SYSTEM AND METHOD FOR ASSESSMENT OF GROUND GLASS OPACITY IN A COMPUTED TOMOGRAPHY IMAGE

Abstract

ABSTRACT SYSTEM AND METHOD FOR ASSESSMENT OF GROUND GLASS OPACITY IN A COMPUTED TOMOGRAPHY IMAGE The present invention describes a technique for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image. At first, a CT image is acquired. A first filtering of pixels of the CT image is performed (104) to extract lung masks comprising two largest blobs in the CT image. Further, a second filtering of pixels of the CT image is performed (106) to extract vessel and GGO masks. The vessel masks from the lung masks is eliminated (108) to segment out only the GGO masks. A fractal image of the segmented GGO masks is created (110). A fractal dimensions based on the fractal image of the segmented GGO masks is determined (112). Finally, a lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, is performed (114) to quantify texture and distribution of the GGO patterns in the CT image. (Fig. 1)

Specification

Description:[001] FIELD OF THE INVENTION
[002] The present invention relates to lung disease diagnosing, and particularly, to a method and system for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image.

[003] BACKGROUND OF THE INVENTION
[004] Background description includes information that may be useful in understanding the present invention
[005] The computed tomography (CT) imaging forms the mainstay in assessing the pattern, location, and regional distribution of involvement in the diagnosis of several lung disorders such as COVID-19, pneumonia, and lung cancer. According to one research study, the tomography imaging allows high resolution assessment of lung parenchyma and surrounding structures to evaluate disease severity. According to another research study, they can demonstrate in-depth analysis on morphological and functional characteristics of lung regions with high sensitivity. The advent of high-Resolution CT (HRCT) images provides valuable information that cannot be determined from clinical history and other diagnostic tests. According to another study, the HRCT plays major roles such as assisting in determining prognosis, monitoring for the efficacy of treatment, detecting progression of disease, and evaluating patients with acute symptoms.
[006] According to another research study, the chest CT has a significant role in the very early stages of the infection. Detecting the abnormalities in the lung tissue in an early phase improves the treatment efficiency and gives the patient a better chance of survival. Pulmonary nodules are one of the important lesions and early symptoms of major diseases such as lung cancer and COVID-19 Pneumonia. The CT makes it possible to visualize the small or low-opacity nodules that are hardly seen in the other conventional medical imaging techniques. Pulmonary nodules can be divided into solid nodules and ground glass nodules (GGNs).
[007] The GGNs often referred to as GGOs are more likely to be malignant than solid nodules. According to another research study, the GGOs are non-specific imaging signs defined as hazy lung opacities that do not obscure the underlying vascular or bronchial margins. GGOs are further sub-divided into pure and mixed GGOs based on the existence of solidity. GGOs are the most common findings in major lung diseases.
[008] According to another research study, in early stages of the disease, the CT findings are bilateral distribution of ground glass opacities with or without consolidation in the posterior and peripheral lungs. According to another research study, the characteristics of GGO is highly complex as it is characterized by patchiness with an irregular shape, which may be inflammation, hemorrhage, granulomatous lesions, or tumors.
[009] The development of an effective computer-aided diagnosis (CAD) technique for lung disease diagnosis is of great clinical importance and can increase the patient's chance of survival. According to another research study, the CAD also offers second opinion by assisting radiologists and helps to overcome interpretation difficulties that occur due to intensity variations and anatomical mis-judgements in CT scans. These techniques improve accuracy of diagnosis, speed and automation level. According to another research study, a typical CAD technique for lung diseases involves steps such as image acquisition, pre-processing, lung segmentation, nodule detection and characterization. The lungs segmentation is performed as a pre-processing step to nodule detection step in order to eliminate intensity similarities between lung wall and nodular opacities and obtain high detection rates. According to another research study, the lung field segmentation is considered to be challenging due to involvement of vascular structures and nodules attached to the lung walls.
