Identifying candidate elements in image using atlas elements to determine disease state

Abstract

The invention relates to identifying candidate elements in an image using atlas elements to determine a disease state. Systems and methods for determining a disease state of an image element are disclosed herein. An example method may include providing a quantitative image and a non-quantitative image, determining a corresponding mapping of the quantitative image to an atlas, and determining candidate elements of the quantitative image by comparing the elements of the quantitative image to atlas elements of the atlas. The example method may also include locating the candidate element in the non-quantitative image, classifying the candidate element based on an attribute of the non-quantitative image, and determining a disease state of the candidate element.

Description

[0001] Divisional application

[0002] This application is a divisional application of the invention patent application with application date of April 20, 2018, application number 201880036870.0, and invention name “Using atlas elements to identify candidate elements in an image to determine disease status”.

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004] This application claims the benefit under 35 U.S.C. §119 of an earlier filing date of U.S. Provisional Application Serial No. 62 / 487,848, filed on April 20, 2017, the entire contents of which are hereby incorporated by reference in their entirety for all purposes. Technical Field

[0005] Examples described herein relate to a method of determining a disease state of a tissue using quantitative images and a system for performing the method. Background Art

[0006] Medical images consist of intensity values ​​at different points in space, which are usually arranged in a regular Cartesian grid. The grid can be, for example: two-dimensional, such as for planar x-ray imaging or MR slice imaging; three-dimensional, such as for 3D CT or 3D MRI imaging; or four-dimensional, such as for respiratory-related 4D CT imaging. The intensity values ​​depend on the imaging mode used; for example, the intensity values ​​can correspond to x-ray attenuation, or be related to the concentration of water molecules. Intensity values ​​are usually relative, for example, the value of the intensity value itself has no meaning, but only provides contrast between adjacent points in the grid. For example, clinicians can use changes in contrast to visually distinguish cancerous tissue from non-cancerous tissue.

[0007] In quantitative medical imaging, acquisition and reconstruction techniques attempt to produce images whose intensity values ​​are absolute, i.e., related to a physical or functional attribute of the patient at that point. One potential advantage of quantitative imaging is that images can be used to diagnose malignant areas in a patient based on the intensity values ​​themselves rather than subjectively comparing intensity changes between images. In practice, quantitative medical imaging is confounded by the large uncertainty in the relationship between intensity values ​​and patient attributes.

[0008] Visually searching through large medical imaging datasets and interpreting intensity values ​​based on these variations is both time consuming and requires a high level of experience, especially for whole body imaging of limited areas with no known suspected disease. Methods for automated diagnosis are confounded by the uncertainty of intensity values. The present disclosure is intended to meet this need. Summary of the invention

[0009] Disclosed herein are example methods and systems for determining candidate elements and / or disease states of the candidate elements. The example methods may include: providing at least one quantitative image of a subject, each of the at least one quantitative image comprising an element; providing an atlas corresponding to the subject, the atlas comprising atlas elements, each of the atlas elements comprising a metric distribution; determining a corresponding mapping associating elements of the at least one quantitative image with corresponding atlas elements; and determining a candidate element by comparing values ​​of certain elements of the at least one quantitative image with the metric distribution of the associated atlas element of the corresponding mapping.

[0010] The example method may also include: providing at least one additional image of the subject, each of the at least one additional image containing an element; locating the candidate element to a corresponding element of the at least one additional image; classifying the candidate element into at least one category based on attributes of the corresponding element of the at least one additional image; and determining the disease state of the candidate element by analyzing the at least one category of the at least one additional image corresponding to the candidate element.

[0011] An example method may include: locating a candidate element of a quantitative image to a corresponding element of at least one additional image; classifying the candidate element into at least one category based on an attribute of the corresponding element of the at least one additional image; and determining a disease state of the candidate element by analyzing the at least one category of the at least one additional image corresponding to the candidate element.

[0012] An example system may include at least one processing unit and at least one computer-readable medium encoded with instructions that, when executed, cause the system to: provide at least one quantitative image and at least one additional image of a subject, each of the at least one quantitative image and the at least one additional image comprising an element; provide an atlas corresponding to the subject, the atlas comprising atlas elements, each of the atlas elements comprising a metric distribution; determine a corresponding mapping that associates elements of the at least one quantitative image with corresponding atlas elements; determine candidate elements by comparing values ​​of certain elements of the at least one quantitative image with the metric distribution of associated atlas elements of the corresponding mapping; locate the candidate elements to corresponding elements of the at least one additional image; classify the candidate elements into at least one category based on attributes of the corresponding elements of the at least one additional image; and determine a disease state of the candidate element by analyzing the at least one category of the at least one additional image corresponding to the candidate element. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The foregoing and other features of the present disclosure will become more fully apparent from the following description and appended claims taken in conjunction with the accompanying drawings. It should be understood that these drawings depict only a few examples according to the present disclosure and are therefore not to be considered limiting of the scope of the present disclosure, which will be described with additional specificity and detail through the use of the accompanying drawings, in which:

[0014] Figure 1 is a flow chart of a method for determining a disease state of a tissue arranged in accordance with an embodiment of the present disclosure;

[0015] Figure 2 is a schematic diagram depicting a system for determining a disease state of a tissue arranged in accordance with an embodiment of the present disclosure;

[0016] Figure 3 is a flow chart of a method of locating candidate elements arranged according to an embodiment of the present disclosure;

[0017] Figure 4 is a schematic diagram of a corresponding mapping arranged according to an embodiment of the present disclosure;

