Method for generating quantitative data from medical imaging data

Abstract

A computer implementation method for generating quantitative parameters of pancreatic and biliary ducts and lesions of visceral organs from volumetric medical imaging scan data in which the scan shows all or part of the visceral organs is described. The method includes the steps of: identifying the location of at least one lesion within the visceral organ on the medical imaging scan data; segmenting at least one detected lesion and determining one or more quantitative structural parameters for at least one segmented lesion; performing volume segmentation of the medical imaging scan data to identify tubular tubes of the visceral organs within the volume of the medical imaging scan; calculating at least one set of quantitative structural parameters for at least one location along the length of the tubular tube for at least one segmented tubular tube within the volume of the medical imaging scan data; and generating at least one set of quantitative structural parameters for at least one segmented tubular tube. Apparatus for performing the method is also described.

Description

Technical Field

[0001] The present invention relates to the analysis of medical scan imaging data. In particular, it relates to the evaluation of cystic lesions within visceral organs and pancreaticobiliary ducts by analyzing medical scan imaging data.

Background Art

[0002] Cystic lesions, particularly pancreatic cystic lesions, are a common incidental finding during abdominal imaging examinations. Most cystic lesions remain asymptomatic and benign throughout the patient's lifetime. However, a small portion of the lesions become malignant and progress to pancreatic cancer. Considering that pancreatic cancer is usually detected only at a very late stage with a very poor prognosis, the ability to monitor incidentally detected cystic lesions provides an opportunity to diagnose pancreatic cancer at an early stage or ideally before the lesion becomes malignant. This early diagnosis may open the way to radical surgical treatment that is impossible at a later stage of the disease. The remaining challenge is to distinguish cystic lesions that require this intervention from benign lesions.

[0003] Accurately differentiating between benign and malignant pancreatic lesions is particularly important due to the abundance of detected lesions. More than 2% of the population and more than 20% of the population over 70 years old have asymptomatic pancreatic cystic lesions, and in the United States, more than 800,000 cystic lesions are incidentally detected every year. As mentioned above, these lesions can serve as precursors to pancreatic cancer and therefore require monitoring. However, since most are completely benign, it is important to identify rare high-risk lesions. The confusion surrounding pancreatic cystic lesions leads to over-treatment and monitoring, costing an estimated $5.3 billion per year in the United States.

[0004] Because pancreatic cystic lesions exhibit good image contrast on T2-weighted magnetic resonance imaging (MRI), current monitoring programs rely on non-invasive MRI using magnetic resonance cholangiopancreatography (MRCP) to evaluate lesions. Figure 1 shows a schematic diagram of a portion of a human subject showing the locations of various different organs, including the digestive system, liver, gallbladder, and part of the pancreas. MRI scans using MRCP sequences are recommended in guidelines and are commonly used to monitor patient health. Figure 2 shows MRI images of this region from a human subject. MRCP is an MRI technique used to image the bile ducts and pancreatic ducts. Pancreatic cystic lesions also exhibit high contrast in MRCP images. When monitoring pancreatic cystic lesions, MRCP images are used to manually assess the size of the pancreatic ducts and pancreatic lesions and to qualitatively evaluate the interaction between the lesion and the pancreatic ducts.

[0005] Known methods exist for quantitatively measuring bile ducts and pancreatic ducts. This method is described in European Patent No. 3443533B and is also shown in Figure 3. As shown in Figure 3, known techniques include standardized protocols for image acquisition, tubular enhancement and segmentation, a quantitative biliary tree model, and determination of metrics for the overall tree and individual ducts. This method forms the basis of an existing product called MRCP+. Figure 4(a) is an example of a report showing the results of MRCP analysis in a healthy subject. Figure 4(b) is an example of a report showing the results of MRCP analysis in a subject with chronic pancreatitis. However, this existing method does not measure pancreatic lesions or their interaction with any ducts.

[0006] Figures 4(a) and 4(b) show the MRCP+ models and corresponding reports obtained for healthy volunteers and subjects with chronic pancreatitis for pancreatic duct quantification. As shown, no stenosis or dilation was detected in healthy subjects, but multiple stenosis and dilation were detected in subjects with chronic pancreatitis.

[0007] Clinical guidelines recommend that radiologists manually measure features on medical scan images, such as the maximum diameter of cystic lesions or pancreatic ducts, and qualitatively assess features of concern, such as mural nodules, connectivity between cystic lesions and pancreatic ducts, and changes in the lesion over time. These assessments are then used in a binary manner to risk-stratify the lesions.

[0008] Current clinical assessments are performed manually and heavily rely on subjective characteristics, resulting in considerable variability between different interpreters at different points in time, and even among the same interpreter.

[0009] Furthermore, these subjective assessments make it difficult to distinguish between benign, pre-malignant, and malignant lesions. Due to the variability in assessments and the difficulty in identifying high-risk cases, many patients are diagnosed at a late stage or receive excessive treatment, and the majority of patients ordered to undergo surgery for pancreatic cystic lesions do not show high-grade dysplasia or malignancy on post-resection histological examination.

[0010] Progress has been made in the risk stratification of lesions by academic consortia. For example, the COMP Cystic Consortium has developed a method based on a combination of genetic markers obtained from invasive evaluation of cystic lesions and standard clinical assessments. However, these methods still involve only manual qualitative evaluation from imaging data.

[0011] This invention utilizes novel quantitative metric measurements from medical scan imaging data of internal organs. This method may improve the stratification of patients' current and future malignancy risk by characterizing the biophysical features of cystic lesions, the interaction between the lesions and the pancreatic duct, common bile duct, and pancreatic parenchyma, or other visceral ducts. Using these methods, the inventors aim to improve the accuracy of cystic lesion assessment, reduce the time required to perform the assessment, and minimize operator variability compared to conventional methods. [Prior art documents] [Patent Documents]

[0012] [Patent Document 1] European Patent No. 3443533B [Overview of the project]

[0013] There is a method for generating quantitative parameters for the pancreatic and bile ducts and lesions in internal organs from volumetric medical imaging scan data obtained from medical imaging scans of all or part of the internal organs. This method may include the step of identifying the location of at least one lesion and at least one pancreatic and bile duct within the internal organs on the volumetric medical imaging scan data.

[0014] This method may include the step of performing volume segmentation within a medical imaging scan using volumetric medical imaging scan data in order to segment at least one lesion. This method may include the step of segmenting at least one pancreatic and bile ducts using volumetric medical imaging scan data. Within the volume, there may be viscera imaged by the medical imaging scan. Within the volume, the viscera may include all or part of an internal organ.

[0015] The method may include the step of determining at least one structural parameter of at least one segmented lesion. The method may include the step of determining at least one structural parameter of at least one segmented pancreaticobiliary duct.

[0016] The pancreaticobiliary duct is typically tubular and is therefore called a tubular pancreaticobiliary duct. The pancreaticobiliary duct is involved in the transport of bile and pancreatic juice, specifically the bile duct, pancreatic duct, and their associated tubules and tubular branches. Therefore, the procedure for segmenting and identifying the pancreaticobiliary duct is the procedure for segmenting and identifying the tubular ducts involved in the transport of bile and pancreatic juice, specifically the bile duct, pancreatic duct, and their associated tubules and tubular branches, within the internal organs of a medical imaging scan volume.

[0017] The structural parameters of at least one segmented pancreaticobiliary duct may include quantitative parameters.

[0018] In a preferred embodiment of the present invention, a set of structural or quantitative structural parameters for at least one segmented pancreaticobiliary duct or tubular duct includes at least one parameter representing at least one of the following: The position of the midline of at least one segmented pancreaticobiliary duct at at least one location; the orientation of the midline of at least one segmented pancreaticobiliary duct at at least one location; Radius of at least one segmented pancreaticobiliary duct at at least one location; The diameter of at least one segmented pancreaticobiliary duct at at least one location; Orientation of a cross-section of at least one segmented pancreaticobiliary duct at at least one location; Cross-sectional profiles of at least one segmented pancreaticobiliary duct at at least one location; indication of the presence of at least one bifurcation of the segmented pancreaticobiliary duct at at least one location; and The orientation of at least one segmented pancreaticobiliary bifurcation at at least one location.