[0010] The chest radiologists rely on the segmentation analysis of GGO to extract quantitative information that evaluate the disease severity. However, it may be difficult to segment and analyze patterns of GGO since it usually contains unclear clear boundaries. The CT value of GGOs is always lower than that of blood vessels; and similar to normal lung tissues. Therefore, GGOs may not always be obvious on CT images, and they may be missed. The recognition of GGO is based on a subjective assessment of lung attenuation at CT, but manual observation is labor-intensive and time-consuming, and the results of examination may often be misleading. According to some research studies, several GGO based segmentation CAD technique using methods such as thresholding, region growing, clustering, pixel classification and deep learning have been developed. However, simpler automated analysis of GGO identification, segmentation and localization has not been established due to the low detection sensitivity in locating the suspicious areas that includes both the true nodules and a high number of false positives. Therefore, effective delineation using digital IP technique for CAD to assist physicians and radiologists were required.
[0011] According to another research studies, morphological/structural features of GGOs are important to identify the aggressiveness of the disease and helps to differentiate abnormal patterns. Further, multiscale textural features such as fractals are necessary to represent internal structures of GGOs based on regularity and coarseness to indicate malignancy. The fractal dimension of a structure provides a measure of its texture complexity. Another fractal measure, lacunarity measures the irregular shapes of the fractal data, providing metainformation in the image. The higher the lacunarity, the more inhomogeneous the examined fractal area and vice versa. Therefore, this technique deals with GGO characterization by determining fractal feature vectors for regions of interest, and using these vectors as predictors for tumor severity. According to some research study, investigation and classification of GGOs according to their aggressiveness and determining how FD relates to real medical key indicators is important.
[0012] Visualizing and interpreting the abnormalities help to validate the model's localization performance in assessing subtle characteristics and suspicious biomarkers in medical images. According to another research study, besides quantitative analysis using morphological features, heatmap based visualization is essential to provide a visual diagnostic approach that could be clinically useful to assist radiologists to take precise decisions in a minimal amount of time in healthcare settings. According to some research studies, different schemes have been used to represent heatmaps with colour overlay for GGO, lungs and vessels.
[0013] The existing methods of GGO identification involve manual determination in detecting subtle GGO patterns, which can be laborious and subject to misinterpretation leading to diagnostic errors. An early and accurate detection of GGO is crucial for timely intervention, particularly in severe lung conditions.
[0014] Thus, there is a desired need for an automatic technique for an early and accurate detection of GGO.

[0015] OBJECTS OF THE INVENTION:
[0016] Some of the objects of the present disclosure, which at least one embodiment herein satisfies, are listed herein below.
[0017] The primary objective of the present invention is to develop a computer-based detection technique to automatically identify the GGO structures in the lungs that characterize COVID-19 and pneumonia conditions, provide an enhanced visualization and severity analysis of these regions through critical image markers allowing clinicians to easily distinguish GGO from other lung structures, and obtain insights on disease progression for further follow-up, treatment and management.
[0018] Another objective of the present invention is to accurate detect GGO patterns that characterize pneumonia conditions for its early detection.
[0019] Another objective of the present invention is to develop a resource-intensive and efficient solution to identify GGO structures from the lungs using comprehensible analysis.
[0020] Another objective of the present invention is to visualize GGO structures to assist clinicians and radiologists for improved triaging and diagnosis.
[0021] Another objective of the present invention is to extract reliable and localized image markers for assessing the severity of GGO structures particularly useful in emergency conditions.
[0022] Another objective of the present invention is to automate CAD analysis to reduce the risk of human error and enhance the reliability of disease detection in clinical settings.
[0023] These and other objects and advantages will become more apparent when reference is made to the following description and accompanying drawings.

[0024] SUMMARY OF THE INVENTION
[0025] This summary is provided to introduce concepts related to assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
[0026] In an aspect of the present invention, an improved method of assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image is described. The method includes the step of acquiring the CT image. The method further includes the step of performing a first filtering of pixels of the CT image, with pixel intensities from a first value to a second value, to extract lung masks comprising two largest blobs in the CT image. The method further includes the step of performing a second filtering of pixels of the CT image, with pixel intensities from the first value to a third value, to extract vessel and GGO masks. The second value is greater than the third value. The method further includes the step of eliminating the vessel masks from the lung masks, to segment out only the GGO masks. The method further includes the step of creating a fractal image of the segmented GGO masks, while preserving an original texture of the GGO masks. The method further includes the step of determining fractal dimensions based on the fractal image of the segmented GGO masks. Finally, the method includes the step of performing lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, to quantify texture and distribution of the GGO patterns in the CT image.