[0018] Figure 5 is a schematic diagram of a metric distribution arranged according to an embodiment of the present disclosure;

[0019] Figure 6 is a flowchart of a method of filtering candidate elements arranged according to an embodiment of the present disclosure; and

[0020] Figure 7 is a flow chart of a method for locating and filtering candidate elements according to the present disclosure;

[0021] All of these figures are arranged in accordance with at least some embodiments of the present disclosure. DETAILED DESCRIPTION

[0022] In the following detailed description, reference is made to the accompanying drawings, which form a part of this specification. In the drawings, similar symbols generally identify similar parts, unless the context otherwise indicates. The illustrative examples described in the detailed description, drawings, and claims are not meant to be limiting. Other examples may be utilized and other changes may be made without departing from the spirit or scope of the subject matter presented herein. It is readily understood that the various aspects of the present disclosure, as generally described herein and shown in the drawings, may be arranged, replaced, combined, separated, and designed in a variety of different configurations, all of which are implicitly contemplated herein.

[0023] The present disclosure relates in particular to methods, systems, products, devices and / or apparatus generally related to providing at least one quantitative image of a subject, each of which contains an element. An atlas corresponding to the subject may be provided. The atlas contains atlas elements, each of which contains a metric distribution. A corresponding mapping associating the elements of the at least one quantitative image with the corresponding atlas elements may be determined. Candidate elements may be determined by comparing the values ​​of certain elements of the at least one quantitative image with the metric distribution of the associated atlas elements of the corresponding mapping. The present disclosure may also relate to providing at least one additional image of the subject, each of which contains an element. The candidate element may be located to a corresponding element in the elements of the at least one additional image. The candidate element may be classified into at least one category based on the attributes of the corresponding element of the at least one additional image. The disease state of the candidate element may be determined by analyzing the at least one category of the at least one additional image corresponding to the candidate element.

[0024] Figure 1 1 is a flow chart of a method for determining a disease state of a tissue arranged in accordance with an embodiment of the present disclosure. The example method may include one or more operations, functions, or actions, as shown in one or more of blocks 100, 102, 104, 105, 106, and / or 108. The operations described in blocks 100 to 108 may be performed in response to executing (such as by one or more processors described herein) computer-executable instructions stored in a computer-readable medium (such as a computer-readable medium of a computing device or some other controller of similar configuration).

[0025] Method 100 includes at least one image 102. Block 104 depicts "locating candidate elements" and may accept the at least one image 102 as input. Block 104 may output candidate elements 105. Block 104 may be followed by block 106, which may accept the candidate elements 105 as input and depict "filtering candidate elements." Block 106 may output diseased elements 108 of the tissue.

[0026] One or more images 102 may be medical images of a subject. The images 102 may be acquired during the method 100, or may be acquired in advance. If the images 102 are acquired in advance, they may be stored on a storage medium (such as a computer-readable medium) until ready for use in the method 100. The images 102 may have been acquired at the same location where the method 100 is being performed, or may have been transmitted from a remote location. The images 102 may be transmitted to the location where the method 100 is being performed via a wired or wireless connection. The images 102 may be a single image or multiple images, which may include a mixture of different imaging modes. In some embodiments, the modes may include magnetic resonance imaging (MRI), ultrasound, optical imaging, x-rays, computed tomography (CT), positron emission tomography (PET), or a combination thereof. The images 102 may be quantitative images, non-quantitative images, or a combination. The at least one image 102 may include elements such as pixels, voxels, groups of voxels corresponding to anatomical structures, surfaces, or a combination thereof.

[0027] Box 104 states "Locate candidate elements". The at least one image 102 is analyzed for abnormal areas. The attributes of each of the elements (or groups of elements) of the image 102 are determined and analyzed. The elements can be components of the image, and can be raw image data (e.g., intensity values ​​for each pixel), processed image data (e.g., normalized or background-corrected intensity values ​​for each pixel), groupings of image components to identified features (e.g., coordinates of pixel centroids identified as tissue structures), or combinations thereof. There can be multiple different types of elements in the image 102. Different attributes and values ​​can be associated with different types of elements (e.g., location, intensity, texture, tissue type, etc.). In some instances, the elements can be pixels, voxels, groups of voxels that have been identified as anatomical structures, surfaces that have been extracted from the data by a segmentation algorithm, annotations, points in the image coordinate space, or a combination thereof. Other elements can be used in other instances. The elements can be compared with normal ranges or probability distributions corresponding to different tissues. The elements determined to be of interest are output as candidate elements 105. Candidate elements 105 may represent a subset of elements of image 102. Candidate elements 105 may be regions with abnormal tissue properties and may be regions of interest for further clinical investigation. Candidate elements 105 may be used alone, for example, to guide medical treatment of a subject. Method 100 may continue to block 106, where candidate elements 105 are filtered.

[0028] Box 106 states "Filtering Candidate Elements". Box 106 may accept candidate elements 105 as input. Box 106 may use additional attributes of the at least one image 102 to determine which, if any, of the candidate elements 105 represent diseased tissue. Box 106 may use the same or different information from the at least one image 102 as box 104. Box 104 and box 106 may use different images corresponding to the same patient. Box 106 may output a set of diseased elements 108 that represent a subset of the candidate elements 105. The diseased elements 108 may represent areas in the image that correspond to diseased tissue of the subject.

[0029] Although method 100 shows blocks 104 and 106 working together, it should be understood that each block can be used independently. For example, a clinician can determine a candidate region of interest by some other method, and then use the region as an input to block 106 for filtering. Similarly, a clinician can analyze a set of images using block 104, and then use the resulting candidate elements 105 without continuing to block 106.