[0019] More preferably, the set of quantitative structural parameters for at least one segmented endovascular lesion includes at least one parameter representing at least one of the following: Maximum diameter of segmented lesions; volume of segmented lesions; Determination of the presence of septa within segmented lesions; Determination of the shape of segmented lesions; determination of the contents of segmented lesions.

[0020] The structural parameters of at least one segmented lesion may include quantitative parameters or qualitative parameters or both. In a preferred embodiment of the present invention, the method uses one or more of the structural parameters or qualitative structural parameters for at least one segmented lesion and one or more of the structural parameters or quantitative structural parameters for the segmented pancreaticobiliary duct to determine the distance between the segmented duct and the segmented lesion.

[0021] Preferably, the method further includes dividing the identified segmented pancreaticobiliary duct into regions upstream of the lesion, regions downstream of the lesion, and regions that interact with the lesion, and for each of these different regions, determining the diameter and diameter variation of the pancreaticobiliary duct or tubular duct.

[0022] In one embodiment of the present invention, the width profile of at least one of the upstream region, downstream region, and interaction region of the tubular pancreaticobiliary duct or tubular duct is used to determine additional metrics of the duct.

[0023] More preferably, the method further includes detecting a change pattern within at least one segmented pancreaticobiliary duct based on the median diameter estimate along the length of the at least one segmented pancreaticobiliary duct. Preferably, the method includes detecting a change pattern within at least one segmented pancreaticobiliary duct by applying at least one pattern detection window to the median diameter estimate data.

[0024] In a preferred embodiment of the present invention, the method includes detecting at least one of stenosis, dilation, and beading.

[0025] More preferably, at least one lesion is identified based on the maximum intensity projection of the voxels constituting the volumetric medical imaging scan data.

[0026] Preferably, at least one lesion is identified by applying a threshold to all medical imaging scan data and examining the volume rendering of all voxels exceeding the threshold.

[0027] Preferably, the internal organ is the pancreas.

[0028] In a preferred embodiment of the present invention, the method further includes the step of performing tubular enhancement on volumetric medical imaging scan data to derive at least one tubularity measurement for each voxel in the volumetric medical imaging scan data, wherein the tubular enhancement includes, for each voxel in the volume of medical imaging scan data, calculating at least one metric that distinguishes tubular structures from other structures, and calculating a tubularity measurement for each voxel based at least partially on the at least one metric.

[0029] Preferably, the method includes a step of using a multiscale Hessian-based highlighting of tubular pancreatic and biliary duct structures.

[0030] In a preferred embodiment of the present invention, volumetric medical imaging scan data includes imaging scan data obtained via magnetic resonance imaging modality.

[0031] Preferably, the method further includes the step of determining the total number of pancreatic lesions shown in the volumetric imaging scan data.

[0032] This method may be performed by a system. The method may include the step of locating the position of at least one lesion within an internal organ on medical imaging scan data. The segmented structural parameters of at least one pancreatic and bile duct may include quantitative parameters. The method may include the steps of segmenting at least one detected lesion and determining one or more quantitative structural parameters of at least one segmented lesion.

[0033] This method may include the step of segmenting a volume of medical imaging scan data to identify tubular structures involved in the transport of bile and pancreatic juice within the internal organs of the medical imaging scan volume, specifically bile ducts, pancreatic ducts, and their associated tubules and tubular branches.

[0034] The method is a procedure for at least one segmented pancreaticobiliary duct in a volume of medical imaging scan data, which may include a procedure for calculating at least one set of quantitative structural parameters for at least one location along the length of the segmented pancreaticobiliary duct.

[0035] This method may include the step of generating at least one set of quantitative structural parameters for at least one segmented tubular tube.

[0036] The image processing system may be configured to generate quantitative parameters for pancreaticobiliary ducts and lesions in visceral organs from volumetric medical imaging scan data, wherein the scan shows all or part of the visceral organ, and the image processing system comprises at least one processing device configured to: locate the location of at least one lesion in the visceral organ on the medical imaging scan data; segment the at least one detected lesion and determine one or more quantitative structural parameters for at least one segmented lesion; perform volume segmentation of the medical imaging scan data to identify tubular pancreaticobiliary ducts in the visceral organs within the volume of the medical imaging scan; calculate at least one set of quantitative structural parameters for at least one location along the length of the tubular pancreaticobiliary duct for at least one segmented tubular pancreaticobiliary duct within the volume of the medical imaging scan data; and generate at least one set of quantitative structural parameters for at least one segmented tubular pancreaticobiliary duct.

[0037] In exemplary embodiments of the present invention, a set of quantitative structural parameters for at least one segmented pancreaticobiliary duct includes at least one parameter representing at least one of the following: The position of the midline of at least one segmented pancreaticobiliary duct at at least one location; the orientation of the midline of at least one segmented pancreaticobiliary duct at at least one location; Radius of at least one segmented pancreaticobiliary duct at at least one location; The diameter of at least one segmented pancreaticobiliary duct at at least one location; Orientation of a cross-section of at least one segmented pancreaticobiliary duct at at least one location; Cross-sectional profiles of at least one segmented pancreaticobiliary duct at at least one location; indication of the presence of at least one bifurcation of the segmented pancreaticobiliary duct at at least one location; and The orientation of at least one segmented pancreaticobiliary bifurcation at at least one location.

[0038] More preferably, the set of quantitative structural parameters for at least one segmented endovascular lesion includes at least one parameter representing at least one of the following: Maximum diameter of segmented lesions; volume of segmented lesions; Determination of the presence of septa within segmented lesions; determination of the shape of segmented lesions; determination of the contents of segmented lesions.

[0039] In a preferred embodiment of the present invention, the processing device is further configured to perform the step of determining the distance between a segmented pancreaticobiliary duct and a segmented lesion using one or more qualitative structural parameters for at least one segmented lesion and one or more quantitative structural parameters for a segmented pancreaticobiliary duct.

[0040] Preferably, the processing device is further configured to perform the steps of dividing the identified segmented pancreatic and bile ducts into upstream regions of the lesion, downstream regions of the lesion, and regions interacting with the lesion, and determining the diameter and diameter variability of the pancreatic and bile ducts for each of these different regions.

[0041] More preferably, the width profiles of different regions of the tubular pancreaticobiliary duct are used to determine additional metrics for the duct.

[0042] Preferably, the processing device is further configured to perform the step of detecting a pattern of change within at least one segmented pancreaticobiliary duct based on median diameter estimates along the length of at least one segmented pancreaticobiliary duct.

[0043] In a preferred embodiment of the present invention, the processing device is further configured to perform the step of detecting at least one pattern of change within a segmented pancreatic and bile duct by applying at least one pattern detection window to median diameter estimate data.

[0044] More preferably, the processing device is further configured to perform the step of detecting at least one of narrowing, dilation, and beading.

[0045] Preferably, the processing device is further configured to identify at least one lesion by applying a threshold to all medical imaging scan data and examining the volume rendering of all voxels exceeding the threshold.

[0046] More preferably, the processing device is further configured to perform tubular enhancement on volumetric medical imaging scan data to derive at least one tubularity measurement for each voxel in the volumetric medical imaging scan data, the tubular enhancement comprising, for each voxel in the volume of medical imaging scan data, calculating at least one metric that distinguishes tubular structures from other structures, and calculating a tubularity measurement for each voxel based at least partially on the at least one metric.

[0047] In preferred embodiments of the present invention, any reference to the terms “tube” or “pancreaticobiliary duct” refers to the tubules involved in the transport of bile and pancreatic juice, and specifically includes all bile ducts, pancreatic ducts, and their associated tubules and tubular branches.

