Adaptive diffusion-based metric for quantification of lung disease

Abstract

Disclosed are various embodiments directed to an adaptive diffusion-based metric for quantification of lung disease, an imaging biomarker, and methods of use thereof. The systems and methods described herein combine imaging techniques and image processing techniques with knowledge of structural impacts on disease progression in order to produce a model for predicting disease progression. A system can be arranged to capture and interpret images of diseased tissue in a patient. The system can capture images using a diagnostic imaging technique, identify affected voxels within the images, apply adaptive smoothing to the voxels surrounding the clusters, generate a weight map, and predict the progression of the disease.

Description

[0001] TITLE: ADAPTIVE DIFFUSION-BASED METRIC FOR QUANTIFICATION OF

[0002] LUNG DISEASE

[0003] Inventors: Surya Prakash Bhatt, Sandeep Bodduluri, and Arie Nakhmani

[0004] CROSS-REFERENCE TO RELATED APPLICATIONS

[0005] This application claims priority to and the benefit of U.S. Provisional Application entitled “ADAPTIVE DIFFUSION-BASED METRIC FOR QUANTIFICATION OF LUNG DISEASE” and having serial number 63 / 611 ,830, filed on December 19, 2023, which is incorporated herein by reference in its entirety.

[0006] BACKGROUND

[0007] Emphysema is a lung disease characterized by the breakdown of alveolar walls which results in larger air spaces within the lungs. With the degradation of the alveolar walls, the overall surface area inside the lungs decreases, resulting in a reduced capacity for the necessary gas exchanges to occur within the lungs. Additionally, air can get trapped in damaged tissue, which can further reduce the ability of the lungs to perform gas exchange. Emphysema can result in difficulty breathing, coughing, and a sensation of tightness in the chest.

[0008] As emphysema progresses, a patient can lose much lung function, as well as have an increased risk of downstream complications for numerous other bodily systems. Emphysema can lead to the development of bullae, or large holes in the lungs, as well as to an increased risk for pneumothorax. Additionally, emphysema can lead to heart problems over time since the heart will compensate for the lack of oxygenated blood coming from the lungs by pumping more blood through at a higher rate.

[0009] SUMMARY

[0010] In accordance with the purpose(s) of this disclosure, as embodied and broadly described herein, the disclosure, in various aspects, relates to an adaptive diffusion-based metric for quantification of lung disease, an imaging biomarker, and methods of use thereof. The solutions described herein combine imaging techniques and image processing techniques with knowledge of structural impacts on disease progression in order to produce a model for predicting disease progression.

[0011] Aspects of the present disclosure provide for an imaging system and a computing system to be used for capturing images and producing the model. Embodiments of the present disclosure include: an imaging system, a computing device having a processor and a memory, and machine-readable instructions stored in the memory that, when executed by the processor, cause the computing device to at least perform a number of actions. These actions can include capturing one or more images using the imaging system; identifying one or more clusters of affected voxels within the one or more images; applying smoothing to each of the one or more clusters of affected voxels; generating a weight map based at least in part on the smoothing; and predicting the progression of a disease based at least in part on the weight map. The instructions can be executed to further cause the computing device to segment the one or more images using an organ mask; apply a disease mask to identify a plurality of affected voxels; and apply a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels. In addition, the system can apply the disease mask to identify the plurality of affected voxels by labeling each voxel having a value of less than -950 HU as an affected voxel. The system can also cause the computing device to estimate a diameter of each of the one or more clusters; assign a weight to each voxel based at least in part on the diameter of the cluster and a distance from each of the one or more clusters; blur the voxels surrounding each of the one or more clusters based at least in part on the diameter of each of the one or more clusters; and sum the weight of each voxel to obtain a total weight. In some embodiments, the system can generate the weight map based at least in part on the total weight. The system can further identify vasculature near the one or more clusters of affected voxels and predict the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature. In some embodiments, the imaging system can be a computed tomography (CT) system, a magnetic resonance imaging (MRI) system, a positron emission tomography (PET) system, an ultrasound, an x-ray system, or any other imaging system as can be appreciated.

[0012] Other systems, methods, devices, features, and advantages of the devices and methods will be or become apparent to one with skill in the art upon examination of the following drawings and detailed description. It is intended that all such additional systems, methods, devices, features, and advantages be included within this description, be within the scope of the present disclosure, and be protected by the accompanying claims.

[0013] BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Further aspects of the present disclosure will be more readily appreciated upon review of the detailed description of its various embodiments, described below, when taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Moreover, in the drawings, like reference numerals designate corresponding parts throughout the several views.

[0015] FIG. 1 is a drawing of a network environment according to various embodiments of the present disclosure.

[0016] FIG. 2 is a flowchart illustrating one example of functionality implemented as portions of an application executed in network environment of FIG. 1 according to various embodiments of the present disclosure.

[0017] FIG. 3 shows a graphic representation of the spatial mapping methodology according to various embodiments of the present invention. Euclidean distance was calculated for each normal voxel to the nearest emphysema voxel in 3D, and the surrounding normal tissue was clustered into different groups based on their spatial distance to the nearest emphysema voxel. The lung distance color map represents distances of each normal voxel to the nearest emphysema voxel from 1 to 5 mm. The color red represents emphysema regions, labeled as 0 mm on the color scale.

[0018] FIGS. 4A-D are illustrations depicting correlations between different values according to various embodiments of the present disclosure. The top panel shows correlations between a 3D Mechanically Affected Lung and AFEVi (FIG 4A) as well as AEmphysema (FIG. 4B). The bottom panel shows correlations between %Convex Hull and AFEVi (FIG. 4C) and AEmphysema (FIG. 4D). Data is shown for 253 participants with GOLD stage 0 and 1 COPD.

[0019] FIGS. 5A-C are illustrations of a computation of Gaussian Weights for normal lung parenchyma according to various embodiments of the present disclosure. FIG. 5A shows a lung mask (blue boundary) and emphysema masks (white) overlaid on a CT slice; FIG. 5B shows a modeled mechanical effect; and FIG. 5C shows a modeled mechanical effect after excluding emphysema voxels.

[0020] FIG. 6 is an illustration of a color map of a 3D Mechanically Affected Lung surrounding areas of emphysema according to various embodiments of the present disclosure.

[0021] DETAILED DESCRIPTION

[0022] Before the present disclosure is described in greater detail, it is to be understood that this disclosure is not limited to particular embodiments described, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present disclosure will be limited only by the appended claims.

[0023] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit (unless the context clearly dictates otherwise), between the upper and lower limit of that range, and any other stated or intervening value in that stated range, is encompassed within the disclosure. The upper and lower limits of these smaller ranges may independently be included in the smaller ranges and are also encompassed within the disclosure, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure.