[0030] Method 100 may involve feedback from an operator (e.g., a doctor or technician), or may be automatically operated in a user-independent manner. Candidate elements 105 may be presented to an operator for review. Before box 106, the operator may be required to accept the located candidate elements 105 or edit them. Method 100 may display candidate elements 105 and / or diseased elements 108 as results to the operator. The operator may review, verify and / or edit the results. The operator may manage patient health based on the results. As an example, further tests may be arranged based on candidate elements 105, or treatment may be managed based on the state of diseased elements 108. Method 100 may be used to automatically manage patient health using an automatic health management system (e.g., a dose calculator) using candidate elements 105 and / or diseased elements 108. Therefore, the identification of one or more diseased elements may enable diagnosis-e.g., cancer, inflammation, tumors. In response to the diagnosis, various drugs may be administered to the patient, and / or treatment (e.g., biopsy, chemotherapy, dialysis, etc.) may be performed for the patient.

[0031] Figure 2 2 is a schematic diagram of a system for determining a disease state of a tissue. The system 200 can be used to perform a method for determining a disease state of a tissue, such as Figure 1The method 100 of the present invention is disclosed. The system 200 includes a subject (e.g., a patient) 201, a candidate location (or region) 205', a diseased location (or region) 208', an imaging unit 210, and an operating unit 212. The operating unit 212 may include an input / output unit (e.g., a display) 214, a controller 216, a processor 218, and a memory 220. The imaging unit 210 may generate an image 202, which may be sent to the operating unit 212 to be stored in the memory 220 and / or displayed on the input / output unit 214. The image 202 includes a candidate element 205 and a diseased element 208. The memory 220 may include instructions 222 and an atlas 224, the instructions including instructions for locating the candidate element 205 and filtering the candidate element 206.

[0032] The subject 201 may include a candidate region 205' and / or a diseased location 208' in the subject 201. The imaging unit 210 may scan or image all or part of the subject 201 to generate data that is sent to the operation unit 212 to form the image 202. The operation unit 212 may include an input / output unit 214, a controller 216, a processor 218, and a memory 220. The memory stores instructions 222 that can be accessed to enable the processor 218 to perform certain operations. The processor 218 and / or the memory may receive data from the imaging unit 210 to generate the image 202. The processor 218 may execute the instructions 222 of the memory, such as locating the candidate elements 204 and / or filtering the candidate elements 206, to generate the candidate elements 205 and the diseased elements 208 of the image 202. The candidate elements 205 of the image 202 may correspond to the candidate locations 205' of the subject 201, and the diseased elements 208 of the image 202 may correspond to the diseased locations 208' of the subject 201. When executing one or more instructions 222 , processor 218 may access atlas 224 .

[0033] Subject 201 can be a mammal, such as a human. Subject 201 can be a patient who exhibits symptoms or is receiving treatment and / or monitoring of a disease or condition. Subject 201 can be a member of the general population, or can be asymptomatic. Subject 201 can have one or more candidate locations 205' and / or one or more diseased areas 208'. Areas 205', 208' can be located on the outer surface of subject 201 (e.g., on the skin), or can be located inside. The one or more diseased areas 208' can be a subset of the one or more candidate locations 205'. Candidate location 205' can be an area of ​​tissue with an attribute that falls outside the normal clinical range of the tissue or with sufficient probability to fall outside the normal clinical range of the tissue. The candidate location can also be detected using an artificial intelligence algorithm. The one or more diseased areas 208' can be a location containing a structure that is harmful to subject 201 or may be harmful to subject 201 in the future. In one example, diseased area 208' can be a cancer cell area, such as a tumor. Other diseased tissues or conditions can be studied in other examples.

[0034] The imaging unit 210 can generate one or more images (such as image 202) of the subject 201. The imaging unit can generate a single image or multiple images of the subject 201. The imaging unit 210 can generate images of all or part of the subject 201. The imaging unit 210 can include multiple imaging modes, such as MRI, ultrasound, optical imaging, x-ray, CT, PET, or a combination thereof. In some cases, multiple imaging modes can be physically installed in different locations, and the patient can be transported to the different locations to complete a separate imaging session. The imaging unit 210 can generate a full body scan of the subject 201. The imaging unit can generate quantitative images, non-quantitative images, or a combination. The imaging unit 210 can be coupled to the operating unit 212 via wired and / or wireless communication. The imaging unit 210 can send raw data to the operating unit 212, or can send processed data to the operating unit 212. The imaging unit 210 and the operating unit 212 can be located away from each other. The data can be stored by the imaging unit 210 on a non-transitory medium (such as a computer-readable medium) and retrieved later by the operating unit 212.

[0035] The image 202 generated by the imaging unit 210 can be used as Figure 1The image 102 of the subject 201 can be stored, displayed and / or analyzed by the operating unit 212. The image 202 can represent the subject 201. The image can include elements such as pixels, voxels, groups of voxels corresponding to anatomical structures, surfaces, or combinations thereof. The image 202 can include a quantitative image, wherein the value (e.g., intensity) of each element of the image 202 has an absolute value related to the property of the subject 201 at the position corresponding to the element. The image can be image processed by the processor 218 and / or the imaging unit 210. The image processing can include correction for distortion or artifacts, normalization of values, or combination. The values ​​of the elements of the image 202 can be normalized so that the values ​​are, for example, between 0 and 255. Other ranges can be used in other examples. The imaging unit 210 can be calibrated. The calibration can involve using the imaging unit 210 to generate an image of a tissue phantom (not shown) with known properties.

