine 202 can localize four points automatically: the rectum, the cecum, the sphlenic and the hepatic flexures. The processing engine 202 can localize the rectum using metrics such as location, size, intensity and connectivity to the outer body. The processing engine 202 can localize the cecum, as the cecum is typically the fattest segment of the colon. The processing engine 202 can localize the sphlenic and hepatic flexures based on their proximity R, L edges of the scan and the lung. The processing engine 202 can then generate an initial segmentation by region growing and heuristic approaches. After generating the initial segmentation, the processing engine 202 can link colon blobs that may have been missed as weaker candidate segmentations using a daisy chaining procedure. After segmentation, the processing engine 202 can extract a minimal energy path through the ordered set of points using a front propagation. The processing engine 202 can guide the propagation by forces such as a force given by the distance to the colon segmentation surface, a force derived from the CT intensity, a force from weaker candidate colon segmentations, and a geodesic repulsive force, where the other points exhibit an repelling force in their voronoi partition, the force proportional to the geodesic distance to the point. The processing engine 202 can extract a path through the colon using a path extraction method that is the energy minimized geodesic favouring centeredness, punching through gaps, traversing in so far as possible through lower intensity regions and ordered through the localized points. The processing engine 202 can then store the scanned and processed data in the PACS 203, wherein the processed data comprises of the scan of the colon with the extracted path superimposed on the scan of the colon.
In an embodiment herein, the processing engine 202 can store the scanned and processes data in another location, than the PACS 203, such as local memory, the cloud, a file server, a data server, and so on.
In an embodiment herein, the processing engine 202 can enable a user to access the scanned and processed data directly from the processing engine 202.
FIG. 3 illustrates a plurality of components of the processing engine. Referring to FIG 3, the processing engine 202 is illustrated in accordance with an embodiment of the present subject matter. In an embodiment, the processing engine 202 may include at least one processor 301, an input/output (I/O) interface 302 (herein a configurable user interface), and a memory 303. The at least one processor 301 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the at least one processor 301 is configured to fetch and execute computer-readable instructions stored in the memory 303.
The I/O interface 302 may include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface such as a display screen, and the like. The I/O interface 302 may allow the processing engine 202 to communicate with other devices, such as the scanner 201, the PACS 203, and so on. The I/O interface 302 may facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example, Local Area network (LAN), cable, etc., and wireless networks, such as Wireless LAN, cellular, Device to Device (D2D) communication network, Wi-Fi networks and so on. The modules 304 include routines, programs, objects, components, data structures, and so on, which perform particular tasks, functions or implement particular abstract data types. In one implementation, the modules 304 may include a device operation module 305. The device operation module 305 can be configured to handle one or more tasks. The device operation module 305 can comprise of a cleansing engine 306, a localization engine 307, a segmentation engine 308, and a path extraction engine 309.
The cleansing engine 306 can discard the outer-body-air and lungs region growing from the faces of the bounding box of the dataset to extract air-filled connected regions. The cleansing engine 306 can locate and discard non-relevant regions such as outer-air, the scanning table, bones, lungs, fecal matter, and other unrelated tissue and material using mathematical morphology based methods that take into account intensity, location, connectivity and size measures. The cleansing engine 306 can use a suitable method such as sparse Expectation Maximization method.
Subsequently, the localization engine 307 can localize and detect the rectum (a first end point), using location, size, intensity and proximity to the outer body. Specifically, the localization engine 307 can traverse the most inferior axial slice upwards, examining air filled connected components that lie within the body and are roughly centered with reference to the body medial axis, with a constraint on the minimum area, accumulated volume and its sphericity. Then the localization engine 307 can locate the cecum (a second endpoint) by locating the thickest air filled sac below the lung on the right third of the patient. Then the localization engine 307 can propagate an isotropic front from this point towards its inferior and right. The localization engine 307 can use a constraint on the distance to the boundary of 5mm, so as to not traverse into the appendix. The localization engine 307 can assume the farthest point on the front to be the cecum. If the rectum and the cecum do not belong to the same connected chunk, the localization engine 307 can extract the sphlenic and hepatic flexures. The localization engine 307 can localize these flexures based on their proximity to the R, L edges of the scan and the lung by choosing the points with a sufficiently large thickness as determined from a distance map. The localization engine 307 can locally center the localized points (the rectum, and the cecum and in some cases, the sphlenic and hepatic flexures).