[0024] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Although any methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the present disclosure, the preferred methods and materials are now described.

[0025] As will be apparent to those of skill in the art upon reading this disclosure, each of the individual embodiments described and illustrated herein has discrete components and features which may be readily separated from or combined with the features of any of the other several embodiments without departing from the scope or spirit of the present disclosure. Any recited method can be carried out in the order of events recited or in any other order that is logically possible.

[0026] Embodiments of the present disclosure will employ, unless otherwise indicated, imaging and image processing techniques and the like, which are within the skill of the art. Such techniques are explained fully in the literature.

[0027] The following examples are put forth to provide those of ordinary skill in the art with a complete disclosure and description of how to perform the methods and use the compositions and compounds disclosed and claimed herein. Efforts have been made to ensure accuracy with respect to numbers (e.g., amounts, measurements, etc.), but some errors and deviations should be accounted for.

[0028] Before the embodiments of the present disclosure are described in detail, it is to be understood that, unless otherwise indicated, the present disclosure is not limited to particular materials, machines, computing processes, or the like, as such can vary. It is also to be understood that the terminology used herein is for purposes of describing particular embodiments only and is not intended to be limiting. It is also possible in the present disclosure that steps can be executed in different sequence where this is logically possible.

[0029] All publications and patents cited in this specification are cited to disclose and describe the methods and / or materials in connection with which the publications are cited. Publications and patents that are incorporated by reference, where noted, are incorporated by reference as if each individual publication or patent were specifically and individually indicated to be incorporated by reference. Such incorporation by reference is expressly limited to the methods and / or materials described in the cited publications and patents and does not extend to any lexicographical definitions from the cited publications and patents. Any lexicographical definition in the publications and patents cited that is not also expressly repeated in the instant application should not be treated as such and should not be read as defining any terms appearing in the accompanying claims. Any terms not specifically defined within the instant application, including terms of art, are interpreted as would be understood by one of ordinary skill in the relevant art; thus, is not intended for any such terms to be defined by a lexicographical definition in any cited art, whether or not incorporated by reference herein, including but not limited to, published patents and patent applications. The citation of any publication is for its disclosure prior to the filing date and should not be construed as an admission that the present disclosure is not entitled to antedate such publication by virtue of prior disclosure. Further, the dates of publication provided could be different from the actual publication dates that may need to be independently confirmed.

[0030] It should be noted that ratios, amounts, and other numerical data can be expressed herein in a range format. It is to be understood that such a range format is used for convenience and brevity, and thus, should be interpreted in a flexible manner to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. To illustrate, a numerical range of “about 0.1 % to about 5%” should be interpreted to include not only the explicitly recited values of about 0.1 % to about 5%, but also include individual values (e.g., 1%, 2%, 3%, and 4%) and the sub-ranges (e.g., 0.5%, 1.1 %, 2.2%, 3.3%, and 4.4%) within the indicated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure, e.g., the phrase “x to y” includes the range from ‘x’ to ‘y’ as well as the range greater than ‘x’ and less than ‘y’. The range can also be expressed as an upper limit, e.g., ‘about x, y, z, or less’ and should be interpreted to include the specific ranges of ‘about x’, ‘about y’, and ‘about z’ as well as the ranges of ‘less than x’, less than y’, and ‘less than z’. Likewise, the phrase ‘about x, y, z, or greater’ should be interpreted to include the specific ranges of ‘about x’, ‘about y’, and ‘about z’ as well as the ranges of ‘greater than x’, greater than y’, and ‘greater than z’. In some embodiments, the term “about” can include traditional rounding according to significant figures of the numerical value. In addition, the phrase “about ‘x’ to ‘y’”, where ‘x’ and ‘y’ are numerical values, includes “about ‘x’ to about ‘y’”.

[0031] Discussion

[0032] Disclosed are various approaches for predicting the progression of disease using various imaging and image processing techniques to generate a quantitative and predictive model. While the Example herein is focused on predicting emphysema progression based on CT images of lungs, the principles of the present disclosure can also be extended to various other diseased tissues and can utilize various other forms of images. The approaches disclosed herein provide a new means for quantifying the progression of disease from diseased tissues to healthy tissues. In the example of emphysema, as well as numerous other diseases, some individuals diagnosed with the disease will experience slow progression over many years while others will experience a very rapid progression which could lead to life-threatening complications. For many diseases, the progression or spread of disease from affected tissues to healthy tissues can be a factor of proximity - e.g., healthy tissues which are closer in proximity to affected tissues are often more likely to become affected by the disease sooner than healthy tissues which are farther from the affected tissues. In the example of emphysema, data has shown that the spread of the disease can be related to the mechanical stress inflicted upon nearby healthy lung tissue from compensating for damaged tissues.

[0033] Current technologies for quantifying lung disease include parametric response mapping (PRM), a technique which pairs inspiratory and expiratory images voxel-by-voxel. PRM can provide a color map, which represents normal lung tissue, small airway disease, or emphysema, and gives insight into the extent and localization of disease. However, there is limited understanding of the mechanisms of progression of structural disease, and PRM does not account for mechanical influences of multiple surrounding emphysematous areas which may impact disease progression. Additionally, there is limited understanding of whether the distribution of emphysema affects disease progression. Prior studies show that predominant upper lobe emphysema is associated with faster progression than lower lobe emphysema, but studies examining spatial distributions at the regional and voxel level are lacking. If the mechanical effects of emphysema result in progression of emphysema, then the distribution of emphysema should also be important in disease progression, as more widespread distribution of a given degree of emphysema should result in a larger penumbra of lung at-risk for progression.

[0034] Accordingly, various embodiments of the present disclosure are directed to systems and methods for modeling and quantifying lung disease progression which accounts for emphysema surrounding a normal voxel in three dimensions. To do this, a system can be arranged to capture and interpret images of diseased tissue in a patient. The system can capture images using a diagnostic imaging technique, identify affected voxels within the images, apply adaptive smoothing to the voxels surrounding the clusters, generate a weight map, and predict the progression of the disease.

[0035] In the following discussion, a general description of the system and its components is provided, followed by a discussion of the operation of the same. Although the following discussion provides illustrative examples of the operation of various components of the present disclosure, the use of the following illustrative examples does not exclude other implementations that are consistent with the principles disclosed by the following illustrative examples.

[0036] With reference to FIG. 1 , shown is a network environment 100 according to various embodiments. The network environment 100 can include a computing environment 103, an imaging system 106, and a client device 109, which can be in data communication with each other via a network 113.