[0036] The operating unit 212 receives data from the imaging unit 210. The operating unit 212 can be a computer. The operating unit 212 generates the image 202 using the data from the imaging unit 210. The data can be saved in the memory 220. The controller 216 can cause the processor 218 to render the image 202 to the input / output 214. In some examples, the input / output 214 can be a display, such as a computer monitor. The controller 216 can also send instructions to the imaging unit 210 to control various aspects of the imaging of the subject 201. The controller 216 can cause the processor 218 to execute the instructions 222 stored in the memory.

[0037] Instructions 222 in memory 220 may be used to analyze image 202. Instructions 222 may include instructions for locating candidate elements 204 (e.g., Figure 1 Instructions 222 may include instructions for filtering candidate elements 206 (e.g., Figure 1 The instructions 204, 206 may be run sequentially on the same image 202, or may be run as separate steps. The instructions 204, 206 may generate candidate elements 205 and / or diseased elements 208 for the image 202. The input / output 214 may display the image 202 and one or more of the candidate elements 205 and the diseased elements 208. The memory 220 may store additional information, such as an atlas 224 that may be used in one or more of the instructions 222.

[0038] Figure 3304 is a flow chart of a method 304 of locating candidate elements arranged according to an embodiment of the present disclosure. The example method may include one or more operations, functions, or actions, as shown in one or more of blocks 302, 302a, 302b, 324, 326, 327, 328, 330, 332, and / or 334. The operations described in blocks 302 to 334 may be performed in response to executing (such as by one or more processors described herein) computer-executable instructions stored in a computer-readable medium (such as a computer-readable medium of a computing device or some other controller of similar configuration).

[0039] Method 304 includes at least one image 302, the at least one image including a quantitative image 302a. Image 302 may also include additional images 302b, which may be quantitative images, non-quantitative images, or a combination. The method also includes an atlas 324, which includes atlas elements 326 and a metric distribution 327. The example process may begin at box 328, which states "determine corresponding mappings" and accepts image 302 and atlas 324 as inputs. Box 328 may be followed by box 330, which states "assign metric distributions to quantitative image elements." Box 330 may be followed by box 332, which states "locate candidate elements." Box 332 may be followed by box 334, which states "store candidate elements or filter."

[0040] Method 304 may be a detailed method for locating candidate elements, such as Figure 1-2 Methods 104 and 204. Image 302 may be associated with Figure 1 Image 102 and Figure 2 302a and 302b. The quantitative image 302a and 302b are similar or identical to the image 202 of the present invention. The image 302 may include a mixture of a quantitative image 302a and another image 302b. The quantitative image 302a and the other image 302b are both composed of elements as described herein. The quantitative image 302a has elements with absolute values, for example, the values ​​of the elements are related to the physical or functional properties of the image at the point. The quantitative image 302a may be calibrated. The quantitative image 302a may include quantitative MRI, quantitative PET, quantitative ultrasound, or a combination. Examples of quantitative images include whole-body diffusion weighted imaging MRI (DWI MRI). Examples of non-quantitative images include short TI inversion recovery (STIR) MRI, T1-weighted MRI images, T2-weighted MRI images, CT, PET, single photon emission computed tomography (SPECT), ultrasound, or a combination.

[0041] Atlas 324 is a representation of average values ​​from a normal patient population. Atlas 324 can be constructed from one or more images previously acquired from a normal patient population, and can be artificially generated or combined from statistics or generalizations about the normal patient population. Atlas 324 can contain a single composite image of a common or representative patient, or multiple such images. Atlas 324 can represent all normal patients, or it can represent a subset of the normal patient population. The subset can be limited by one or more of, for example, gender, age, weight, height, body shape, or other factors. Atlas 324 can be selected or filtered to correspond to the subject of image 302 (e.g., Figure 2 In one example, if the subject is a 15-year-old female, the atlas may be filtered by "female" and "teenager", and an atlas may be generated that averages only patient scans of female teenagers. Other filters may be used in other examples.

[0042] Atlas 324 may include atlas elements 326 and metric distributions 327. Atlas elements 326 may each be associated with one or more of metric distributions 327. Atlas elements 326 may be image pixels, image voxels, segmented surfaces, reference points, reference structures, or combinations thereof. Metric distributions 327 may be a range of "normal" values ​​for each atlas element 326. Metric distributions 327 may include intensity values ​​or values ​​calculated from intensity values, such as average intensity values ​​or texture values. In an embodiment where quantitative image 302a is a DWI MRI image, metric distributions 327 are ranges of expected diffusion values ​​for each atlas element 326. Other quantitative images and metric distributions may be used in other instances. Metric distributions 327 may be normalized values. Metric distributions 327 may be a probability density distribution range for intensity values, a mean value and standard deviation for intensity values, a range of expected texture values, or a combination thereof. Atlas 324 may include multiple images, each of which includes atlas elements with metric distributions.

[0043] Box 328 describes "determine corresponding mapping", and accepts input including quantitative image 302a and atlas 324. During box 328, each element of quantitative image 302a is matched with one or more of atlas elements 326. Atlas element 326 can be mapped to an element of quantitative image 302a, or the element of quantitative image 302a can be mapped to atlas element 326. Mapping can be adjusted for the difference between the size and shape of one or more normal patients in atlas 324 and the size and shape of the subject of quantitative image 302a. In one embodiment, corresponding mapping can be performed by using a deformable registration algorithm. In another embodiment, quantitative image 302a can be segmented to divide the quantitative image into one or more structural elements, and atlas elements 326 can be defined as structural elements. Quantitative image 302a elements can be matched with atlas elements 326 by comparing the size, morphology and / or positioning of the elements. For example, a lymph node chain may be approximately 2 cm in size in the neck and approximately 1 cm in size in the chest, which may be combined with morphological information to identify corresponding structures in the image. Other methods of determining the corresponding mapping may be used in other examples, such as using artificial intelligence algorithms.