[0048] Further details, aspects, and embodiments of the present invention are described merely as examples with reference to the drawings. In the drawings, the same reference numerals are used to identify the same or functionally similar elements. Elements in the drawings are shown for simplicity and clarity and are not necessarily drawn to scale. [Brief explanation of the drawing]

[0049] [Figure 1] This is a schematic diagram showing various different human organs (conventional technology). [Figure 2] This figure shows an exemplary MRCP image of the region including the bile ducts, pancreas, and gallbladder (prior art). [Figure 3] This figure shows the process of performing quantitative imaging of the biliary tree using MRCP+ (conventional technique). [Figure 4(a)] This figure shows an example of a report illustrating the results of MRCP analysis in healthy subjects. [Figure 4(b)] This figure shows an example of a report illustrating the results of MRCP analysis in subjects with chronic pancreatitis. [Figure 5(a)] This figure shows an example of a medical scan image used in one embodiment of the present invention. [Figure 5(b)] This figure shows another example of a medical scan image used in one embodiment of the present invention. [Figure 5(c)] This figure shows yet another example of a medical scan image used in one embodiment of the present invention. [Figure 6(a)] This figure shows a medical scan image and the resulting data obtained during a harsh stage of one embodiment of the method of the present invention. [Figure 6(b)] This is another figure showing a medical scan image and the resulting data obtained during a harsh stage of one embodiment of the method of the present invention. [Figure 6(c)] This is yet another figure showing a medical scan image and resulting data obtained during a harsh stage of one embodiment of the method of the present invention. [Figure 6(d)] This is yet another figure showing a medical scan image and resulting data obtained during a harsh stage of one embodiment of the method of the present invention. [Figure 7] This is a schematic diagram of a measurement between a duct and a cystic lesion according to one embodiment of the present invention. [Figure 8] This flowchart shows some of the steps in the method of the present invention. [Figure 9] This flowchart shows the steps for loading data for processing. [Figure 10] This flowchart shows the steps involved in segmenting cystic lesions. [Figure 11] This flowchart shows the steps for segmenting a pipe according to one embodiment of the present invention. [Figure 12] This flowchart shows the steps for calculating one or more metrics according to one embodiment of the present invention. [Figure 13] This is a simplified flowchart illustrating an example of a method for detecting and quantifying narrowing and expansion within pipe structures. [Figure 14] This is a simplified block diagram of an example image processing system. [Modes for carrying out the invention]

[0050] The present invention will now be described with reference to the accompanying drawings. The drawings illustrate examples of methods and apparatus for generating quantitative data on biliary and pancreatic duct structures and / or information on cystic or malignant lesions from volumetric medical imaging scan data. However, it will be understood that the present invention is not limited to the specific examples described herein and shown in the accompanying drawings.

[0051] Furthermore, since most of the illustrated embodiments of the present invention can be implemented using electronic components and circuits known to those skilled in the art, the following description will not go into detail beyond what is deemed necessary to understand and appreciate the basic concepts of the present invention, nor will it obscure or distract from the teachings of the present invention.

[0052] Preferably, one embodiment of the present invention enables quantitative measurement of cystic or malignant lesions present in visceral organs and their interactions with the ducts of the visceral organs from medical scan images. In a preferred embodiment of the present invention, the medical scan image is an MRCP image. Preferably, the visceral organ is the pancreas, but the method may be used for other organs. Another embodiment of the present invention also enables the creation of a risk score for malignant tumors of visceral organs by combining a series of automated quantitative measurements, which can assist in clinical decision-making. In one embodiment of the present invention, the method improves the characterization of ducts within visceral organs, such as the pancreatic duct.

[0053] The present invention can be used to predict the current and future malignant risk of lesions in visceral canals based on the proximity of the lesion to the duct, the size of the lesion, and one or more quantitative indicators showing the impact of the lesion on one or more ducts of the visceral canals. The present invention includes measuring the morphology of the visceral canals downstream and upstream of the lesion, as well as the morphology of the ducts in the area of ​​the approach closest to the lesion. Since cystic lesions of the head of the visceral canals may also interact with the common bile duct, in some embodiments of the present invention, the morphology of the common bile duct is also measured.

[0054] Furthermore, one embodiment of the present invention may also incorporate information from additional MRI sequences to characterize the contents of the cyst and the parenchyma of the endovascular organs. These may include T1 mapping (normalized for magnetic field intensity), proton density lipid fraction (PDFF), and diffusion-weighted imaging (DWI).

[0055] Figures 5(a) to 5(c) show the maximum intensity projections of MRCP coronal images with image-enhanced pancreatic cystic lesions in 502, 504, and 506.

[0056] Figure 6 illustrates some of the steps used to evaluate medical scan images of a patient's internal organs. Preferably, the organ is the pancreas, and the evaluation relates to medical scan images of the pancreatic system. In a preferred embodiment of the present invention, the medical scan image is an MRI scan, and MRI data showing the patient's pancreas is loaded into the software. This data must include at least an MRCP scan. Scan data from additional 2D or 3D MRI sequences (e.g., T1, PDFF, or DWI data) may also be loaded.

[0057] First, in a preferred embodiment of the present invention, the MRCP scan is a scan of the pancreas, as well as pancreatic lesions, the common bile duct, and the pancreatic duct. These are identified in the MRCP data, as shown in Figure 6(a), and the lesions are identifiable on the maximum intensity projection (MIP) of the 3D MRI array obtained from the MRCP data, with Figure 6(b) showing the 3D coordinates. In a preferred embodiment of the present invention, these selected regions of the scan are segmented. Segmentation of medical scan images and general digital images is a well-known process of identifying objects and boundaries in an image and assigning labels to each pixel / voxel in the image so that pixels sharing certain characteristics are assigned the same label.

[0058] The simplest method of image segmentation is called "thresholding," in which grayscale digital image data is converted to binary image data based on a comparison of the grayscale data of each pixel / voxel with a threshold. Preferably, segmentation is performed using local thresholds automatically calculated at the location of the lesion and within the pancreaticobiliary tree, as shown in Figure 6(c).

[0059] In a preferred embodiment of the present invention, a tubular enhancement algorithm that labels pixels according to their likelihood of containing tubular structures is also incorporated into the segmentation of the pancreatic duct and common bile duct. In a preferred embodiment of the present invention, the method also includes the step of performing tubular enhancement on volumetric medical imaging scan data to derive at least one tubularity measure for each voxel in the volumetric medical imaging scan data, wherein the tubular enhancement includes, for each voxel in the volume of medical imaging scan data, calculating at least one metric that distinguishes tubular structures from other structures, and calculating a tubularity measure for each voxel based at least partially on the at least one metric. More preferably, the method includes the step of using multiscale Hessian-based tubular bile duct structure enhancement.

[0060] As shown in Figure 6(d), one or more metrics of lesion morphology (e.g., maximum diameter, volume, sphericity, interaction with the pancreatic duct) are measured directly from the lesion segmentation mask. Duct morphology is assessed by first measuring the centerlines of the duct segmentation mask and then calculating the diameter of the duct at all points along these centerlines. An interpolation-based method is used to provide sub-pixel scale diameter estimates. From these point-by-point diameter estimates, duct morphological metrics (e.g., median diameter, local diameter change region) are calculated.

[0061] If additional MRI parametric maps (e.g., T1, PDFF) are loaded into the software, the values ​​for these parametric maps are calculated within the parenchyma of visceral tissues and, if requested by the user, within lesions. These values ​​can be calculated using either a user-defined region of interest or an automated segmentation mask of visceral tissues within the parametric map (e.g., via a convolutional neural network).

[0062] Finally, one or more of the calculated metrics and any visualizations of the medical images and segmentation are output from the present invention in a digital machine-readable file format (e.g., a JSON file, PDF, PNG, or other standard format for images).

[0063] Figure 7 shows different measurements that can be performed according to the method of the present invention. An increase in the diameter of the upstream duct due to interaction with the lesion can be an indicator of malignancy. In one embodiment of the present invention, the interaction between a pancreatic cystic lesion and the main pancreatic duct (MPD) can indicate malignancy. The output of this function enables the automated detection and quantification of phenomena qualitatively assessed by radiologists, such as cystic lesions causing upstream dilation of the pancreatic or common bile duct, which are known high-risk features that may require surgical removal of the lesion.