[0037] The network 113 can include wide area networks (WANs), local area networks (LANs), personal area networks (PANs), or a combination thereof. These networks can include wired or wireless components or a combination thereof. Wired networks can include Ethernet networks, cable networks, fiber optic networks, and telephone networks such as dial-up, digital subscriber line (DSL), and integrated services digital network (ISDN) networks. Wireless networks can include cellular networks, satellite networks, Institute of Electrical and Electronic Engineers (IEEE) 802.11 wireless networks ( / .e., WIFI®), BLUETOOTH® networks, microwave transmission networks, as well as other networks relying on radio broadcasts. The network 113 can also include a combination of two or more networks 113. Examples of networks 113 can include the Internet, intranets, extranets, virtual private networks (VPNs), and similar networks.

[0038] The computing environment 103 can include one or more computing devices that include a processor, a memory, and / or a network interface. For example, the computing devices can be configured to perform computations on behalf of other computing devices or applications. As another example, such computing devices can host and / or provide content to other computing devices in response to requests for content.

[0039] Moreover, the computing environment 103 can employ a plurality of computing devices that can be arranged in one or more server banks or computer banks or other arrangements. Such computing devices can be located in a single installation or can be distributed among many different geographical locations. For example, the computing environment 103 can include a plurality of computing devices that together can include a hosted computing resource, a grid computing resource, or any other distributed computing arrangement. In some cases, the computing environment 103 can correspond to an elastic computing resource where the allotted capacity of processing, network, storage, or other computing-related resources can vary over time. Various applications or other functionality can be executed in the computing environment 103. The components executed on the computing environment 103 include a processing service 116 and other applications, services, processes, systems, engines, or functionality not discussed in detail herein.

[0040] The processing service 116 can be executed to process images 119 obtained from the imaging system 106. The processing service 116 can be configured to obtain images 119 from the imaging system 106, use various processing techniques to identify diseased tissues within each image 119. For example, the processing service 116 can use one or more imaging masks to identify relevant portions of an image 119 and reduce or eliminate background noise. By estimating the size and location of the diseased tissues, the processing service 116 can assign a weight to each of the nearby voxels in an image 119 and blur each cluster based at least in part on the assigned weight. The processing service 116 can be executed to generate a weight map and use the weight map to predict disease progression.

[0041] Also, various data is stored in a data store 123 that is accessible to the computing environment 103. The data store 123 can be representative of a plurality of data stores 123, which can include relational databases or non-relational databases such as object- oriented databases, hierarchical databases, hash tables or similar key-value data stores, as well as other data storage applications or data structures. Moreover, combinations of these databases, data storage applications, and / or data structures may be used together to provide a single, logical, data store. The data stored in the data store 123 is associated with the operation of the various applications or functional entities described below. This data can include images 119 and potentially other data. The images 119 can represent files which contain graphic data, such as CT scans, MRI scans, x-rays, or other medical image. The images 119 can be many different file types, such as an ITK Metalmage, Neuroimaging Informatics Technology Initiative (NlfTI), Analyze, Digital Imaging and Communications in Medicine (DICOM), Nearly Raw Raster Data (Nrrd), Medical Imaging NetCDF (MINC), Guys Image Processing Lab (GIPL), Medical Research Council (MRC), Bio-Rad (PIC), LSM (Zeiss) microscopy images, Stimulate / Signal Data (SDT), Join Photographic Experts Group (JPEG / JPG), bitmap (BMP), Portable Network Graphics (PNG), Tagged Image File Format (TIFF), Hierarchical Data Format (HDF5), Photoshop Document (PSD), Portable Document Format (PDF), a RAW format, or other file types. In some embodiments, the images 119 can be stored as black-and-white, grayscale, or color images 119, such as cyan, yellow, magenta, and key (CYMK) or red, green, and blue (RGB) images 119.

[0042] The imaging system 106 can be representative of a device used for the purpose of capturing images 119 of a patient. In some embodiments, the imaging system 106 can be an CT machine, an MRI machine, an x-ray machine, or another type of diagnostic imaging system. The imaging system 106 can include a computing device configured to execute an imaging application 126. The imaging application 126 can be executed to control the different mechanisms within the imaging system 106 to capture different images 119. In some embodiments, the imaging application 126 can send the captured images 119 to a data store 123, the processing service 116, or another system, service, or application in the network environment 100.

[0043] The client device 109 is representative of a plurality of client devices that can be coupled to the network 113. The client device 109 can include a processor-based system such as a computer system. Such a computer system can be embodied in the form of a personal computer (e.g., a desktop computer, a laptop computer, or similar device), a mobile computing device (e.g., personal digital assistants, cellular telephones, smartphones, web pads, tablet computer systems, music players, portable game consoles, electronic book readers, and similar devices), media playback devices (e.g., media streaming devices, BluRay® players, digital video disc (DVD) players, set-top boxes, and similar devices), a videogame console, or other devices with like capability. The client device 109 can include one or more displays 129, such as liquid crystal displays (LCDs), gas plasma-based flat panel displays, organic light emitting diode (OLED) displays, electrophoretic ink (“E-ink”) displays, projectors, or other types of display devices. In some instances, the display 129 can be a component of the client device 109 or can be connected to the client device 109 through a wired or wireless connection.

[0044] The client device 109 can be configured to execute various applications such as a client application 133 or other applications. The client application 133 can be executed in a client device 109 to access network content served up by the computing environment 103 or other servers, thereby rendering a user interface 136 on the display 129. To this end, the client application 133 can include a browser, a dedicated application, or other executable, and the user interface 136 can include a network page, an application screen, or other user mechanism for obtaining user input. The client device 109 can be configured to execute applications beyond the client application 133 such as email applications, social networking applications, word processors, spreadsheets, or other applications.

[0045] Next, a general description of the operation of the various components of the network environment 100 is provided. Although the following description provides an example of the interactions between the various components of the network environment 100, other interactions or sequences of interactions are encompassed by the various embodiments of the present disclosure.

[0046] To begin, an imaging application 126 can be executed on an imaging system 106 to acquire one or more images 119 of a patient. Next, the imaging application 126 can send the images 119 to a processing service 116. Once the processing service 116 obtains the images 119, the processing service 116 can be executed to identify and isolate the target organ(s) or system(s) depicted in the images 119. The processing service 116 can isolate the targeted system using a mask. Next, the processing service 116 can be executed to identify the diseased or affected tissues within the targeted system. In some embodiments, the processing service 116 applies a mask to identify the affected tissues. Once the voxels representing the affected tissues have been identified, the processing service 116 can be executed to identify clusters of voxels representing the affected tissues and estimate the size and location of each cluster. Next, the processing service 116 can apply smoothing to the images 119 which, in some embodiments, can include assigning a weight to each voxel surrounding the clusters and blurring the voxels based at least in part on their assigned weights. The processing service 116 can use the results of the smoothing to generate a weight map. Once the weight map has been generated, the processing service 116 can be executed to predict disease progression.