[0044] Figure 4 is a schematic diagram of a corresponding mapping, which can be used as Figure 3 The corresponding mapping of box 328. Figure 4 The atlas 424 having atlas elements 426a and the image 402 having mapped atlas elements 426b are included. Figure 4 In the example of , atlas element 426a represents a structure being mapped onto a corresponding element of image 402. As indicated by the arrows, atlas element 426a undergoes a change in size and shape as it is mapped onto image 402 to become mapped element 426b. Each group of corresponding elements 426a, 426b may have a distribution of metrics corresponding to a structure, tissue type, or other characteristic common to the group.

[0045] Reference Figure 3 After determining the corresponding mapping, the method 304 proceeds to block 332, which states "Locate candidate elements". The elements of the quantitative image 302a are compared to the metric distribution 327 of the corresponding atlas element 326 determined by the corresponding mapping. The comparison may involve a direct comparison (such as comparing a value to a threshold), or may involve a machine learning, deep learning, or other artificial intelligence algorithm. In some embodiments, a user (e.g., a clinician) may provide input to the system in order to locate the candidate elements.

[0046] Figure 5 A schematic diagram of a metric distribution is depicted, which can be Figure 3An instance of a positioning candidate element of box 332. Figure 5 The example comprises elements 536a-c, which may be elements of a quantitative image, and ranges 527a-c, which may be metric distributions of an atlas. In this example, the comparison of the value of the element of the quantitative image with the metric distribution of the corresponding atlas element is accomplished by determining whether the value of the quantitative image element falls within the range of values ​​of the metric distribution. Figure 5 Three elements 536a-c are depicted, each of which is associated with a separate range of values ​​527a-c. Elements 536a and 536c have values ​​that fall within ranges 527a and 527c, respectively, and would be considered normal (e.g., not a candidate element). Element 536b falls outside of range 527b and would not be considered normal. Element 536b may be reported as a candidate element (e.g., Figure 1 In other examples, other methods of comparing the metric distribution with the values ​​of the quantitative image may be used.

[0047] Reference Figure 3 The method 304 involves block 334, which states "Store candidate elements or filter". The located candidate elements can be stored in a memory (such as Figure 2 The candidate elements may be further processed. The coordinates or centroids of the candidate elements may be identified. The candidate elements may be grouped together into unified elements. Figure 2 The system 200 of can display the candidate elements after positioning. The candidate elements can be displayed as an overlay on the image 302.

[0048] Figure 6 Flowchart of a method 606 for filtering candidate elements arranged according to an embodiment of the present disclosure. The example method may include one or more operations, functions, or actions, as shown in one or more of blocks 602, 602a, 602b, 304, 640, 642, 644, 646, 608. The operations described in blocks 602 to 646 may be performed in response to executing (such as by one or more processors described herein) computer-executable instructions stored in a computer-readable medium (such as a computer-readable medium of a computing device or some other controller of similar configuration).

[0049] Method 606 includes at least one image 602, the at least one image including a quantitative image 602a and an additional image 602b. The example process may begin at box 640, which states "locate corresponding candidate elements in additional images" and accepts as input the additional image 602b and the results of box 304 "locate candidate elements". Box 640 may be followed by box 642, which states "classify corresponding candidate elements". Box 642 may be followed by box 644, which states "determine the disease state of the candidate elements". Box 644 accepts as input box 646 which states "logical relationship". Box 644 may output diseased elements 608.

[0050] Method 606 can be used to Figure 1 and 2 The method 606 may be a continuation of the method 304, or may be an independent method. Similarly, the image 602 including the quantitative image 602a and the additional image 602b may be respectively Figure 1-3 The frame 304 is shown as stating "Positioning Candidate Elements" and may represent Figure 3 Method 606 may also be used alone for a new image 602 and a set of candidate elements determined by a different process (eg, identified by a clinician).

[0051] Box 640 states "Locate corresponding candidate elements in another image". During this step, the candidate elements identified in the quantitative image 602a (such as by box 304) are located in the corresponding other image 602b. The other image 602b can be a quantitative image or a non-quantitative image. The other image 602b can be composed of elements similar to the elements of the quantitative image 602a. The candidate elements can be individual elements or groups of elements pre-identified for further review. In one instance, the coordinates of the candidate elements can be matched with the coordinates of the elements in the other image 602b. In another example, the quantitative image 602a can be registered to the other image 602b using a non-rigid registration algorithm. In yet another example, the common elements of the images 602a, 602b can be identified by segmenting the structures in each image. Other positioning methods can be used in other examples.

[0052] Box 642 states "classify the corresponding candidate element". The attributes of the additional image 602b corresponding to the candidate element can be assigned to different categories. The categories can be binary, such as "dark" and "bright", or "hard" and "soft". Categories may also involve more than two elements, such as "dark", "bright" and "undetermined" or multiple discrete values ​​assigned (for example, brightness in the range of 1-10). The category of a given candidate element can be determined by comparing the mean, standard deviation or texture value within the candidate element with the average surrounding value. The comparison can be to determine whether the value within the candidate element is higher or lower than the average of the surrounding elements. In some instances, a machine learning algorithm can be used to assign categories to the candidate elements.