[0064] The various metrics that can be obtained by the method of the present invention include, but are not limited to, the following: measurement of lesion morphology in the visceral tract (e.g., maximum diameter of the lesion); detection of one or more interactions between the lesion and the MPD / common bile duct (CBD); measurement of MPD / CBD morphology (median diameter, stenosis, dilation); lesion shape; detection of septa within the lesion; number of septa present within the lesion; number of lesions present (monocystic, oligocystic, polycystic); lesion content; changes in MRI signal detected in the pancreatic parenchyma between the lesion and the duct; and changes in MRI signal detected in a remote region of the pancreatic parenchyma away from the lesion.

[0065] Figure 8 shows a schematic diagram of the steps of an exemplary embodiment of method 900 for analyzing medical scan images. In a preferred embodiment of the present invention, the medical scan image is an image obtained via magnetic resonance imaging modality. In step 902, data for analysis is loaded. The following step 904 is segmentation of lesions of visceral organs shown in the scan data. The steps for lesion segmentation are described in more detail in Figure 11 below. In step 906, tubular structures in the medical scan image are segmented. This is followed by step 908, in which segmentation of lesions and / or tubules is potentially mapped to other image data. In step 910, quantitative metrics for lesions and / or tubules are calculated. In step 912, a risk score is calculated, and in step 914, all results are saved. In one embodiment of the present invention, a method is described for generating quantitative parameters of tubules and lesions of visceral organs from volumetric medical imaging scan data, where the scan shows all or part of the visceral organs. Preferably, the method includes the steps of: identifying the location of at least one lesion within an internal organ on medical imaging scan data; segmenting the at least one detected lesion and determining one or more quantitative structural parameters for the at least one segmented lesion; performing segmentation of the volume of medical imaging scan data to identify both internal and adjacent tubular tubes within the volume of the medical imaging scan; calculating at least one set of quantitative structural parameters for at least one location along the length of the tubular tube for the at least one segmented tubular tube in the volume of medical imaging scan data; and generating at least one set of quantitative structural parameters for the at least one segmented tubular tube.

[0066] Examples of the types of quantitative structural parameters that can be generated by the above method include: The location of one or more central lines of segmented tubular pancreaticobiliary structures; The orientation of the central line of the segmented tubular pancreaticobiliary structure at one or more locations along the length of the segmented tubular pancreaticobiliary structure; Radius values ​​of segmented tubular pancreaticobiliary structures at one or more locations along the length of the pancreaticobiliary structure; The diameter values ​​of segmented tubular pancreaticobiliary structures at one or more locations along the length of the pancreaticobiliary structure; Orientation of the cross-section of a segmented pancreaticobiliary structure at one or more locations along the length of the pancreaticobiliary structure; Cross-sectional profiles (shapes) of segmented tubular pancreaticobiliary structures at one or more locations along the length of the pancreaticobiliary structure; An indicator of the presence of branching points for segmented tubular pancreaticobiliary structures at one or more locations along the length of the pancreaticobiliary structure; and The orientation of the branching points of segmented tubular pancreaticobiliary structures at one or more locations along the length of the pancreaticobiliary structure.

[0067] In particular, in some exemplary embodiments of the present invention, as will be described in more detail below, the method may include the step of determining such quantitative structural parameters at multiple locations along the length of each identified tubular pancreaticobiliary structure.

[0068] Advantageously, by generating quantitative tubular and lesion data, including quantitative structural parameters generated from volumetric scan data obtained, for example, via magnetic resonance imaging modality, clinicians are provided with enhanced information about a patient's tubular organs, enabling them to more accurately assess the pathology of the patient's tubular organs, in particular, to more accurately assess the topological and structural features of one or more tubular organs and / or lesions within the tubular organs, such as the presence of tubular stenosis and dilation. Thus, embodiments of the present invention provide improved non-invasive techniques that enable clinicians to detect changes in tubular organs, more generally, improved non-invasive diagnostic techniques for providing clinicians with an accurate assessment of the interaction between tubular organs and lesions.

[0069] Figure 9 shows step 1000, which is necessary to load medical scan data into the system in order to perform this method, and corresponds to step 902 in Figure 9. In step 1002, the directory is loaded with the cases, in step 1004, the validity of each case in the directory is checked, and in step 1006, the valid cases are loaded into the application.

[0070] Figure 10 illustrates the steps required for the method of the lesion segmentation workflow 1100. In step 1102, the lesion location is identified on a medical scan image. The input in this step is preferably a 3D MRI image, stored as a three-dimensional array of voxels (where a voxel is a 3D pixel), and the output is the (row, column, slice) coordinates of any voxel within the lesion in the 3D array. Of course, in alternative embodiments of the present invention, 2D images may also be used.

[0071] In a preferred embodiment of the present invention, a maximum intensity projection (MIP) of a 3D MRI array is utilized. The MIP is a 2D array of pixels calculated by taking the maximum intensity of the 3D image along one direction. When calculating the MIP, a lookup table is also calculated that records the position of each pixel in the MIP within the 3D MRI image. Preferably, at least one lesion is identified based on the maximum intensity projection of the voxels that make up the volumetric medical imaging scan data.

[0072] Pixels within the MIP containing the lesion to be measured are identified. From the selected pixels, the 3D position of the lesion voxel is obtained from a lookup table.

[0073] In an alternative embodiment of the present invention, 3D MRI data alone can be used to locate a lesion on a medical scan image and directly identify and measure voxels within the lesion. Alternatively, a threshold can be applied to the entire image, and the volume rendering of all voxels exceeding this threshold can be presented to the user, who can then directly click on the portion of the volume rendering that they believe to be a lesion. In one embodiment of the present invention, at least one lesion is identified by applying a threshold to all of the medical imaging scan data and examining the volume rendering of all voxels exceeding the threshold.

[0074] The next step is lesion segmentation using local thresholds in step 1104. This step has a 3D MRI image as input and the coordinates of any voxel within the lesion (output from step 1102), the output being a 3D lesion mask, i.e., a voxel array the same size as the input image, where all voxels containing the lesion are set to 1 and all other voxels are set to 0.

[0075] In a preferred embodiment of the present invention, all pixels in a 3D MRI image within a defined distance of a selected point within a lesion are selected. Using these selected pixels, an intensity threshold is calculated that separates these pixels into foreground (MRI voxels above the threshold) and background (MRI voxels below the threshold) classes. This may be done, for example, by calculating the variability between classes at each threshold for a set of thresholds ranging from the lowest to the highest image intensity values, and returning the threshold that maximizes this variability (commonly known as the Otsu method).

[0076] A 3D mask is calculated in which all voxels in a 3D MRI image exceeding a threshold are set to 1, and all others are set to 0. The connected components of this mask are then identified. A connected component is any set of foreground voxels that form a continuous object; that is, for each pair of voxels in a connected component, a continuous pathway must exist between them via a voxel array containing only other foreground voxels. The connected component containing selected voxels within the lesion is identified, and all other components are removed from the mask. This final mask is returned as the 3D lesion mask.

[0077] However, in an alternative embodiment of the present invention, step 1104 can be omitted, and the method proceeds directly to step 1106 to find a threshold, or skips to step 1110 to manually outline the lesion using a paintbrush tool. Another valid alternative is to use a selected point as a seed point in the 3D image and iteratively grow a lesion mask from that seed point, with each iteration identifying MRI voxels adjacent to the current mask and adding any voxels whose intensity is within a specific threshold for voxels in the current mask to the mask. This method is repeated until the lesion mask stops growing or a certain number of iterations are reached.

[0078] Step 1106 is manual fine-tuning of the threshold by arbitrary selection. The inputs to this step are preferably the 3D MRI image, the coordinates of any voxels within the lesion (output from step 1102), and the 3D cyst mask from step 1104, if step 1104 has been performed. In a preferred embodiment of the present invention, the user is presented with a visualization of the 3D MRI image with the current lesion mask overlaid thereon. The user may manually input a different image threshold than the one calculated in step 1104, and the lesion mask is recalculated with the manually selected threshold by applying the same method.