[0047] Moving now to FIG. 2, shown is a flowchart that provides one example of the operation of a portion of the processing service 116. The flowchart of FIG. 2 provides merely an example of the many different types of functional arrangements that can be employed to implement the operation of the depicted portion of the processing service 116. As an alternative, the flowchart of FIG. 2 can be viewed as depicting an example of elements of a method implemented within the network environment 100.

[0048] Beginning with block 200, the processing service 116 can be executed to obtain or receive one or more images 119. The images 119 can be received from an imaging application 126 on an imaging system 106. In some embodiments, the processing service 116 can obtain the images 119 from a data store 123. The processing service 116 can obtain the images 119 in response to receipt of an initiation request sent from a client application 133, an interaction with a user interface 136, a notification from the imaging application 126 that images 119 have been captured, or some other trigger received by the processing service 116. In some embodiments, a trigger could be received by the processing service 116 after the images 119 have been stored in the data store 123. In these instances, the trigger could include an identifier of the images 119 to allow the processing service 116 to search for and obtain the images 119 from the data store 123.

[0049] Next, at block 203, the processing service 116 can be executed to segment the images 119. In some embodiments, the processing service 116 can segment images 119 using morphological operations, region-growing, machine learning-based classification of voxels, or other sophisticated algorithms. In some embodiments, the processing service 116 can segment the images 119 obtained at block 200 by thresholding the voxels in an image 119. The processing service 116 can segment the images 119 using a threshold value by which to measure the voxels in an image 119 to identify the structure of interest within the image 119. For example, the processing service 116 can use a threshold value to identify voxels on one side of the threshold as background and voxels on the other side of the threshold as a structure of interest. The structure of interest can be an organ, a tissue, a particular system, or other region of interest captured in the image. The mask can be applied to an image 119 to identify and isolate the structure of interest by removing the background or otherwise eliminating noise. In some embodiments, once the processing service 116 has segmented the images 119 to isolate the structure of interest, the segmented images 119 can be stored in a data store 123. In some embodiments, the segmented images 119 can be sent to a user interface 136 of a client device 109 for a user to review for accuracy or edit.

[0050] Next, at block 206, the processing service 116 can apply a disease mask to the images 119. Much like segmenting the images 119, application of a disease mask to the images 119 can be used to identify the diseased tissues within the images 119. The processing service 116 can apply the disease mask to reduce noise in an image 119 and, in some examples, label each voxel in an image 119 as affected or unaffected. According to various examples, the processing service 116 can use the disease mask to label voxels of less than -950HU as affected voxels, and voxels of greater than -950HLI as unaffected voxels. In some embodiments, the disease mask applied is unique to the disease that is being evaluated and is configured to identify patterns in an image 119 attributable to the disease. The processing service 116 can apply the disease mask to images 119 obtained at block 200, or images 119 which have been segmented at block 203.

[0051] Next, at block 209, the processing service 116 can be executed to identify clusters of affected or diseased tissues within the images 119. In some embodiments, the processing service 116 can apply a binary connected component algorithm to identify clusters within the images 119. According to various examples, the binary connected component algorithm can assign individual affected voxels to a cluster if the affected voxel is connected to a group of at least 5 other affected voxels. The processing service 116 can apply a binary connected component algorithm to the images 119 obtained at block 200, the images 119 segmented at block 203, or the images 119 which have received the disease mask at block 206. In some embodiments, the processing service 116 can use other image processing techniques to identify clusters of affected voxels.

[0052] At block 213, the processing service 116 can be executed to estimate the equivalent diameter of the clusters identified at block 209. The processing service 116 can be configured to use various data about the patient, the structure of interest, and the scale of the images 119 to determine a conversion of voxel to unit of distance. In some embodiments, the processing service 116 estimates the diameter of each cluster based at least in part on the number of voxels in the cluster. The processing service 116 can estimate the diameter of each cluster individually and separately.

[0053] Moving to block 216, the processing service 116 can assign a weight to each voxel. In some embodiments, the processing service 116 can implement smoothing to assign a weight to each voxel. For example, the processing service 116 can use Gaussian smoothing to assign a weight to each voxel. The processing service 116 can assign a value to each unaffected voxel near an affected cluster identified at block 209. The value assigned to each nearby unaffected voxel can represent the local weighted average of the adjacent affected cluster voxels. In some embodiments, the weight assigned to each unaffected voxel is based at least in part on the size and location of nearby clusters. For example, the farther away and smaller the adjacent cluster is to an unaffected voxel, the lower the weight assigned to that unaffected voxel. Next, at block 219, the processing service 116 can blur one or more voxels surrounding the clusters. In some embodiments, the processing service 116 can blur the surrounding voxels for each cluster of affected voxels separately by convolution or through evolving a variety of partial differential equations. For example, the processing service 116 can apply the following partial differential equation to blur the surrounding voxels: du x,y, z; t) ,

[0054] - — - = div with the initial conditions: u x,y,z,- 0) = M(x,y, z) and the diffusion coefficient c(x,y,z; t) computed as c(x,y, z; t) = a(u(x,y, z; t * K(x,y, z)). In the equations above, u is 3D- MAL map updated over space (x,y,z) and time t, M is the original emphysema binary mask, a is the conductance gain, the asterisk denotes a convolution, and K is the blurring (lowpass filter) kernel. The Eq.1 is iteratively solved numerically, while all values of u > 1 and within the emphysema mask M are replaced with 1 at each iteration. In some embodiments, the processing service 116 blurs each voxel based at least in part on the weight assigned at block 216. In some embodiments, the processing service 116 blurs the voxels surrounding the one or more clusters identified at block 209 as a part of the smoothing used to assign a value to each unaffected voxel at block 216.

[0055] At block 223, the processing service 116 can be executed to generate a weight map. The processing service 116 can generate a weight map based at least in part on the weight assigned at block 216 or the blurring at block 219. In some embodiments, the weight map generated by the processing service 116 can represent the effect of diseased voxels on nearby unaffected voxels based at least in part on the size and location of nearby affected clusters identified at block 209. According to various examples, the processing service 116 generates a Gaussian weight map based at least in part on the Gaussian smoothing conducted at block 216.