The segmentation engine 308 can create an initial segmentation of the colon by initiating a region growing with a suitable threshold, (for example, -750HU) from the localized points provided by the localization engine (which can be 4 or 2). The segmentation engine 308 can daisy chain large blobs, which may have been part of the colon segmentation, but are not included in the segmentation, using an outwardness measure. The segmentation engine 308 can consider connected components within the body (discounting the lung) with volume > 15mm3 and with a thickness > 6mm and elongation > 3 (computed as ratio of the major axis length to the diameter of a sphere representative of the candidate volume) as candidates. For these candidates, the segmentation engine 308 can perform voting to determine the outwardness of the connected components, on the coronal (XZ) plane. The segmentation engine 308 can project the colon segmentation and candidates onto the XZ plane. At each projected voxel, a vector value is present, that is indicative of the number of voxels from the current segmentation and from each of the candidates at that (x,z) coordinate. The segmentation engine 308 can fire rays from each of the candidates to the closest outward faces of the volume. If a sufficient number of rays intersect the current segmentation or other candidate components, the segmentation engine 308 can deem it as not outward (i.e., inward to those intersecting segmentation/candidates) and discarded. This ensures that outward disconnected fragments on the coronal projection are retained (typically colon in an abdominal scan) and inward fragments (extracolonic regions such as small bowel) are discarded. The segmentation engine 308 can use a pre-defined vote threshold (i.e., no more than the pre-defined vote threshold in percentage of the outward facing rays must intersect the current colon or other segmentations). For efficiency, the segmentation engine 308 can restrict the number of rays cast on each candidate to a predefined limit, by sampling the candidate segmentations randomly for points up to the pre-defined limit. Fig 4 depicts examples of regions that are included in the colon segmentation using daisy chaining for two datasets in a darker shade (wherein a lighter shade depicts segments linked by the daisy chaining procedure to the initial colon segmentation, shown in a darker shade). Note that the intent of this process is to capture large disconnected chunks in the colon; and not to capture smaller isolated chunks. Note also that these are chunks that are not connected (via simple thresholding on the cleansed volume) to the detected rectum, cecum, hepatic or sphlenic flexures.
The path extraction engine 309 can extract a medial line through the localized points, wherein the medial line traces out a path through the colon. The path extraction engine 309 can follow the following criteria: the path should be centered (to the extent permitted; the path must be able to punch through breakages in the colon segmentation (the jump should be preferred through air filled ?gaps and should be centered through these gaps); the path must traverse from the rectum to the ?sphlenic to the hepatic flexure and to the cecum and C1 continuity for smooth endoluminal navigation along the centerline using a virtual camera. ?
The minimal energy path, ? from a point p1 to p2 along its parametric length, s, can be written as ?
e(?)= ?¦??(?(s))ds? 0=s=1 (1)
where ?(x)>0 is the cost function.
The path extraction engine 309 can obtain a solution of the minimization of the energy, e through the computation of the geodesic distance, ?. The path extraction engine 309 can regard the values of ? as the arrival times of a front propagating from the source p1 with velocity ?F = 1/ ?. The function ? satisfies the Eikonal equation ?
||??(x) ||F = 1 ?(?_1 )= 0 (2)
? has only one global minimum, the source point p1, and its flow lines satisfy the Euler-Lagrange equation of functional eqn (1). The path extraction engine 309 can retrieve the minimal energy path, C_(p1,p2), with a gradient descent on ? from ?_2 to ?_1 by solving the following differential equation using a suitable standard numerical method such as Runge-Kutta. ?
?dC?_(?_1,?_2 )/ds (s) ? - ??(C_(?_1,?_2 ) (s)) (3)
with C_(?_1,?_2 ) (0)= ?_1 and C_(?_1,?_2 ) (1)= ?_2
The path extraction engine 309 can incorporate an additive term guided by intensity, to enable the path to transcend gaps, in so far as possible through airish regions. To ensure centeredness of the path, in those colonic segments that have not have been included in the segmentation, the path extraction engine 309 can incorporate the distance to the remaining air sacs into the velocity. The modified velocity at a location x can be computed as:
F(x)= aD_s (x)+ ßD_a (x)+ µS(I(x) ) (4)
where D_s is the euclidean distance to the colon segmentation boundary, D_a is the euclidean distance to the boundaries of the remaining air filled regions (discounting the outer body and the lungs as described earlier); I(x) is the CT intensity at x and S is a sigmoid function that prefers lower intensities. The weight a is at least an order of magnitude greater than ß and µ, to ensure that the propagation is primarily guided by the distance to the segmented colon surface.
Fig. 6 shows front evolution from the cecum to the sphlenic flexure. The front is guided by distance to the boundary as well as the intensity as in eqn (4). Note the jump across the gap from (d) to (e). The path extraction engine 309 can extract the centerline successfully.
Unfortunately, with this formulation, in datasets with larger disconnections, the path occasionally jumps directly from the rectum (through the pertonium and appendix) into the cecum due to its close spatial proximity to the rectum (as depicted in FIG. 7). To avoid this, the path extraction engine 309 can perform a multiple segmental reconstruction (rectum to sphlenic flexure, to hepatic flexture, to cecum). FIG. 3 shows the front traversing from the rectum to the sphlenic flexure via the cecum. The front reaches the cecum in FIG. 3(c). The weight ß could be increased, but note in this dataset that a front that reaches the cecum will travel faster than it would through the narrow sigmoid. Also, ß needs to kept low so as to not impact centeredness of the path.