[0079] Step 1110 is an optional step for separating the mask from the surrounding structure. In a preferred embodiment of the present invention, this step presents the user with a visualization of a 3D MRI image on which the current lesion mask (output from either step 1104 or 1106) is overlaid. If the lesion is connected to another anatomical structure that has high intensity in the MRI image, such as the pancreatic duct, the mask may include both the mask and the connected structure. If this situation occurs, the user can use a watershed segmentation algorithm to separate the lesion from the surrounding structure.

[0080] In this method, the distance transformation of the lesion mask is calculated. This returns a 3D voxel array the same size as the 3D lesion mask, and the appropriate distance of each voxel from each voxel to the nearest background voxel in the 3D lesion mask is recorded. The locations of the maxima in the distance-transformed image are then recorded. Using the maxima as seed points, a series of masks are iteratively grown from these seed points until adjacent masks meet each other at "ridges" in the distance map and all voxels in the original foreground region are labeled. The mask containing the selected points in the lesion is considered the new lesion mask. If this division also divides the lesion into sub-components, any sub-masks contained within the lesion can be recombined in step 1112.

[0081] In an alternative embodiment of the present invention, separation can also be performed by iteratively eroding a combined mask containing both the tubule and the lesion until the tubule is no longer visible, and then iteratively expanding this eroded mask containing only the lesion for the same number of iterations required to remove the tubule. This alternative approach takes advantage of the fact that the tubule is generally thinner than most lesions and therefore disappears with fewer erosion iterations than the lesion mask.

[0082] The final step 1114 of Method 1100 allows for the editing of the lesion mask using the paintbrush tool (optional). In a preferred embodiment of the present invention, in this step of the method, the user is presented with a visualization of the 3D MRI image with the current lesion mask overlaid thereon. If the mask does not properly cover the lesion, the user can manually edit the lesion mask using the paintbrush tool, even after applying Methods 1106, 1110, and 1112. Here, the user can scroll through slices of the 3D MRI image, manually mark any voxels within the lesion, and manually remove any voxels outside the lesion.

[0083] Figure 11 is a flowchart of the steps required for tubular segmentation, as shown in step 906 of Figure 9. This is a more detailed description of the process performed in step 906. In step 1202, the tubular is selected automatically or by the user and thresholding is performed using local contrast. In the following step 1204, the start and end of the pancreatic duct and / or the start and end of the common bile duct are selected. The next step is step 1206, in which tubular structure is detected using tubularity measurements. In step 1208, pathfinding is performed to move from the start to the end of the tubular to generate a centerline model. Finally, in step 1210, the diameter and diameter change of the tubular are quantified. In a preferred embodiment of the present invention, a set of quantitative structural parameters for at least one segmented tubular includes at least one parameter representing at least one of the following: The position of the centerline of at least one segmented pipe at at least one location; the orientation of the centerline of at least one segmented pipe at at least one location; The radius of at least one segmented pipe at at least one location; The diameter of at least one segmented tube at at least one location; The orientation of the cross-section of at least one segmented pipe at at least one location; Cross-sectional profile of at least one segmented pipe at at least one location; indication of the presence of at least one branch point of the segmented pipe at at least one location; and The orientation of at least one branching point of a segmented pipe at at least one location.

[0084] Figure 13 illustrates a workflow 1300 for measuring quantitative metrics regarding lesions and their interactions with the pancreatic duct and common bile duct. The steps of this method are an extension of step 910 of determining the quantitative metrics in Figure 9 above. In a preferred embodiment of the present invention, a set of quantitative structural parameters for at least one segmented visceral lesion includes at least one parameter representing at least one of the following: Maximum diameter of segmented lesions; volume of segmented lesions; Determination of the presence of septa within segmented lesions; determination of the shape of segmented lesions; determination of the contents of segmented lesions.

[0085] The first step, 1302, is to calculate the lesion dimensions. The inputs to this step are the 3D lesion segmentation mask generated from the method described above, as shown in Figure 11, and the 3D MRI image metadata (including information on voxel dimensions in millimeters). The output of this step is the maximum diameter of the lesion in three field planes, as well as the 3D volume of the lesion.

[0086] In a preferred embodiment of the present invention, in this step, the lesion mask is stored as a 3D voxel array of dimensions Nx × Ny × Nz, where all voxels containing the lesion are set to 1 and all others to 0. To calculate the maximum lesion diameter along the first field plane, all slices along the first dimension are looped through (i.e., from slice 1 to slice Nx), and for each slice containing at least one voxel set to 1, the edge voxels of the lesion are identified (i.e., voxels that are set to 1 but have at least one adjacent voxel set to 0), and the maximum distance between any pair of edge voxels is measured. This method is repeated for the other two field planes, and the second and third dimensions are looped through to obtain the maximum diameters of the second and third field planes, respectively. The volume of the lesion is also recorded.

[0087] In step 1304, the distance between the lesion and the pancreatic duct and common bile duct is measured. The input to this step is a 3D lesion segmentation mask including the pancreatic duct and common bile duct, or a 3D segmentation mask, or both, (obtained using the same method outlined in the MRCP+ patent). The output from this step is the nearest distance between the lesion and the pancreatic duct and common bile duct, as well as the (row, column, slice) coordinates of the nearest point on both of these ducts relative to the lesion. In a preferred embodiment of the present invention, one or more qualitative structural parameters for at least one segmented lesion and one or more quantitative structural parameters for a segmented duct are used to determine the distance between the segmented duct and the segmented lesion.

[0088] In a preferred embodiment of the present invention, a distance transformation of the lesion mask is calculated. This returns a 3D voxel array of the same size as the 3D lesion mask, and the distance from each background voxel to the nearest foreground pixel in the lesion mask is recorded. For all foreground voxels in the pancreatic duct and common bile duct segmentation masks, the corresponding values ​​in the distance transformation image are examined, and the foreground voxel with the minimum distance transformation value is recorded. The location of this voxel and its corresponding distance transformation value are the nearest point on the duct to the lesion and the minimum distance between the duct and the lesion, respectively.

[0089] If there are several ductal voxels located at the same minimum distance from the lesion, the average position is calculated (by taking the average row, column, and slice coordinates of each voxel).

[0090] If there is a very serious interaction between the lesion and the duct, the duct may not be modeled in the vicinity of the lesion, resulting in a gap in the duct mask near the lesion. In such cases, the proximity from the lesion to the duct is calculated using the duct's centerline representation generated by MRCP+. This is an ordered list of coordinates (x,y,z,d) of points along the duct's centerline, starting from the furthest downstream point, where the first three digits record the coordinates of the point in the image, and the fourth record the diameter of that point. First, the (x,y,z,d) coordinates of any gaps in the centerline model are estimated using interpolation, such as a spline. Then, at each point on the interpolated centerline, the distance transformation value of the lesion mask is calculated. The centerline point with the minimum distance transformation value is the centerline point closest to the lesion. The closest distance from the duct to the lesion is the distance transformation value of the closest centerline point to the lesion minus the radius of the duct at this point. If several centerline points are found to have the same minimum distance from the lesion, the median of the centerline points (ordered from the beginning to the end of the duct) is used as the closest point.

[0091] Alternatively, for every foreground voxel in the tubular mask, the distance to each foreground voxel in the lesion mask can be calculated. The closest distance between the lesion and the tubular is the minimum across all these distances, and the voxel in the tubular mask with this minimum distance is the closest point on the tubular to the lesion.

[0092] In step 1306, the connectivity between the lesion and the pancreatic duct and common bile duct is determined. In a preferred embodiment of the present invention, if the minimum distance (from step 1304) is below a certain threshold, they are recorded as connected. Otherwise, they are recorded as not connected.

[0093] If the above case also fails, and the user can visually confirm that the lesion and the tube are connected, there is also the option to manually label them as connected, and the distance from step 1304 is set to 0.