[0056] Next, at block 226, the processing service 116 can be executed to predict disease progression. In some embodiments, the processing service 116 can predict disease progression based at least in part on the weight map generated at block 223. The processing service 116 can analyze the weight map and, in some embodiments, predict the likelihood of the disease moving to nearby voxels based at least in part on the weight assigned to each voxel. In some embodiments, the processing service 116 can predict progression of the disease based at least in part on other identified features from the images 119. For example, the processing service 116 can predict disease progression based at least in part on the distance of a particular voxel from a blood vessel. After block 226, the flowchart of FIG. 2 comes to an end.

[0057] A number of software components previously discussed are stored in the memory of the respective computing devices and are executable by the processor of the respective computing devices. In this respect, the term "executable" means a program file that is in a form that can ultimately be run by the processor. Examples of executable programs can be a compiled program that can be translated into machine code in a format that can be loaded into a random access portion of the memory and run by the processor, source code that can be expressed in proper format such as object code that is capable of being loaded into a random access portion of the memory and executed by the processor, or source code that can be interpreted by another executable program to generate instructions in a random access portion of the memory to be executed by the processor. An executable program can be stored in any portion or component of the memory, including random access memory (RAM), read-only memory (ROM), hard drive, solid-state drive, Universal Serial Bus (USB) flash drive, memory card, optical disc such as compact disc (CD) or digital versatile disc (DVD), floppy disk, magnetic tape, or other memory components.

[0058] The memory includes both volatile and nonvolatile memory and data storage components. Volatile components are those that do not retain data values upon loss of power. Nonvolatile components are those that retain data upon a loss of power. Thus, the memory can include random access memory (RAM), read-only memory (ROM), hard disk drives, solid-state drives, USB flash drives, memory cards accessed via a memory card reader, floppy disks accessed via an associated floppy disk drive, optical discs accessed via an optical disc drive, magnetic tapes accessed via an appropriate tape drive, or other memory components, or a combination of any two or more of these memory components. In addition, the RAM can include static random-access memory (SRAM), dynamic random access memory (DRAM), or magnetic random access memory (MRAM) and other such devices. The ROM can include a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or other like memory device.

[0059] Although the applications and systems described herein can be embodied in software or code executed by general purpose hardware as discussed above, as an alternative the same can also be embodied in dedicated hardware or a combination of software / general purpose hardware and dedicated hardware. If embodied in dedicated hardware, each can be implemented as a circuit or state machine that employs any one of or a combination of a number of technologies. These technologies can include, but are not limited to, discrete logic circuits having logic gates for implementing various logic functions upon an application of one or more data signals, application specific integrated circuits (ASICs) having appropriate logic gates, field-programmable gate arrays (FPGAs), or other components, etc. Such technologies are generally well known by those skilled in the art and, consequently, are not described in detail herein.

[0060] The flowcharts show the functionality and operation of an implementation of portions of the various embodiments of the present disclosure. If embodied in software, each block can represent a module, segment, or portion of code that includes program instructions to implement the specified logical function(s). The program instructions can be embodied in the form of source code that includes human-readable statements written in a programming language or machine code that includes numerical instructions recognizable by a suitable execution system such as a processor in a computer system. The machine code can be converted from the source code through various processes. For example, the machine code can be generated from the source code with a compiler prior to execution of the corresponding application. As another example, the machine code can be generated from the source code concurrently with execution with an interpreter. Other approaches can also be used. If embodied in hardware, each block can represent a circuit or a number of interconnected circuits to implement the specified logical function or functions.

[0061] Although the flowcharts show a specific order of execution, it is understood that the order of execution can differ from that which is depicted. For example, the order of execution of two or more blocks can be scrambled relative to the order shown. Also, two or more blocks shown in succession can be executed concurrently or with partial concurrence. Further, in some embodiments, one or more of the blocks shown in the flowcharts can be skipped or omitted. In addition, any number of counters, state variables, warning semaphores, or messages might be added to the logical flow described herein, for purposes of enhanced utility, accounting, performance measurement, or providing troubleshooting aids, etc. It is understood that all such variations are within the scope of the present disclosure.

[0062] Also, any logic or application described herein that includes software or code can be embodied in any non-transitory computer-readable medium for use by or in connection with an instruction execution system such as a processor in a computer system or other system. In this sense, the logic can include statements including instructions and declarations that can be fetched from the computer-readable medium and executed by the instruction execution system. In the context of the present disclosure, a "computer- readable medium" can be any medium that can contain, store, or maintain the logic or application described herein for use by or in connection with the instruction execution system. Moreover, a collection of distributed computer-readable media located across a plurality of computing devices (e.g., storage area networks or distributed or clustered filesystems or databases) may also be collectively considered as a single non-transitory computer-readable medium.

[0063] The computer-readable medium can include any one of many physical media such as magnetic, optical, or semiconductor media. More specific examples of a suitable computer-readable medium would include, but are not limited to, magnetic tapes, magnetic floppy diskettes, magnetic hard drives, memory cards, solid-state drives, USB flash drives, or optical discs. Also, the computer-readable medium can be a random access memory (RAM) including static random access memory (SRAM) and dynamic random access memory (DRAM), or magnetic random access memory (MRAM). In addition, the computer-readable medium can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or other type of memory device.

[0064] Further, any logic or application described herein can be implemented and structured in a variety of ways. For example, one or more applications described can be implemented as modules or components of a single application. Further, one or more applications described herein can be executed in shared or separate computing devices or a combination thereof. For example, a plurality of the applications described herein can execute in the same computing device, or in multiple computing devices in the same computing environment 103.

[0065] In addition to the foregoing, the various embodiments of the present disclosure include, but are not limited to, the embodiments set forth in the following clauses.

[0066] Clause 1. A system, comprising: an imaging system; at least one computing device comprising a processor and a memory; and machine-readable instructions stored in the memory that, when executed by the processor, cause the computing device to at least: capture one or more images using the imaging system; identify one or more clusters of affected voxels within the one or more images; apply smoothing to each of the one or more clusters of affected voxels; generate a weight map based at least in part on the smoothing; and predict the progression of a disease based at least in part on the weight map. Clause 2. The system of clause 1 , wherein the machine-readable instructions which, when executed, cause the computing device to identify one or more clusters of affected voxels, further cause the computing device to at least: segment the one or more images using an organ mask; apply a disease mask to identify a plurality of affected voxels; and apply a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels.

[0067] Clause 3. The system of clause 2, wherein the machine-readable instructions which, when executed, cause the computing device to apply the disease mask to identify the plurality of affected voxels further causes the computing device to at least label each voxel having a value of less than -950 HU as an affected voxel.

[0068] Clause 4. The system of any one of clauses 1 to 3, wherein the machine-readable instructions which, when executed, cause the computing device to apply smoothing to each of the one or more clusters, further cause the computing device to at least: estimate a diameter of each of the one or more clusters; assign a weight to each voxel based at least in part on the diameter and a distance from each of the one or more clusters; and blur each of the one or more clusters based at least in part on the weight of each voxel.