To address this, the path extraction engine 309 can increase a repulsion field, applicable in the context of a segmental centerline reconstruction where an opposing node can repel an oncoming front. Consider a set of nodes T = {T1 , ., Tn }. During front propagation from Ti to Ti+1, all other nodes in {T } (except i, i + 1), exhibit a repelling force. This force by a node, Tq, for a point in its geodesic vicinity is proportional to the geodesic distance of the point, x to Tq. The path extraction engine 309 can compute the geodesic vicinities or zones of a node using a simple isotropic integer front propagation from all nodes simultaneously on the binary colon segmentation until all pixels in the segmentation are visited. This path extraction engine 309 can perform the front propagation in a single pass, with initial seed values at the nodes. In an embodiment herein, consider that there are four nodes, the initial seed values can be of 10000, 20000, 30000 and 40000 at each of the nodes. The most significant digit therefore serves as the zone identifier. The zones of influence (geodesic vicinities) form a voronoi partition of the segmentation with partitions centered at the nodes; cutoff distances being geodesic. The repulsion at any voxel within an opposing zone of influence is proportional to the geodesic distance from its epicenter. Consequently, when the path extraction engine 309 is extracting the path from the rectum to the sphlenic flexure, the hepatic and cecum zones repel the front, but the repulsion is absent in the voronoi partition that contains the rectum and the sphlenic flexure. The velocity to extract a path from Tj to Tj+1 is:
F(x)= aD_s (x)+ ßD_a (x)+ µS(I(x) )- ? ?_¦(i?j@?J+1)^n¦D_s^g (x,T_i) (5)
where Ds (x, Ti ) is the geodesic distance of x to the node Ti, when x is inside its voronoi partition and is zero when outside. F(x) is clamped to 0 to ensure that the velocity is a non-negative function. FIG. 8 shows the evolution of the front, on the same dataset, again from the rectum to the sphlenic flexure using the velocity term described above. The path is now correctly extracted. Comparing the front advancement in FIG. 8d-e with that in FIG. 7c-d, it can be seen that the cecum repels advances of the front. FIG. 7 also shows centerlines extracted on two other fragmented datasets. The modules 304 may include programs or coded instructions that supplement applications and functions of the processing engine 202. The data 310, amongst other things, serves as a repository for storing data processed, received, and generated by one or more of the modules 304. Further, the names of the other components and modules of the processing engine 202 are illustrative and need not be construed as a limitation.
FIG. 8 depicts the dataset from the ACRIN (American College of Radiology Imaging Network (ACRIN)) CTC trial with a 55mm tumor in the rectum, effectively separating the rectum from the sigmoid and its extracted centerline. The ACRIN database contains several datasets with poor bowel preparation. Embodiments herein performed a visual evaluation of the centerlines in 57 datasets. Of these, 49 centerlines were completely and correctly extracted. The rectum was detected success- fully in all scans. The multi-segment centerline extraction is parallelizable. The process takes < 30s, with cleansing taking 15s, colon segmentation and centerline extraction taking 11s on an Intel Xeon dual core 3GHz with 8GB RAM.
FIG. 9 illustrates the effect of the geodesic repulsion field, according to embodiments as disclosed herein. The effect of the geodesic repulsion field is evidenced here. During the extraction of the path from the rectum to the sphlenic flexure, the cecum repels. It can be seen that the front traverses much faster from the rectum to the sphlenic flexure. In spite of the front having jumped into the cecum in (b), which is significantly thicker, the front does not manage to get very far down that road. Without such a repulsion force, the front would have reached the sphlenic flexure through this incorrect path. FIG. 10 illustrates the execution of front propagation(s) between each pair of consecutive points on the speed image. FIG. 11 illustrates collect centerline components from each marching joining together to get the complete centerline, according to embodiments as disclosed.
The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the spirit and scope of the embodiments as described herein.

CLAIMS
We claim:
A method for determining a path through a colon, the method comprising
performing cleansing of scanned data to remove non-relevant regions by a processing engine;
localizing points in the colon in the cleansed scanned data by the processing engine, wherein the localized points comprise a rectum and a cecum;
creating a segmentation of the colon by initiating a region growing from the localized points by the processing engine; and
extracting a path through the localized points by the processing engine, wherein the path punches through breakages in the segmentation and the path traverses from the rectum to the cecum.
The method, as claimed in claim 1, wherein the localized points further comprise of sphlenic and hepatic flexures.