[0094] In an alternative embodiment of the present invention, the interaction between the lesion and the tubule is detected by first obtaining the lower of the image thresholds used for either the lesion or the tubule, and creating a mask in which any voxels in the original MRI image greater than this threshold are set to 1 and all other voxels are set to 0. If any connecting component in this mask overlaps with both the lesion and tubule masks, the lesion and tubule are marked as connected, and the distance calculated in step 1304 is overwritten as 0.

[0095] Another method involves repeatedly applying the expansion to either the lesion mask or the tubular segmentation mask and recording the number of iterations required to connect one mask to the other. If the number of iterations falls below a certain threshold, the two are considered connected.

[0096] The inventors can further supplement the methods outlined herein and alternative methods by also measuring the difference in pancreatic duct diameter upstream and downstream of the lesion, and if the diameter increases from downstream to upstream, or if the change in diameter deviates from a pre-characterized normal range, this suggests the possibility of interaction (in the absence of interaction, the diameter of the pancreatic duct is expected to decrease as it moves from downstream to upstream). If the diameter of the pancreatic duct increases from downstream to upstream, or if the output from any of the other methods outlined herein indicates that the duct is connected to the lesion, the lesion can be flagged as interacting with the duct. In a preferred embodiment of the invention, the method also includes the steps of dividing the identified segmented duct into an upstream region of the lesion, a downstream region of the lesion, and a region interacting with the lesion, and determining the diameter of the duct and the variability of the diameter for each of these different regions. More preferably, the width profiles of different regions of the tubular duct are used to determine additional metrics for the duct.

[0097] In step 1308, the dimensions of the tube are measured. In a preferred embodiment of the present invention, a (x,y,z,d) point along the centerline of the tube, starting from the head of the pancreas and ending at the tail, is used. The first three coordinates record the position of that point in the image, and the fourth is the measured diameter of the tube at that point.

[0098] The coordinates of the point on the pancreatic duct closest to the lesion are most likely to be on the surface of the pancreatic duct. First, we find which point on the pancreatic duct midline is closest to this point. This is recorded as the midline point closest to the cyst. If there is a gap in the duct mask and a midline-based method was used to find the closest point, the closest midline point from this method is used instead. Each point on the midline that is within r mm of the midline point closest to the lesion is labeled as being in an "interaction zone," where r is half the maximum diameter of the lesion calculated in step 1302. Since the lesion midline points are ordered from head to tail (i.e., downstream to upstream), all midline points before an interaction zone are labeled as downstream, and all midline points occurring after an interaction zone are labeled as upstream. The median diameter values ​​of the duct in the upstream, downstream, and interaction zones are returned, as is the median diameter of the entire duct.

[0099] Figure 13 shows a simplified flowchart 1400 of an example of a method for detecting and quantifying change patterns indicating at least one of stenosis, dilation, and beading within the biliary tree or pancreatic duct structure. The method in Figure 13 begins at 1410 and proceeds to 1420 where quantitative biliary tree and pancreatic duct data are loaded. In step 1430, the centerline points and duct widths are measured (steps 1208 and 1210) as described in Figure 11. Each duct is represented as an ordered list of centerline coordinates, including both the spatial position and duct width at each centerline point. In the illustrated example, the extracted duct width estimates can be optionally smoothed in step 1440 to remove noise that may cause false detections, for example, by applying a Gaussian kernel or local averaging.

[0100] In step 1450, local minimums are detected along the width profile of the ordered centerline points. The significance, or "severity," of each local minimum is characterized by measuring the absolute and relative differences in tube width at the minimum point compared to those of adjacent local maximums. If either or both of these differences exceed a given threshold for a given local minimum, that local minimum is labeled as a constriction.

[0101] In step 1460, local maximums are detected along the width profiles of the ordered centerline points. The significance, or "severity," of each local maximum is characterized by measuring the absolute and relative differences in pipe width at the maximum point compared to those of adjacent local minimums. If either or both of these differences exceed a given threshold for a given local maximum, that local maximum is labeled as an expansion.

[0102] Beading can also be detected, for example, based on the detected sequence of stenosis and dilation. In step 1470, a quantitative index of each detected stenosis and dilation, for example, the severity defined in steps 1450 and 1460, is calculated. Finally, in 1480, the stenosis and dilation data, including location and each quantitative index, as well as summary metrics such as the total number of stenosis and dilation and the percentage of the length of each tube containing stenosis and dilation, are recorded.

[0103] Figure 14 shows a simplified block diagram of an example of an image processing system 1500 that may be adapted according to embodiments of the present invention. The system 1500 comprises one or more processing devices 1510 configured to perform various processing functions. In some exemplary embodiments, one or more of the processing devices 1510 may include one or more processor cores configured to execute computer program code. Additionally / alternatively, at least one of the processing devices may include a hardware processing device such as an application-specific integrated circuit (ASIC) device or a hardware accelerator module, which has hardware circuitry configured to perform predetermined processing of data provided thereto.

[0104] The system 1500 further comprises one or more memory elements 1520. One or more memory elements 1520 may consist of one or more non-temporary computer program products, such as optical storage devices like hard disks and CD-ROM devices, magnetic storage devices, read-only memory, ROM, programmable read-only memory, PROM, erasable programmable read-only memory, EPROM, electrically erasable programmable read-only memory, EEPROM, and flash memory. The memory elements 1520 may additionally / alternatively include one or more volatile memory elements, such as random access memory (RAM) and cache memory.

[0105] For the sake of simplification and ease of understanding, we will refer below to a single processing device 1510 and a single memory element 1520. However, it should be understood that such references to a single processing device 1510 or a single memory element 1520 are intended to encompass multiple processing devices 1510 and multiple memory elements 1520, respectively.

[0106] The memory element 1520 may store executable computer program code that is executed by the processing device 1510. The memory element 1520 may further store data that is accessed and / or processed by the processing device 1510 when the computer program code is executed.

[0107] The system 1500 shown in Figure 14 further comprises one or more output devices, collectively indicated as 1530. Such output devices may include, for example, a display device, a printer device, a network interface device, and the like. The system 1500 shown in Figure 14 further comprises one or more user input devices, collectively indicated as 1540. Such input devices may include, for example, a keyboard, a keypad, a mouse, a touchscreen, and the like.

[0108] According to some examples of the present invention, the processing device 1510 is configured to generate quantitative data of biliary and pancreatic duct structures from volumetric medical imaging scan data in accordance with the various methods described above, for example with reference to the accompanying drawings. Thus, the processing device 1510 is configured to generate quantitative data of biliary and pancreatic duct structures from volumetric medical imaging scan data in accordance with the various methods described above with reference to the accompanying drawings. To identify tubular structures within a volume of medical imaging scan data, segmentation of the volume of medical imaging scan data is performed. For each location along the length of the identified tubular biliary and pancreatic structures, a set of quantitative structural parameters is calculated. Outputs quantitative biliary tree data and pancreatic data, including a set of quantitative structural parameters for identified tubular biliary structures and pancreatic structures. It is configured in this way.

[0109] For example, the processing device 1510 may be configured to execute computer program code stored in the memory element 1520 in order to perform at least some of the steps of the various methods described above with reference to the attached drawings.

[0110] As shown in Figure 14, the image processing system 1500 may be operably coupled to at least one data storage device 1550 from which the processing device 1510 can retrieve imaging scan data, and the processing device 1510 can output quantitative biliary tree data and pancreatic data, including a set of quantitative structural parameters for segmented tubular biliary structures and pancreatic structures.

[0111] The various values, metrics, measurements, and other data generated by the image processing system 1500 when generating quantitative biliary tree data and pancreatic data may also be output to and stored in the data storage device 1550, either collectively or separately, and such data may include, just as an example: Hessian volume; Eigenvectors; Ra ratio; Rb ratio; Intensity metric S; Tubular measurement value (To / To(σ)); Scale (σ); Segmented (binarized) volume centerline point; Estimated pipe width; Branch node; Medialness value; and Narrowing and / or dilating data.