[0027] In an embodiment of the present invention, the first, second, and third values are 1, 190, and 120, respectively.
[0028] In another embodiment of the present invention, the lacunarity analysis grades the texture and distribution of the GGOs in the CT image, to determine an extent of pathology in lungs.
[0029] In another embodiment of the present invention, the method further includes the step of determining an average fractal dimension for a selected GGO area in the segmented GGO masks, and a standard deviation of the fractal dimension of the selected area.
[0030] In another embodiment of the present invention, the average fractal dimension indicates the overall complexity of the GGO patterns in the CT image.
[0031] In another embodiment of the present invention, the standard deviation reflects variability in the GGO patterns in the CT image.
[0032] In another embodiment of the present invention, the lung masks, vessel and GGO masks, and GGO masks are overlaid in a predetermined first, second, and third colors, respectively, to distinguish one of the other masks.
[0033] In another embodiment of the present invention, the fractal image provides insights into the spatial complexity of GGO patterns in the CT image.
[0034] In another aspect of the present invention, a system for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image is described. The system includes a CT module, a filtering module, an eliminating module, a fractal image module, and a lacunarity analysis module. The CT module is configured for capturing the CT image. The filtering module is coupled to the CT module and configured for receiving the CT image subsequently performing (i) a first filtering of pixels of the CT image, with pixel intensities from a first value to a second value, to extract lung masks comprising two largest blobs in the CT image, and (ii) a second filtering of pixels of the CT image, with pixel intensities from the first value to a third value, to extract vessel and GGO masks. The second value is greater than the third value. The eliminating module is coupled to the filtering module and configured for eliminating the vessel masks from the lung masks, to segment out only the GGO masks. The fractal image module is coupled to the eliminating module and configured for creating a fractal image of the segmented GGO masks, while preserving an original texture of the GGO masks. The fractal dimension module is coupled to the fractal image module and configured for determining fractal dimensions based on the fractal image of the segmented GGO masks. The lacunarity analysis module coupled to the fractal dimension module and configured for performing lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, to quantify texture and distribution of the GGO patterns in the CT image.
[0035] In an embodiment of the present invention, the first, second, and third values are 1, 190, and 120, respectively.
[0036] In another embodiment of the present invention, the lung masks, vessel and GGO masks, and GGO masks are overlaid in a predetermined first, second, and third colors, respectively, to distinguish one of the other masks.
[0037] Various objects, features, aspects, and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which like numerals represent like components.

[0038] BRIEF DESCRIPTION OF DRAWINGS:
[0039] The illustrated embodiments of the subject matter will be understood by reference to the drawings, wherein like parts are designated by like numerals throughout. The following description is intended only by way of example and simply illustrates certain selected embodiments of devices, apparatus, systems, and methods that are consistent with the subject matter as claimed herein, wherein:
[0040] FIG. 1 illustrates a block diagram depicting a method for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image, in accordance with an exemplary embodiment of the present disclosure;
[0041] FIG. 2 illustrates a block diagram depicting a system configured for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image, in accordance with an exemplary embodiment of the present disclosure;
[0042] FIG. 3 illustrates a schematic work-flow diagram of having stages (i) CT image as input data, (ii) extraction of lung masks from the CT image, (iii) extraction of Vessel and GGO masks from the CT image, and (iv) lacunarity based fractal analysis estimating the fractal dimension and lacunarity score, in accordance with an exemplary embodiment of the present disclosure;
[0043] FIGs. 4(i)-(iii) illustrate an adaptive pixel threshold segmentation for masked images at various threshold levels (a) input CT image (b) 71-80, (c) 81-90, (d) 91-100, (e) 101-110, (f) 111-120, (g) 121-130, (h) 131-140, (i) 141-150, (j) 151-160, (k) 161-170, (l) 171-180, (m) 181-190, (n) 191-200, and (o) 201-210, in accordance with an exemplary embodiment of the present disclosure;
[0044] FIGs. 5(i)-(ii) illustrate a sequential processing of lung and Ground Glass Opacity (GGO) Masks (a) input CT image (b) lung mask (1-190) (c) overlaid lung mask (blue) (d) GGO and vessels mask (1-120) (e) overlaid GGO and vessels mask (f) GGO mask (g) overlaid GGO mask (h) final overlaid (lung, GGO and vessels), in accordance with an exemplary embodiment of the present disclosure;
[0045] FIG. 6 illustrates a schematic diagram (a) chest CT image, (b) FD image, and (c) segmented GGO image, in accordance with an exemplary embodiment of the present disclosure;
[0046] FIG. 7 illustrates a schematic diagram (a) input image and (b) FD analysis of GGO in Region of Interest (ROI), in accordance with an exemplary embodiment of the present disclosure; and
[0047] FIG. 8 illustrates a schematic diagram (a) chest CT Image, and (b) segmented GGO image, in accordance with an exemplary embodiment of the present disclosure.