[0069] Clause 5. The system of clause 4, wherein the machine-readable instructions which, when executed, cause the computing device to generate the weight map, further cause the computing device to generate the weight map based at least in part on the assigned weight of each voxel.

[0070] Clause 6. The system of any one of clauses 1 to 5, wherein the machine-readable instructions which, when executed, cause the computing device to predict the progression of the disease, further cause the computing device to at least: identify vasculature near the one or more clusters of affected voxels; and predict the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature.

[0071] Clause 7. The system of any one of clauses 1 to 6, wherein the imaging system comprises a computed tomography (CT) system, a magnetic resonance imaging (MRI) system, or an x-ray system.

[0072] Clause 8. A method, comprising: capturing, with an imaging system, one or more images; identifying, by a computing device, one or more clusters of affected voxels within the one or more images; applying, by the computing device, smoothing to each of the one or more clusters of affected voxels; generating, by the computing device, a weight map based at least in part on the Gaussian smoothing; and predicting, by the computing device, the progression of a disease based at least in part on the weight map.

[0073] Clause 9. The method of clause 8, wherein identifying one or more clusters of affected voxels further comprises at least: segmenting, by the computing device, the one or more images using an organ mask; applying, by the computing device, a disease mask to identify a plurality of affected voxels; and applying, by the computing device, a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels.

[0074] Clause 10. The method of clause 9, wherein applying the disease mask to identify the plurality of affected voxels further comprise at least labeling, by the computing device, each voxel having a value of less than -950 HU as an affected voxel.

[0075] Clause 11 . The method of any one of clauses 8 to 10, wherein applying smoothing to each of the one or more clusters further comprises at least: estimating, by the computing device, a diameter of each of the one or more clusters; assigning, by the computing device, a weight to each voxel based at least in part on the diameter and a distance from each of the one or more clusters; blurring, by the computing device, each of the one or more clusters based at least in part on the diameter of each of the one or more clusters; and summing, by the computing device, the weight of each voxel to obtain a total weight.

[0076] Clause 12. The method of clause 11 , wherein generating the weight map further comprises generating the weight map based at least in part on the total weight.

[0077] Clause 13. The method of any one of clauses 8 to 12, wherein predicting the progression of the disease further comprises at least: identifying, by the computing device, vasculature near the one or more clusters of affected voxels; and predicting, by the computing device, the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature.

[0078] Clause 14. The method of any one of clauses 8 to 13, wherein the imaging system comprises a computed tomography (CT) system, a magnetic resonance imaging (MRI) system, or an x-ray system.

[0079] Clause 15. A system, comprising: an imaging system; a non-transitory, computer- readable medium, comprising machine-readable instructions that, when executed by a processor, cause a computing device to at least: capture one or more images using the imaging system; identify one or more clusters of affected voxels within the one or more images; apply smoothing to each of the one or more clusters of affected voxels; generate a weight map based at least in part on the smoothing; and predict the progression of a disease based at least in part on the weight map. Clause 16. The system of clause 15, wherein the machine-readable instructions which, when executed, cause the computing device to identify one or more clusters of affected voxels, further cause the computing device to at least: segment the one or more images using an organ mask; apply a disease mask to identify a plurality of affected voxels; and apply a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels.

[0080] Clause 17. The system of clause 16, wherein the machine-readable instructions which, when executed, cause the computing device to apply the disease mask to identify the plurality of affected voxels further causes the computing device to at least label each voxel having a value of less than -950 HU as an affected voxel.

[0081] Clause 18. The system of any one of clauses 15 to 17, wherein the machine- readable instructions which, when executed, cause the computing device to apply smoothing to each of the one or more clusters, further cause the computing device to at least: estimate a diameter of each of the one or more clusters; assign a weight to each voxel based at least in part on the diameter and a distance from each of the one or more clusters; blur each of the one or more clusters based at least in part on the diameter of each of the one or more clusters; and sum the weight of each voxel to obtain a total weight.

[0082] Clause 19. The system of clause 18, wherein the machine-readable instructions which, when executed, cause the computing device to generate the weight map, further cause the computing device to generate the weight map based at least in part on the total weight. Clause 20. The system of any one of clauses 15 to 19, wherein the machine- readable instructions which, when executed, cause the computing device to predict the progression of the disease, further cause the computing device to at least: identify vasculature near the one or more clusters of affected voxels; and predict the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature.

[0083] Disjunctive language such as the phrase “at least one of X, Y, or Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., can be either X, Y, or Z, or any combination thereof (e.g., X; Y; Z; X or Y; X or Z; Y or Z; X, Y, or Z; etc.). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each be present.

[0084] It should be emphasized that the above-described embodiments of the present disclosure are merely possible examples of implementations set forth for a clear understanding of the principles of the disclosure. Many variations and modifications can be made to the above-described embodiments without departing substantially from the spirit and principles of the disclosure. All such modifications and variations are intended to be included herein within the scope of this disclosure and protected by the following claims.

[0085] EXAMPLE A

[0086] 3D-Mechanically Affected Lung (3D-MAL)

[0087] The small conducting airways <2 mm in internal diameter that offer little resistance to airflow in normal subjects become the major site of airway obstruction in chronic obstructive pulmonary disease (COPD). As lung function declines, these airways show progressive thickening of their walls, infiltration of walls with inflammatory cells, and occlusion of the lumen by inflammatory exudates and mucus. It has been postulated that centrilobular emphysema develops beyond the surviving terminal bronchioles. Although “small airway disease precedes emphysema" is now the most prevalent view of the sequence of structural changes in COPD, much of this data is based on cross-sectional analysis of explanted lungs, surgical specimens, or in vivo computed tomography (CT) images. COPD is, however, characterized by both airway remodeling and alveolar destruction, and it is likely that both processes contribute to disease initiation and progression. In a systematic study of 1 ,508 current and former smokers from COPDGene, the parametric response mapping (PRM), a technique pairing inspiratory and expiratory images voxel-by-voxel, was used to quantify lung disease as emphysema and non-emphysematous air trapping or functional small airway disease (PRMfSAD). Although PRMfSADwas associated with forced expiratory volume (FEVi) decline, especially in mild-to-moderate stage COPD, both emphysema and small airway disease were found to be associated with lung function decline. Population cohorts also indicate that emphysema and airway disease on CT are both associated with accelerated lung function decline and predict incident spirometric airflow obstruction. Clinical observation and cohort studies indicate that emphysema also occurs very early in the course of the disease. In the population-based Multi-Ethnic Study of Atherosclerosis (MESA) study of individuals without airflow obstruction, the prevalence of percentage emphysema above the upper limit of normal was 5.4%. Multiple studies have demonstrated that individuals with visual and quantitative emphysema on CT have a faster decline in lung function than those without emphysema.