[0053] Box 644 states "Determining the disease state of a candidate element." Based on the category applied to the candidate element, the candidate element can be assigned a diseased or non-disease state. Other states can also be determined, such as, for example, the type of disease. Logical relationships 646 can be applied to the categories of candidate elements from non-quantitative image 602b. Logical relationships 646 can be specific to the type and / or application of image 602. Logical relationships 646 can reflect a logical understanding of what circumstances may cause an element of quantitative image 602a to be identified as a candidate element. Logical relationships 646 can compare the categories of additional images 602b, the values ​​of quantitative image 602a, or both. Logical relationships 646 can be changed to reflect different image types, user preferences, updated scientific knowledge, or other factors.

[0054] Table 1 gives an example of a logical relationship. In this example, there is a single quantitative image (DWI) and two additional non-quantitative images (STIR and T1). Candidate elements are located by examining the DWI images of elements with abnormal water movement. For the purpose of this example, abnormal water movement can be considered to be caused by the presence of fluid or cysts, blood (such as in fat or bone marrow), or tumors. When there is fluid or cysts, the STIR image is bright, and when there is a tumor, the STIR image is dark. When there is fat / bone marrow, the T1 image is bright, and when there is a tumor, the T1 image is dark. The columns of Table 1 show possible imaging modes. The rows of Table 1 show the diagnosis based on the results of each of the three imaging modes. In Table 1, 'X' indicates that the classification of the column is not important for diagnosis and may not need to be considered. Therefore, when a given candidate element is dark in both STIR and T1, a tumor can be diagnosed. When the candidate element in the STIR image is bright, it is identified as fluid and / or cyst, and when the T1 image is bright, it is identified as blood. Other logical relationships can be established for other examples.

[0055] Table 1. Logical relationships

[0056] diagnosis DWI STIR T1 Tumor Candidate dark dark Fluid and / or cysts Candidate Bright X blood Candidate X Bright

[0057] In this way, examples described herein can utilize quantitative images to identify candidate elements using comparisons with atlases (e.g., average image values ​​from a normal population). The candidate elements can then be located in other associated images (e.g., non-quantitative images). Binary classification (e.g., bright / dark) can be used to evaluate the candidate elements in the non-quantitative images to achieve identification of diseased elements.

[0058] Method 606 accordingly outputs diseased elements 608. Diseased elements 608 are elements of image 602 that are identified as corresponding to diseased tissue (e.g., Figure 2 Diseased element 608 may reflect tissue that is currently diseased, or tissue that may become diseased. Method 606 may be modified (e.g., by specifying logical relationship 646) to filter out other elements of interest that are not diseased tissue. The state of each candidate element may be stored, for example, in a memory (e.g., Figure 2 A report may be generated to identify elements that method 606 considers to be diseased.

[0059] Figure 7 is a flow chart of a method for locating and filtering candidate elements according to the present disclosure. The example method may include one or more operations, functions, or actions, as shown in one or more of blocks 748, 750, 704, 740, 742, 752, and diseased elements 708. The operations described in blocks 704 to 752 may be performed in response to executing (such as by one or more processors described herein) computer executable instructions stored in a computer readable medium (such as a computer readable medium of a computing device or some other controller of similar configuration).

[0060] The method may include a box 748, which states "acquire at least one quantitative image". The method may also include a box 750, "acquire at least one additional image". Box 704 may be after box 748 and state "locate candidate elements in the quantitative image". Box 740 may be after box 750 and box 704 and state "co-locate candidate elements in the additional image". Box 742 may be after box 740 and state "classify candidate elements in the additional image". Box 752 may be after boxes 704 and 742 and state "filter candidate elements". Diseased elements 708 may be output from box 752.

[0061] The method can be similar to other methods for determining diseased elements described herein. In the method, the location of candidate elements and the filtering of diseased elements occur as part of a single process. The method can use Figure 2 The system 200 and other systems are used for execution.

[0062] Blocks 748 and 750 state "acquire at least one quantitative image" and "acquire at least one additional image", respectively. The images may be acquired by an imaging unit, such as Figure 2 The imaging unit 210. The image may be acquired in advance and may be stored until called by the method. The image may be, for example, Figure 3 The quantitative image 302a and the additional image 302b of FIG. 750. The additional images acquired in block 750 may be quantitative or non-quantitative. The images may correspond to the same subject. The images may include a plurality of different imaging modes.

[0063] Block 704 states "Locate candidate elements in the quantitative image". As discussed herein, methods may be used to identify candidate elements of the quantitative image. These methods may include locating the quantitative element to an atlas, and comparing the element of the quantitative image to the atlas elements, such as Figure 3 As described in method 304.

[0064] Block 740 states "Co-localize candidate elements in additional images". The candidate elements determined in block 704 are localized in additional images acquired in block 750. The localization may be substantially similar to Figure 6 Positioning of box 640.

[0065] Block 742 states "Classify candidate elements in the additional image". Based on the attributes of the additional image, the candidate elements of the additional image are grouped into a plurality of categories. The classification may be generally similar to Figure 6 Classification of box 642.

[0066] Box 752 states "Filter candidate elements". In this step, the attributes of the candidate elements in the quantitative image and the additional image are compared to determine the disease state of each of the candidate elements. The values ​​of the candidate elements in the quantitative image can be compared with one or more categories of candidate elements in the additional image. Logical relationships can be used to determine the disease state based on the attributes of the candidate elements. The output of the method can be a set of diseased elements 708.