[0112] The present invention will be described with reference to preferred embodiments. The present invention enables the following features: Lesion segmentation, detection of interactions between lesions and the main pancreatic duct and common bile duct; Measurement of pancreatic lesion morphology (e.g., maximum diameter, volume, sphericity); measurement of pancreatic duct and common bile duct morphology (e.g., median diameter, local diameter variation area); Quantification of the interaction between pancreatic cystic lesions and the pancreatic duct; interaction between lesions and the pancreatic duct / CBD; Detection of the point closest to the lesion and the duct; Measurement of the distance between the lesion and the pancreatic duct / CBD; Segmentation of the pancreatic duct / CBD into upstream, downstream, and interaction; detection of changes in pancreatic duct morphology; Detection of branching vessels connected to the lesion; and Quantification of mural nodules / solid components of lesions.

[0113] The method of the present invention can also be used to characterize the pancreatic parenchyma. This provides the following measurements or organ analyses: Characterization of the region between the lesion and the pancreatic duct using multiparametric MRI; Evaluation of pancreatic fat and inflammation in parenchyma using PDFF and T1; and Regional changes in pancreatic volume and regional changes in T1 and PDFF. The method of the present invention also allows for the characterization of lesions. This makes the following possible: Characterization of shape; maximum diameter; volume; Detection of septa within the lesion, determination of the number of septa present within the lesion; and The number of lesions present (including classification as monocystic, oligocystic, or polycystic depending on the presence of cystic lesions).

[0114] The method of the present invention also allows for the determination of cyst contents by changes in MRI contrast, including T1 and proton density fat percentage [PDFF]. These include one or more changes in MRI signals (T1, PDFF, diffusion-weighted imaging [DWI] signals) detected in the pancreatic parenchyma between the lesion and the duct.

[0115] The method of the present invention may also produce temporal changes that predict malignant tumors. These may include one or more of the following: A change in any of the above metrics indicating an increase in risk; Integration of images from the same patient at multiple time points to match lesions over time.

[0116] As described above, the present invention provides novel automated measurements of quantitative metrics characterizing cystic lesions and their interactions with the pancreatic duct, CBD, and pancreatic parenchyma. These offer more robust measurements that reduce inter-clinic variability, improved prediction of pancreatic cancer risk, and time savings through automation.

[0117] Changes can be evaluated in MRI signals detected in distant regions of the pancreatic parenchyma away from the cystic lesion, including T1, PDFF, and DWI signals.

[0118] This method may be applicable to broader assessments involving other modalities (e.g., computed tomography) or other organs (e.g., characterization of renal cysts).

[0119] As described above, the present invention may be implemented in a computer program for execution on an image processing system, the computer program including at least a code portion for performing steps of the method according to the present invention when executed on a programmable device such as an image processing system, or for enabling the programmable device to perform functions of the device or system according to the present invention.

[0120] A computer program is a list of instructions, such as a specific application program and / or an operating system. A computer program may include, for example, one or more of the following: subroutines, functions, procedures, object methods, object implementations, executable applications, applets, servlets, source code, object code, shared libraries / dynamically loaded libraries, and / or other sets of instructions designed to run on a computer system.

[0121] Computer programs may be stored internally on tangible, non-temporary computer-readable storage media, or they may be transmitted to a computer system via computer-readable transmission media. All or part of a computer program may be provided on computer-readable media permanently, removablely, or remotely coupled to an information processing system. Tangible, non-temporary computer-readable media may include, but are not limited to, any number of the following: magnetic storage media, including disk and tape storage media; optical storage media, such as compact disk media (e.g., CD-ROM, CD-R, etc.) and digital video disk storage media; non-volatile memory storage media, including semiconductor-based memory units such as flash memory, EEPROM, EPROM, and ROM; ferromagnetic digital memory; MRAM; and volatile storage media, such as registers, buffers, or caches, main memory, and RAM.

[0122] A computer process typically includes a program or part of a program that is currently running, current program values ​​and state information, and resources used by the operating system to manage the execution of the process. An operating system (OS) is software that manages the sharing of computer resources and provides programmers with interfaces used to access those resources. The OS processes system data and user input and responds by managing and assigning tasks and internal system resources as services to users and programs of the system.

[0123] The above specification has described the present invention with reference to specific examples of embodiments of the present invention. However, it will be clear that various modifications and changes can be made without departing from the scope of the present invention as described in the appended claims, and that the claims are not limited to the above-described examples.

[0124] Those skilled in the art will recognize that the boundaries between logic blocks are merely illustrative, and that alternative embodiments may integrate logic blocks or circuit elements, or impose alternative functional decompositions on various logic blocks and circuit elements. Therefore, it should be understood that the architectures described herein are merely illustrative, and in practice, many other architectures can be implemented to achieve the same functionality.

[0125] Any arrangement of components to achieve the same function is effectively “associated” in such a way that the desired function is achieved. Therefore, any two components in this specification combined to achieve a particular function, regardless of whether they are architectural or intermediate components, can be considered “associated” with each other in such a way that the desired function is achieved. Similarly, any two components thus associated can also be considered “operably connected” or “operably coupled” with each other in such a way that the desired function is achieved.

[0126] Furthermore, those skilled in the art will recognize that the boundaries between the operations described above are merely illustrative. Multiple operations may be combined into a single operation, a single operation may be distributed among additional operations, and operations may be performed with at least partial temporal overlap. Moreover, alternative embodiments may include multiple instances of a particular operation, and the order of operations may be modified in various other embodiments.

[0127] However, other modifications, variations, and substitutions are possible. Therefore, the specification and drawings should be considered illustrative rather than restrictive.

[0128] In the claims, no reference numerals placed in parentheses should be construed as limiting the claims. The word “comprising” does not preclude the existence of other elements or steps other than those enumerated in the claims. Furthermore, as used herein, the terms “a” or “an” are defined as one or more. Also, the use of prefixes such as “at least one” and “one or more” in the claims should not be construed as meaning that the introduction of another claim element by the indefinite article “a” or “an” limits any particular claim containing such introduced claim element to an invention containing only one such element, even if the same claim contains the prefixes “one or more” or “at least one” and the indefinite article “a” or “an.” The same applies to the use of definite articles. Unless otherwise specified, terms such as “first” and “second” are used to arbitrarily distinguish the elements described by such terms. Therefore, these terms are not necessarily intended to indicate any temporal or other prioritization of such elements. The mere fact that certain means are described in different claims does not imply that combinations of these means cannot be used advantageously.

[0129] The present invention is described only as an example. Therefore, the above should be considered only as an illustration of the principles of the present invention. Furthermore, since there are numerous modifications and changes that are readily conceivable to those skilled in the art, it is not desirable to limit the present invention to the exact configuration and operation illustrated and described, and therefore all suitable modifications and equivalents may be adopted, insofar as they fall within the scope of the claims.

Claims

1. A method for generating quantitative parameters for the pancreatic and bile ducts and lesions in internal organs from volumetric medical imaging scan data obtained from medical imaging scans of all or part of the internal organs, wherein the method is: The steps include identifying the location of at least one lesion within the internal organ and at least one pancreatic and bile duct on the volumetric medical imaging scan data, The steps include: using the volumetric medical imaging scan data to perform volume segmentation within the medical imaging scan to identify and segment the at least one lesion and the at least one pancreatic bile duct; A step of determining at least one structural parameter of at least one segmented lesion, The steps include determining at least one structural parameter of the segmented pancreaticobiliary duct and Methods that include...

2. The method according to claim 1, wherein the structural parameters of the segmented at least one pancreaticobiliary duct include quantitative parameters.

3. The method according to claim 1, wherein at least one of the structural parameters of the segmented pancreaticobiliary duct is determined for at least one position along the length of the at least one pancreaticobiliary duct.