[0088] Currently, there is limited understanding of the mechanisms of progression of structural disease. Preclinical and clinical studies have implicated some structural factors in disease progression.

[0089] (1) Mechanotransduction-. It has been postulated that there is a relationship between mechanical stress and parenchymal destruction. Cyclical stretching of the alveolar walls during respiration in the milieu of higher levels of degradative enzymes, as seen in emphysema, is associated with alveolar destruction. The extracellular matrix of the lung is constantly subject to transpulmonary pressure, with intermittent increases during tidal breathing; the magnitude of this pressure is higher with cough and during exacerbations. In emphysematous areas, where the extracellular matrix is already damaged, these pressure changes can cause further destruction of surrounding alveolar walls resulting in the coalescence of emphysema clusters. These findings are consistent with the in-vitro zipper network model proposed for the failure of the extracellular matrix over time, wherein with decreasing number of fibrils, the remaining fibrils become exposed to progressively more pressure load and stretch and are hence more likely to rupture. In 680 subjects enrolled in the COPDGene study, image registration was used to calculate the Jacobian determinant, a measure of local lung expansion and contraction on CT and showed that emphysematous areas exert a mechanical effect up to 2 mm out in the surrounding normal parenchyma, resulting in a penumbra of mechanically affected lung (MAL). This study was, however, limited by the use of the Euclidean distance to calculate the distance of each normal voxel from its nearest emphysema neighbor, and did not account for mechanical influences of multiple surrounding emphysematous areas whose effects may be additive or multiplicative. Simulations have been used in other studies to suggest disease progression is more likely to occur near existing emphysematous areas, with the formation of larger clusters of emphysema over time.

[0090] (2) Disease Distribution: Whether the distribution of emphysema affects disease progression is not known. Prior studies show that predominant upper lobe emphysema is associated with faster progression than lower lobe emphysema, but studies examining spatial distributions at the regional and voxel level are lacking. If the mechanical effects of emphysema result in progression of emphysema, then the distribution of emphysema should also be important in disease progression, as more widespread distribution of a given degree of emphysema should result in a larger penumbra of lung at-risk for progression.

[0091] Cyclical stretching of the alveolar walls during respiration is associated with alveolar wall destruction. In-vitro, a zipper network model has been proposed for the failure of the extracellular matrix over time, wherein with decreasing number of fibrils, the remaining fibrils become exposed to progressively more pressure load and stretch and hence are more likely to rupture. In an objective study of cough frequency in the ambulatory setting, smokers with COPD cough approximately 9 times an hour, almost double that of COPD ex- smokers (4.9 / hour) and healthy smokers (5.3 / hr). In contrast, healthy non-smokers cough infrequently (0.7 / hour). Conservatively, a smoker with COPD aged 40 years will cough 1.97 million times by the time they are 65 years old, resulting in substantial cumulative stretch on alveolar walls. These forces are likely increased during exacerbations, which in turn are associated with lung function decline.

[0092] A spatial relationship exists between areas of emphysema and abnormal lung mechanics in adjacent normal lung, resulting in a penumbra of mechanically affected lung. To improve the anatomic localization and clinical applicability of these relationships between lung mechanics and FEVi change, the spatial distribution of normal voxels in relationship to emphysematous voxels was examined. A distance map was created by plotting the Euclidean distance between each normal voxel to the nearest voxel with emphysema (Figure 3). The percentage of normal voxels within each mm distance from emphysematous voxel was quantified (mechanically affected lung, MAL). After adjustment for age, race, sex, BMI, current smoking status, pack-years of smoking, baseline FEVi, CT scanner protocol, CT emphysema, CT gas trapping and airway wall thickness, the mean Jacobian determinant of normal voxels within 1 mm of emphysema (MALi) was significantly associated with FEVi change (adjusted p= -21.985, 95%CI = -43.791 to -0.179, p = 0.048). Similar relationships were seen for the mean Jacobian determinant of normal voxels within 2 mm, MAL2 (adjusted P= -22.243, 95%CI = -44.317 to -0.169; p = 0.048). However, this relationship was no longer significant for voxels within 3 mm, MAL3 and beyond.

[0093] In order to understand the significance of MAL2 and its implications for lung function change in earlier disease stages, the rate of change of FEVi for varying levels of MAL2 in spirometric GOLD (Global Initiative for Chronic Obstructive Lung Disease) grades 1 and 2 for whom the mean AFEVi was -54.3±61.2 ml was also examined. For participants at or above the median MAL2 (threshold 36.9%), the mean AFEVi was -56.4±68.0 ml / year vs. - 43.2±59.9 ml / year for those below the median (p=0.044). For participants with >75thpercentile of MAL2 (threshold 47.2%), the mean AFEVi was -65.3±75.7 ml / year compared to those <75thpercentile, -44.9±60.2, p= 0.039.

[0094] Three-Dimensional Mechanically Affected Lung (3D-MAL) provides estimates of lung at-risk for disease progression. As MAL2 does not account for the influence of emphysema other than the closest emphysema lesion alone, a new metric termed 3D-MAL was created that accounts for emphysema surrounding a normal voxel in three dimensions (details in the Method section). Figure 4 shows that the higher the 3D-MAL, the greater the lung function decline as well as emphysema progression. 3D-MAL also correlated inversely with the mean Jacobian determinant of normal voxels (Pearson’s r = -0.16; p=0.028).

[0095] Method:

[0096] 3D MAL: After segmenting the lungs from inspiratory CT images using lung masks, we will apply an emphysema mask by labeling voxels <-950HU within the lung mask as emphysema. A binary connected component algorithm will be applied on the emphysema mask to identify individual clusters of emphysema where individual emphysema voxels will be assigned to a cluster if it is connected to a group of at least 5 connected emphysema voxels (Figure 5A). The diameter of each emphysematous cluster will be estimated separately. The 3D effect of emphysema voxels on adjacent normal voxels will be quantified. The mechanical effect is assumed to be a decaying effect with progressively increasing distance and a factor of the size of the emphysema cluster. Since the image and mask domains are bounded, numerical approximations of truncated Gaussians can be used for Gaussian smoothing. This Gaussian smoothing (3D) will be applied separately to each emphysematous cluster in the emphysema mask where the degree of smoothing is determined by the diameter of that specific cluster as follows: (x2+y2+z2)

[0097] IV (x, y, z) = M(x, y, z) * — - — 3 e2O? where (x, y, z) is the voxel location, denotes (V27T<TC) the convolution operation, W is the Gaussian weight, M is a lung mask, and acis a width parameter (crc= equivalent radius of the emphysema cluster c, defined as the radius of a sphere with the same volume as the cluster c). This convolution operation creates a 3D blur which decays with distance (Figure 5B). The Gaussian smoothing assigns a value to nearby normal lung voxels representing the local weighted average of the adjacent emphysema cluster voxels. The farther and smaller the adjacent emphysema cluster, the lower the weight assigned to a given normal voxel. To account for multiple emphysema clusters adjacent to a given normal voxel, each cluster is blurred separately by convolution and the total contribution from all clusters over each normal voxel is summed. Thus, the Gaussian weight map generated will represent the effect of emphysema voxels on nearby normal voxels by taking into account both the size of the emphysema cluster and its distance from the normal voxel (Figure 50 and Figure 6).