[0067] Depend on Figure 1-7One or more output diseased elements in can be used to manage the health of the subject. The diseased elements can be used to diagnose the subject. For example, the viability of the subject can be assessed based on the diseased elements. Subsequent testing can be arranged based on the diseased elements. For example, a biopsy can be performed on the subject at a location corresponding to the diseased element. The diseased elements can be used (alone or together with additional information) to guide and / or monitor the course of treatment. For example, a drug can be delivered to a location on the subject corresponding to the diseased element, or surgery can be performed at the location. Parameters of treatment can be determined based on the diseased elements. For example, the dose of the drug can be determined by locating the volume of the diseased element in the image. The method can be repeated on the same subject over time to monitor the state of the diseased element, such as monitoring disease progression.

[0068] The present disclosure is not limited to the specific examples described in this application, and the examples are intended to serve as illustrations of various aspects. It is obvious to those skilled in the art that many modifications and examples can be implemented without departing from the spirit and scope of the present technology. According to the above description, in addition to the methods and equipment listed herein, functionally equivalent methods and equipment within the scope of the present disclosure will be obvious to those skilled in the art. These modifications and examples are intended to fall within the scope of the appended claims. The present disclosure is only limited by the terms of the appended claims and the full range of equivalents given by these claims. It should be understood that the present disclosure is not limited to specific methods, reagents, compounds, compositions, or biological systems, and of course these can change. It should also be understood that the terms used herein are only for the purpose of describing specific examples and are not intended to be restrictive.

[0069] With respect to the use of substantially any plural terms and / or singular terms herein, those skilled in the art may convert from the plural to the singular and / or from the singular to the plural, depending on the context and / or application. For clarity, various singular / plural arrangements may be expressly set forth herein.

[0070] Those skilled in the art will understand that, in general, the terms used herein, especially in the appended claims (e.g., the bodies of the appended claims), are generally intended to be "open" terms (e.g., the term "including" should be interpreted as "including but not limited to", the term "having" should be interpreted as "having at least", the term "includes" should be interpreted as "including but not limited to", etc.).

[0071] It will be further understood by those skilled in the art that if a specific number of an introduced claim statement is intended, such intent will be explicitly stated in the claim, and in the absence of such a statement, such intent is not present. For example, to aid understanding, the following appended claims may contain the use of the introductory phrases "at least one" and "one or more" to introduce claim statements. However, the use of such phrases should not be interpreted as implying that a claim statement introduced by the indefinite article "a" or "an" limits any particular claim containing such introduced claim statement to only one instance of such statement, even when the same claim contains the introductory phrases "one or more" or "at least one" and an indefinite article such as "a" or "an" (e.g., "a" and / or "an" should be interpreted as meaning "at least one" or "one or more"); the same applies to the use of definite articles used to introduce claim statements. Furthermore, even if a specific number of an introduced claim statement is explicitly stated, one skilled in the art will recognize that such statement should be interpreted to mean at least the stated number (e.g., merely stating "two statements" without other modifiers means at least two statements or two or more statements).

[0072] Furthermore, where a convention similar to “at least one of A, B, C, etc.” is used, generally speaking, such construction is intended to be understood by those skilled in the art (e.g., “a system having at least one of A, B, and C” would include, but not be limited to, systems having A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). Where a convention similar to “at least one of A, B, or C, etc.” is used, generally speaking, such construction is intended to be understood by those skilled in the art (e.g., “a system having at least one of A, B, or C” would include, but not be limited to, systems having A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). Those skilled in the art will further understand that virtually any discrete word and / or phrase presenting two or more alternative terms, whether in the specification, claims, or drawings, should be understood to contemplate the possibility of including one of the terms, either of the terms, or both of the terms. For example, the phrase "A or B" will be understood to include the possibilities of "A" or "B" or "A and B."

[0073] In addition, where features or aspects of the disclosure are described in terms of Markush groups, those skilled in the art will recognize that the disclosure is also thereby described in terms of any individual member or subgroup of members of the Markush group.

[0074] As will be understood by those skilled in the art, for any and all purposes, such as in terms of providing a written description, all ranges disclosed herein also include any and all possible sub-ranges and combinations of sub-ranges thereof. Any listed range can be easily identified as fully describing and making the same range be decomposed into at least equal two, three, four, five, ten, etc. As a non-limiting example, each range discussed herein can be easily decomposed into a lower third, a middle third, and an upper third, etc. As will be understood by those skilled in the art, all languages ​​such as "at most", "at least", "greater than", "less than", etc. contain the cited numbers and refer to the ranges that can be subsequently decomposed into sub-ranges as discussed above. Ultimately, as will be understood by those skilled in the art, the range includes each individual member. Therefore, for example, a group with 1-3 items refers to a group with 1, 2 or 3 items. Similarly, a group with 1-5 items refers to a group with 1, 2, 3, 4 or 5 items, etc.