The method, as claimed in claim 1, wherein creating the segmentation of the colon comprises
creating an initial segmentation by the processing engine; and
creating a segmentation by linking blobs in the colon by the processing engine, wherein the blobs have been missed in the initial segmentation.
The method, as claimed in claim 1, wherein the processing engine extracts the path using a front propagation, wherein the front propagation uses a force given by distance to surface of a colon segmentation, a force derived from intensity of scanning, a force from weaker candidate colon segmentations, and a geodesic repulsive force, where other points exhibit an repelling force in voronoi partition, the force proportional to geodesic distance to the point.
The method, as claimed in claim 1, wherein the extracted path is a central path through the colon.
An electronic device for determining a path through a colon, the electronic device configured to
perform cleansing of scanned data to remove non-relevant regions;
localize points in the colon in the cleansed scanned data, wherein the localized points comprise a rectum and a cecum;
create a segmentation of the colon by initiating a region growing from the localized points; and
extract a path through the localized points, wherein the path punches through breakages in the segmentation and the path traverses from the rectum to the cecum.
The electronic device, as claimed in claim 6, wherein the localized points comprise of sphlenic and hepatic flexures.
The electronic device, as claimed in claim 6, wherein the electronic device is configured for creating the segmentation of the colon by
creating an initial segmentation; and
creating a segmentation by linking blobs in the colon, wherein the blobs have been missed in the initial segmentation.
The electronic device, as claimed in claim 6, wherein the electronic device is configured to extract the path using a front propagation, wherein the front propagation uses a force given by distance to surface of a colon segmentation, a force derived from intensity of scanning, a force from weaker candidate colon segmentations, and a geodesic repulsive force, where other points exhibit an repelling force in voronoi partition, the force proportional to geodesic distance to the point.

The electronic device, as claimed in claim 6, wherein the extracted path is a central path through the colon.

Dated this 08th February 2016 Signature:
Name of the Signatory: Kalyan Chakravarthy

ABSTRACT
Systems and methods for determining a path through the colon in virtual colonography. Embodiments herein relate to the field of medical diagnostics and more particularly to medical diagnostics for performing virtual colonoscopy. Embodiments herein provide methods and systems for providing a path from the rectum to the cecum in the colon, for interactive navigation along the colonic lumen for use in virtual colonoscopy, when the colon is fragmented and has undergone segmentation

FIG. 1
,CLAIMS:CLAIMS
We claim:
1. A method for determining a path through a colon, the method comprising
performing cleansing of scanned data to remove non-relevant regions by a processing engine;
localizing points in the colon in the cleansed scanned data by the processing engine, wherein the localized points comprise a rectum and a cecum;
creating a segmentation of the colon by initiating a region growing from the localized points by the processing engine; and
extracting a path through the localized points by the processing engine, wherein the path punches through breakages in the segmentation and the path traverses from the rectum to the cecum.
2. The method, as claimed in claim 1, wherein the localized points further comprise of sphlenic and hepatic flexures.
3. The method, as claimed in claim 1, wherein creating the segmentation of the colon comprises
creating an initial segmentation by the processing engine; and
creating a segmentation by linking blobs in the colon by the processing engine, wherein the blobs have been missed in the initial segmentation.
4. The method, as claimed in claim 1, wherein the processing engine extracts the path using a front propagation, wherein the front propagation uses a force given by distance to surface of a colon segmentation, a force derived from intensity of scanning, a force from weaker candidate colon segmentations, and a geodesic repulsive force, where other points exhibit an repelling force in voronoi partition, the force proportional to geodesic distance to the point.
5. The method, as claimed in claim 1, wherein the extracted path is a central path through the colon.
6. An electronic device for determining a path through a colon, the electronic device configured to
perform cleansing of scanned data to remove non-relevant regions;
localize points in the colon in the cleansed scanned data, wherein the localized points comprise a rectum and a cecum;
create a segmentation of the colon by initiating a region growing from the localized points; and
extract a path through the localized points, wherein the path punches through breakages in the segmentation and the path traverses from the rectum to the cecum.
7. The electronic device, as claimed in claim 6, wherein the localized points comprise of sphlenic and hepatic flexures.
8. The electronic device, as claimed in claim 6, wherein the electronic device is configured for creating the segmentation of the colon by
creating an initial segmentation; and
creating a segmentation by linking blobs in the colon, wherein the blobs have been missed in the initial segmentation.
9. The electronic device, as claimed in claim 6, wherein the electronic device is configured to extract the path using a front propagation, wherein the front propagation uses a force given by distance to surface of a colon segmentation, a force derived from intensity of scanning, a force from weaker candidate colon segmentations, and a geodesic repulsive force, where other points exhibit an repelling force in voronoi partition, the force proportional to geodesic distance to the point.

10. The electronic device, as claimed in claim 6, wherein the extracted path is a central path through the colon.