[0098] Pulmonary Diffusion:

[0099] The pulmonary vasculature and diffusion of oxygen play an important role in emphysema initiation and progression. The farther a lung region is from a blood vessel, the lower the perfusion is likely to be and hence the greater the likelihood of emphysema. A similar adaptive diffusion method is applied to quantify how well perfused a given lung voxel is from the nearest blood vessels.

Claims

CLAIMSTherefore, the following is claimed:

1. A system, comprising: an imaging system; at least one computing device comprising a processor and a memory; and machine-readable instructions stored in the memory that, when executed by the processor, cause the computing device to at least: capture one or more images using the imaging system; identify one or more clusters of affected voxels within the one or more images; apply smoothing to each of the one or more clusters of affected voxels; generate a weight map based at least in part on the smoothing; and predict the progression of a disease based at least in part on the weight map.

2. The system of claim 1 , wherein the machine-readable instructions which, when executed, cause the computing device to identify one or more clusters of affected voxels, further cause the computing device to at least: segment the one or more images using an organ mask; apply a disease mask to identify a plurality of affected voxels; and apply a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels.

3. The system of claim 2, wherein the machine-readable instructions which, when executed, cause the computing device to apply the disease mask to identify the plurality of affected voxels further causes the computing device to at least label each voxel having a value of less than -950 HU as an affected voxel.

4. The system of claim 1 , wherein the machine-readable instructions which, when executed, cause the computing device to apply smoothing to each of the one or more clusters, further cause the computing device to at least: estimate a diameter of each of the one or more clusters; assign a weight to each voxel based at least in part on the diameter and a distance from each of the one or more clusters; and blur each of the one or more clusters based at least in part on the weight of each voxel.

5. The system of claim 4, wherein the machine-readable instructions which, when executed, cause the computing device to generate the weight map, further cause the computing device to generate the weight map based at least in part on the assigned weight of each voxel.

6. The system of claim 1 , wherein the machine-readable instructions which, when executed, cause the computing device to predict the progression of the disease, further cause the computing device to at least: identify vasculature near the one or more clusters of affected voxels; and predict the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature.

7. The system of claim 1 , wherein the imaging system comprises a computed tomography (CT) system, a magnetic resonance imaging (MRI) system, or an x-ray system.

8. A method, comprising: capturing, with an imaging system, one or more images; identifying, by a computing device, one or more clusters of affected voxels within the one or more images; applying, by the computing device, smoothing to each of the one or more clusters of affected voxels;generating, by the computing device, a weight map based at least in part on the Gaussian smoothing; and predicting, by the computing device, the progression of a disease based at least in part on the weight map.

9. The method of claim 8, wherein identifying one or more clusters of affected voxels further comprises at least: segmenting, by the computing device, the one or more images using an organ mask; applying, by the computing device, a disease mask to identify a plurality of affected voxels; and applying, by the computing device, a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels.

10. The method of claim 9, wherein applying the disease mask to identify the plurality of affected voxels further comprise at least labeling, by the computing device, each voxel having a value of less than -950 HU as an affected voxel.

11. The method of claim 8, wherein applying smoothing to each of the one or more clusters further comprises at least: estimating, by the computing device, a diameter of each of the one or more clusters; assigning, by the computing device, a weight to each voxel based at least in part on the diameter and a distance from each of the one or more clusters; blurring, by the computing device, each of the one or more clusters based at least in part on the diameter of each of the one or more clusters; and summing, by the computing device, the weight of each voxel to obtain a total weight.

12. The method of claim 11 , wherein generating the weight map further comprises generating the weight map based at least in part on the total weight.

13. The method of claim 8, wherein predicting the progression of the disease further comprises at least: identifying, by the computing device, vasculature near the one or more clusters of affected voxels; and predicting, by the computing device, the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature.

14. The method of claim 8, wherein the imaging system comprises a computed tomography (CT) system, a magnetic resonance imaging (MRI) system, or an x-ray system.

15. A system, comprising: an imaging system; a non-transitory, computer-readable medium, comprising machine-readable instructions that, when executed by a processor, cause a computing device to at least: capture one or more images using the imaging system; identify one or more clusters of affected voxels within the one or more images; apply smoothing to each of the one or more clusters of affected voxels; generate a weight map based at least in part on the smoothing; and predict the progression of a disease based at least in part on the weight map.

16. The system of claim 15, wherein the machine-readable instructions which, when executed, cause the computing device to identify one or more clusters of affected voxels, further cause the computing device to at least: segment the one or more images using an organ mask; apply a disease mask to identify a plurality of affected voxels; and apply a binary connected component algorithm to identify one or more clusters of affected voxels based at least in part on the plurality of affected voxels.

17. The system of claim 16, wherein the machine-readable instructions which, when executed, cause the computing device to apply the disease mask to identify the plurality of affected voxels further causes the computing device to at least label each voxel having a value of less than -950 HU as an affected voxel.

18. The system of claim 15, wherein the machine-readable instructions which, when executed, cause the computing device to apply smoothing to each of the one or more clusters, further cause the computing device to at least: estimate a diameter of each of the one or more clusters; assign a weight to each voxel based at least in part on the diameter and a distance from each of the one or more clusters; blur each of the one or more clusters based at least in part on the diameter of each of the one or more clusters; and sum the weight of each voxel to obtain a total weight.

19. The system of claim 18, wherein the machine-readable instructions which, when executed, cause the computing device to generate the weight map, further cause the computing device to generate the weight map based at least in part on the total weight.

20. The system of claim 15, wherein the machine-readable instructions which, when executed, cause the computing device to predict the progression of the disease, further cause the computing device to at least: identify vasculature near the one or more clusters of affected voxels; and predict the progression of the disease to a voxel based at least in part on a distance between the voxel and the identified vasculature.

Images