[0075] Although the above specific embodiments have been described various examples of devices and / or processes by using block diagrams, flow charts and / or examples, such block diagrams, flow charts and / or examples include one or more functions and / or operations, and those skilled in the art will understand that each function and / or operation in such block diagrams, flow charts or examples can be implemented individually and / or collectively by a variety of hardware, software, firmware or almost any combination thereof. In one example, several parts of the subject matter described herein can be implemented by application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), digital signal processors (DSPs) or other integrated formats. However, those skilled in the art will recognize that some aspects of the examples disclosed herein can be implemented in whole or in part in an integrated circuit as one or more computer programs running on one or more computers (e.g., implemented as one or more programs running on one or more computer systems), implemented as one or more programs running on one or more processors (e.g., implemented as one or more programs running on one or more microprocessors), implemented as firmware or implemented as almost any combination thereof, and according to the present disclosure, designing circuits and / or writing codes for software and / or firmware will be completely within the technical scope of those skilled in the art. For example, if the user determines that speed and accuracy are most important, the user may select a tool that is primarily hardware and / or firmware; if flexibility is most important, the user may select a primarily software implementation; or, again optionally, the user may select some combination of hardware, software, and / or firmware.

[0076] Furthermore, those skilled in the art will appreciate that the mechanisms of the subject matter described herein can be distributed as a program product in a variety of forms, and that the illustrative examples of the subject matter described herein apply regardless of the specific type of signal-bearing medium used to actually perform the distribution. Examples of signal-bearing media include, but are not limited to, the following: recordable media such as floppy disks, hard drives, compact disks (CDs), digital video disks (DVDs), digital tapes, computer memory, etc.; and transmission-type media such as digital and / or analog communication media (e.g., fiber optic cables, waveguides, wired communication links, wireless communication links, etc.).

[0077] Those skilled in the art will recognize that it is common in the art to describe devices and / or processes in the manner set forth herein and then use engineering practices to integrate the devices and / or processes described in this manner into a data processing system. That is, at least a portion of the devices and / or processes described herein can be integrated into a data processing system by a reasonable amount of experimentation. Those skilled in the art will recognize that a typical data processing system typically includes a system unit housing, a video display device, a memory (such as volatile and non-volatile memory), a processor (such as a microprocessor and a digital signal processor), a computing entity (such as an operating system, a driver, a graphical user interface, and an application), one or more interactive devices (such as a touch pad or screen), and / or a control system (including feedback loops and control motors (e.g., feedback for sensing position and / or velocity; control motors for moving and / or adjusting components and / or quantities)). A typical data processing system can be implemented using any suitable commercially available components (such as components typically present in data computing / communication systems and / or network computing / communication systems).

[0078] The subject matter described herein sometimes shows different parts that are contained in different other parts or are connected with different other parts.It should be understood that the architecture described in this way is only an example, and many other architectures that realize the same function can actually be implemented.In a conceptual sense, any arrangement of parts for realizing the same function is effectively "associated" so that the desired function is realized.Therefore, any two parts that are combined to realize a specific function herein can be regarded as "associated" with each other so that the desired function is realized, regardless of the architecture or intermediate parts.Equally, any two parts so associated can also be regarded as "operably connected" or "operably coupled" to realize the desired function, and any two parts that can be so associated can also be regarded as "operably coupled" to realize the desired function.The specific examples that can be operably coupled include but are not limited to parts that can be physically paired and / or physically interact and / or parts that can wirelessly interact and / or wirelessly interact and / or parts that logically interact and / or logically interact.

[0079] Although various aspects and examples have been disclosed herein, other aspects and examples will be apparent to those skilled in the art. The various aspects and examples disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the appended claims.

Claims

1. A method for determining a disease state, the method comprising: include: Providing a quantitative image of a subject, wherein the quantitative image is a whole-body image of the subject, and the quantitative image includes structural elements; providing an atlas corresponding to the subject, the atlas comprising atlas elements, each of the atlas elements comprising a metric distribution of intensity values, each of the atlas elements corresponding to an anatomical structure; separating the quantitative image into a plurality of the structural elements, each of the structural elements corresponding to an anatomical structure in the whole-body image; matching each of the plurality of structural elements to the atlas element based on the size, form, and positioning of each of the corresponding structural elements; determining a corresponding mapping associating the structural elements of the quantitative image with the corresponding atlas elements by comparing the size, morphology and positioning of the structural elements of the quantitative image with the corresponding atlas elements; as well as Candidate elements are determined by comparing absolute intensity values ​​of certain structural elements of the quantitative image with the metric distributions of relevant atlas elements of the corresponding map.

2. The method according to claim 1, further comprising: include: providing at least one additional image of the subject, each of the at least one additional image comprising a structural element; locating the candidate element to a corresponding structural element of the at least one other image; classifying the candidate elements into at least one category based on attributes of the corresponding structural elements of the at least one further image; as well as The disease state of the candidate element is determined by analyzing the at least one category of the at least one further image corresponding to the candidate element. 3 . The method of claim 1 , wherein the elements comprise pixels, voxels, groups of voxels corresponding to anatomical structures, surfaces, points, annotations, or a combination thereof. The method of claim 1 , wherein the quantitative image comprises a whole-body diffusion-weighted MRI image. The method of claim 4 , further comprising calibrating the whole-body diffusion-weighted MRI image using a phantom. The method of claim 1 , wherein the atlas comprises pre-acquired quantitative data of a normal patient population.

7. The method of claim 1, wherein the atlas comprises pre-acquired quantitative data for a subset of a normal patient population, the subset being determined at least in part by attributes of the subject.

8. The method of claim 1, wherein the atlas elements include image pixels, image voxels, segmented surfaces, reference points, reference structures, annotations, or a combination thereof.

9. The method of claim 1, wherein the metric distribution comprises a range of expected diffusion values, a probability density distribution of intensity values, a mean and standard deviation of intensity values, a range of expected texture values, or a combination thereof for elements of the atlas.

10. The method of claim 1, wherein said determining said corresponding mapping comprises using a deformable registration technique.

Images