[0048] The figures depict embodiments of the present subject matter for the purposes of illustration only. A person skilled in the art will easily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the disclosure described herein.

[0049] DESCRIPTION OF THE INVENTION:
[0050] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure as defined by the appended claims.
[0051] While the embodiments of the disclosure are subject to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the figures and will be described below. It should be understood, however, that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternatives falling within the scope of the disclosure.
[0052] The terms "comprises", "comprising", or any other variations thereof used in the disclosure, are intended to cover a non-exclusive inclusion, such that a device, system, or assembly that comprises a list of components does not include only those components but may include other components not expressly listed or inherent to such system, or assembly, or device. In other words, one or more elements in a system or device proceeded by "comprises… a" does not, without more constraints, preclude the existence of other elements or additional elements in the system or device.
[0053] The Ground Glass Opacity (GGO) patterns in medical computed tomography (CT) images are critical markers for diagnosing various lung diseases, including COVID-19 and pneumonia. Identifying and analyzing these patterns accurately is vital for early detection and treatment. The present invention is a Computer-Aided Detection (CAD) technique designed to automate the identification, analysis, and visualization of GGO structures in CT images.
[0054] The present invention utilizes advanced CAD based techniques such as adaptive thresholding to precisely delineate GGO regions. The lung segmentation is performed as a pre-processing step to focus the analysis on relevant areas, enhancing the accuracy of GGO identification. Then it visualizes these regions using heatmap color overlays, allowing clinicians to easily distinguish GGO from other lung structures, such as vessels and lung fields. A key feature of the present invention is the use of lacunarity-based fractal analysis, which assesses the severity of GGO patterns, providing insights into the aggressiveness of the lung disease.
[0055] The need for the CAD technique arises from the challenges faced by radiologists in detecting subtle GGO patterns, which can be difficult to identify through manual analysis. Early and accurate detection of GGO is crucial for timely intervention, particularly in severe lung conditions. The present invention offers a novel, efficient, and less resource-intensive solution, assisting radiologists in making informed diagnoses and improving patient outcomes. By automating the analysis process, the present invention reduces the risk of human error and enhances the reliability of GGO detection in clinical settings.
[0056] For better understanding, one or more embodiments of the present invention shall be described with respect to the earlier-mentioned drawings.
[0057] FIG. 1 illustrates a block diagram depicting a method (100) for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image, in accordance with an exemplary embodiment of the present disclosure.
[0058] As illustrated, the method (100) includes the step of acquiring (102) the CT image. The method (100) further includes the step of performing (104) a first filtering of pixels of the CT image, with pixel intensities from a first value to a second value, to extract lung masks comprising two largest blobs in the CT image. The method (100) further includes the step of performing (106) a second filtering of pixels of the CT image, with pixel intensities from the first value to a third value, to extract vessel and GGO masks. The second value is greater than the third value. The method (100) further includes the step of eliminating (108) the vessel masks from the lung masks, to segment out only the GGO masks. The method (100) further includes the step of creating (110) a fractal image of the segmented GGO masks, while preserving an original texture of the GGO masks. The method (100) further includes the step of determining (112) fractal dimensions based on the fractal image of the segmented GGO masks. Finally, the method (100) includes the step of performing (114) lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, to quantify texture and distribution of the GGO patterns in the CT image.