4. The structural parameters for the segmented pancreaticobiliary duct of at least one are, The position of the midline of the at least one pancreatic and bile duct at the at least one of the aforementioned locations; The orientation of the centerline of the at least one pancreaticobiliary duct at the at least one location; the radius of the at least one pancreaticobiliary duct at the at least one location; The diameter of the at least one pancreatic and bile duct at the at least one of the aforementioned locations; The orientation of the cross-section of the at least one pancreatic and bile duct at the at least one of the aforementioned locations; Cross-sectional profile of the at least one pancreatic and bile duct at the at least one of the aforementioned locations; Indication of the presence of at least one branching point of the pancreatic and biliary ducts at the aforementioned at least one location; and The orientation of the branching point of the at least one pancreatic and bile duct at the at least one location, The method according to claim 3, comprising at least one parameter representing at least one of the following.

5. The method according to any one of claims 1 to 4, wherein the structural parameter of the segmented at least one lesion includes a quantitative parameter.

6. The set of structural parameters of the segmented at least one lesion is The maximum diameter of the lesion; the volume of the lesion; Determination of the presence of a septum within the lesion; Determination of the shape of the lesion; determination of the contents of the lesion, The method according to any one of claims 1 to 5, comprising at least one parameter representing at least one of the following.

7. The method according to any one of claims 1 to 6, wherein the structural parameter of the segmented at least one lesion includes a qualitative parameter.

8. The method described above is A step to determine the distance between a segmented lesion and a segmented pancreaticobiliary duct, using at least one of the structural parameters of the at least one segmented lesion and at least one of the structural parameters of the segmented pancreaticobiliary duct. The method according to any one of claims 1 to 7, further comprising:

9. The method described above is The steps include dividing the segmented and identified pancreaticobiliary duct into an upstream region of the lesion, a downstream region of the lesion, and a region of interaction with the lesion, and determining the diameter and diameter variation of the pancreaticobiliary duct for each of these different regions. The method according to any one of claims 1 to 8, further comprising:

10. The method according to claim 9, wherein at least one width profile of the upstream region, the downstream region, and the interaction region of the pancreaticobiliary duct is used to determine an additional metric of the pancreaticobiliary duct.

11. The method according to claim 9 or 10, further comprising the step of detecting a pattern of change within the at least one pancreatic bile duct based on an estimated median diameter along the length of the at least one pancreatic bile duct.

12. The method according to claim 11, further comprising the step of detecting the change pattern in the at least one pancreatic bile duct by applying the at least one pattern detection window to the median diameter estimate data.

13. The method according to claim 11 or claim 12, further comprising the step of detecting at least one of stenosis, dilation, and beading.

14. The method according to any one of claims 1 to 13, wherein the at least one lesion is identified based on the maximum intensity projection of the voxels constituting the volumetric medical imaging scan data.

15. The method according to any one of claims 1 to 14, wherein the segmented at least one lesion is identified by applying a threshold to all of the medical imaging scan data and examining the volume rendering of all voxels exceeding the threshold.

16. The method according to any one of claims 1 to 15, wherein the internal organ is the pancreas.

17. The method according to any one of claims 1 to 16, further comprising the step of performing tubular enhancement on the volumetric medical imaging scan data to derive at least one tubularity measurement for each voxel in the volumetric medical imaging scan data, wherein the tubular enhancement includes, for each voxel in the volume of the medical imaging scan data, calculating at least one metric that distinguishes a tubular structure from other structures, and calculating a tubularity measurement for each voxel based at least in part on the at least one metric.

18. The method according to any one of claims 1 to 17, further comprising the step of using a multiscale Hessian-based enhancement of tubular pancreaticobiliary structures.

19. The method according to any one of claims 1 to 18, wherein the volumetric medical imaging scan data includes imaging scan data obtained via magnetic resonance imaging modality.

20. The method according to any one of claims 1 to 19, further comprising the step of determining the total number of pancreatic lesions shown in the volumetric imaging scan data.

21. An image processing system configured to generate quantitative parameters for the pancreatic and bile ducts and lesions in internal organs from volumetric medical imaging scan data from scans showing all or part of internal organs, wherein the image processing system is The location of at least one lesion in the internal organ is identified on the medical imaging scan data, the identified at least one lesion is segmented, and one or more structural parameters of the segmented at least one lesion are determined. The volume of the medical imaging scan data is segmented to identify the pancreatic and bile ducts within the internal organs in the volume of the medical imaging scan. Generate at least one set of quantitative structural parameters for the segmented pancreaticobiliary duct. An image processing system comprising at least one processing device configured as follows.

22. The image processing system according to claim 1, configured to calculate at least one set of quantitative structural parameters for at least one location along the length of the pancreatic and bile ducts for at least one of the pancreatic and bile ducts identified within the volume of the medical imaging scan data.

23. The set of quantitative structural parameters for the segmented pancreaticobiliary duct, The position of the midline of the at least one pancreatic and bile duct at the at least one of the aforementioned locations; The orientation of the centerline of the at least one pancreaticobiliary duct at the at least one location; the radius of the at least one pancreaticobiliary duct at the at least one location; The diameter of the at least one pancreatic and bile duct at the at least one of the aforementioned locations; The orientation of the cross-section of the at least one pancreaticobiliary duct at the at least one location; the cross-sectional profile of the at least one pancreaticobiliary duct at the at least one location; Indication of the presence of at least one branching point of the pancreatic and biliary ducts at the aforementioned at least one location; and The orientation of the branching point of the at least one pancreatic and bile duct at the at least one location, The image processing system according to claim 22, comprising at least one parameter representing at least one of the following.

24. The image processing system according to claim 21, 22, or 23, wherein the set of structural parameters for at least one segmented lesion includes quantitative structural parameters.

25. The set of structural parameters for the segmented at least one lesion is, The maximum diameter of the lesion; the volume of the lesion; Determination of the presence of septa within the lesion; determination of the shape of the lesion; determination of the contents of the lesion, The image processing system according to claim 21, 22, 23, or 24, comprising at least one parameter representing at least one of the following.

26. The image processing system according to any one of claims 21 to 25, wherein the structural parameters for at least one segmented lesion include qualitative structural parameters.

27. The aforementioned processing device The image processing system according to any one of claims 21 to 26, further configured to perform the step of determining the distance between the pancreatic bile duct and the lesion using one or more of the structural parameters for at least one segmented lesion and one or more of the quantitative structural parameters for the segmented pancreatic bile duct.

28. The aforementioned processing device The steps include dividing the pancreaticobiliary duct into an upstream region of the segmented lesion, a downstream region of the segmented lesion, and an interaction region with the segmented lesion, and determining the diameter and diameter variation of the pancreaticobiliary duct for each of these different regions. An image processing system according to any one of claims 21 to 27, further configured to perform the following:

29. The image processing system according to claim 28, wherein the width profiles of the different regions of the pancreatic and bile ducts are used to determine additional metrics of the ducts.

30. The image processing system according to any one of claims 21 to 29, wherein the processing device is further configured to perform the step of detecting a change pattern within the at least one pancreatic bile duct based on median diameter estimates along the length of the at least one pancreatic bile duct.

31. The image processing system according to any one of claims 21 to 30, wherein the processing device is further configured to perform the step of detecting a pattern of change in at least one pancreatic bile duct by applying at least one pattern detection window to the median diameter estimation data.

32. The image processing system according to any one of claims 21 to 31, wherein the processing device is further configured to perform the step of detecting at least one of stenosis, dilation, and beading.

33. The image processing system according to any one of claims 21 to 32, wherein the processing device is further configured to identify at least one lesion by applying a threshold to all of the medical imaging scan data and examining the volume rendering of all voxels exceeding the threshold.

34. The aforementioned processing device The image processing system according to any one of claims 21 to 33, further configured to perform tubular enhancement on the volumetric medical imaging scan data to derive at least one tubularity measurement for each voxel in the volumetric medical imaging scan data, wherein the tubular enhancement includes calculating at least one metric for each voxel in the volume of the medical imaging scan data that distinguishes a tubular structure from other structures, and calculating a tubularity measurement for each voxel based at least in part on the at least one metric.

Images