[0059] In an embodiment of the present invention, the first, second, and third values are 1, 190, and 120, respectively.
[0060] In another embodiment of the present invention, the lacunarity analysis grades the texture and distribution of the GGOs in the CT image, to determine an extent of pathology in lungs.
[0061] In another embodiment of the present invention, the method (100) further includes the step of determining an average fractal dimension for a selected GGO area in the segmented GGO masks, and a standard deviation of the fractal dimension of the selected area.
[0062] In another embodiment of the present invention, the average fractal dimension indicates the overall complexity of the GGO patterns in the CT image.
[0063] In another embodiment of the present invention, the standard deviation reflects variability in the GGO patterns in the CT image.
[0064] In another embodiment of the present invention, the lung masks, vessel and GGO masks, and GGO masks are overlaid in a predetermined first, second, and third colors, respectively, to distinguish one of the other masks.
[0065] In another embodiment of the present invention, the fractal image provides insights into the spatial complexity of GGO patterns in the CT image.
[0066] FIG. 2 illustrates a block diagram depicting a system (200) configured for assessment of Ground Glass Opacity (GGO) in a medical computed tomography (CT) image, in accordance with an exemplary embodiment of the present disclosure.
[0067] As illustrated, the system (200) includes a CT module (202), a filtering module (204), an eliminating module (206), a fractal image module (208), a fractal dimension module (210) and a lacunarity analysis module (212). The CT module (202) is configured for capturing the CT image. The filtering module (204) is coupled to the CT module (202) and configured for receiving the CT image subsequently performing (i) a first filtering of pixels of the CT image, with pixel intensities from a first value to a second value, to extract lung masks comprising two largest blobs in the CT image, and (ii) a second filtering of pixels of the CT image, with pixel intensities from the first value to a third value, to extract vessel and GGO masks. The second value is greater than the third value. The eliminating module (206) is coupled to the filtering module (204) and configured for eliminating the vessel masks from the lung masks, to segment out only the GGO masks. The fractal image module (208) is coupled to the eliminating module (206) and configured for creating a fractal image of the segmented GGO masks, while preserving an original texture of the GGO masks. The fractal dimension module (210) is coupled to the fractal image module (208) and configured for determining fractal dimensions based on the fractal image of the segmented GGO masks. The lacunarity analysis module (212) coupled to the fractal dimension module (210) and configured for performing lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, to quantify texture and distribution of the GGO patterns in the CT image.
[0068] In an embodiment of the present invention, the first, second, and third values are 1, 190, and 120, respectively.
[0069] In another embodiment of the present invention, the lung masks, vessel and GGO masks, and GGO masks are overlaid in a predetermined first, second, and third colors, respectively, to distinguish one of the other masks.
[0070] In one or more embodiments, the system (200) may be part of a larger computer system and/or maybe operatively coupled to a network (e.g., a second network) with the aid of a communication interface to facilitate the transmission of and sharing data and predictive results. The computer network may be a local area network (LAN), an intranet and/or extranet, an intranet and/or extranet that is in communication with the Internet, or the Internet. The network in some cases is a telecommunication and/or a data network, and may include one or more computer servers. In an example, the communication network includes, but not limited to, 2G network, 3G network, 4G network, LTE network, 5G network, 6G network, and so forth. The network, in some cases with the aid of a computer system, may implement a peer-to-peer network, which may enable devices coupled to the computer system to behave as a client or a server. In other examples, the system, the database, and the server may be integrated network node or a single integrated unit.
[0071] The system (200) may communicate with one or more other systems by the interfaces (e.g., network adapters). The memory or memory locations may be, e.g., random-access memory, read-only memory, flash memory. The system may also include at least one electronic storage units (e.g., hard disks), and peripheral devices, such as cache, other memory, data storage, and/or electronic display adapters.
[0072] The system (200) may also include one or more IO Managers as software instructions that may run on the one or more processors (208) and implement various communication protocols such as User Datagram Protocol (UDP), Modbus, MQ Telemetry Transport (MQTT), Open Platform Communications Unified Architecture (OPC UA), Semiconductor's equipment interface protocol for equipment-to-host data communications (SECS/GEM), Profinet, or any other protocol, to access data in real-time from disparate data sources via any communication network, such as Ethernet, Wi-Fi, Universal Serial Bus (USB), Zigbee, Cellular or 5G connectivity, etc., or indirectly through a device's primary controller, through a Programmable Logic Controller (PLC) or through a Data Acquisition System (DAQ), or any other such mechanism.
[0073] Further, the CPU(s) may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and/or any devices that manipulate data based on operational instructions. Among other capabilities, the one or more processor(s) are configured to fetch and execute computer-readable instructions stored in the memory of the system.
[0074] Further, the memory may store one or more computer-readable instructions or routines, which may be fetched and executed to create or share data units over a network service. The memory may include any non-transitory storage device including, for example, volatile memory such as RAM, or non-volatile memory such as EPROM, flash memory, and the like.
[0075] Further, the processing devices(s) may be implemented as a combination of hardware and programming device(s) (for example, programmable instructions) to implement one or more functionalities of the processing device(s). In examples described herein, such combinations of hardware and programming may be implemented in several different ways. In one example, the programming for the processing device(s) may be processor executable instructions stored on a non-transitory machine-readable storage medium and the hardware for the processing device(s) may include a processing resource (for example, one or more processors), to execute such instructions. In other examples, the processing devices(s) may be implemented by electronic circuitry.
[0076] FIG. 3 illustrates a schematic work-flow diagram having stages (i) CT image as input data (302), (ii) extraction (304) of lung masks from the CT image (310), (iii) extraction (306) of Vessel and GGO masks from the CT image (310), and (iv) lacunarity based fractal analysis (308) estimating the fractal dimension and lacunarity score, in accordance with an exemplary embodiment of the present disclosure.
[0077] The present invention involves the CAD technique designed to identify, extract, analyze, and visualize the GGO pattern from medical CT images. The GGO is the area in the lung that usually indicate various diseases of the lungs such as COVID-19, pneumonia, and lung cancer. The correct identification and characterization of GGOs are crucial in the diagnosis and evaluation of the conditions.
[0078] The present invention describes a technique using image processing like adaptive filtering, blob extraction, and fractal analysis to delineate and assess GGOs in an automated way. Accordingly, a multistep pipeline is proposed that processes CT images by segmenting lung masks from vessels and GGOs and evaluates the severity of GGOs with the help of the fractal analysis based on lacunarity.
[0079] As illustrated, at stage (302), the acquisition of CT images (310) of chest takes place. The chest CT image are taken as raw data input to the system. At step (304), the lung masks (314) from the CT image are extracted. To achieve this, the filtering of pixel intensities within the CT image using pixel intensity filter (312) is done to retain values from 1 to 190 through experimental analysis. This sub-step is a necessary preprocessing for isolating the lung regions from the rest of the CT image. The filtered image returns lung masks (314), denoted by M1, which are the two largest blobs in an image representing the lungs. At step (306), vessel and GGO masks from the CT image are extracted. Thus, a further step of filtering using pixel intensity filter (312) is performed on the CT image, this time, the range chosen goes from 1 to 120 to include vessels and GGOs. The result of this filtering yields vessel masks (316), denoted by M2. To segment out only the GGO masks (318), the vessel masks (M2) from the lung masks (M1) is subtracted i.e., M1-M2=M3. At step (308), the lacunarity-based fractal analysis fractal image and fractal dimension and lacunarity analysis take place. Here, a fractal image (320) is created with the GGO masks (318), whereby the original texture of the GGOs is still preserved. It measures the fractal dimension (322) for quantifying the complexity of the GGO image. In addition, a lacunarity analysis (324) is done that grades the texture and distribution of the GGOs, hence helping in determining the extent of pathology in lungs.
[0080] In one or more embodiments, the present invention includes a variety of components as one whole system for in-depth and detailed analysis of GGO patterns present within chest CT images. At the input stage, the chest CT images are acquired. The chest CT images form the foundation upon which the entire system rests, providing the visual input data from which the normal and abnormal structures of the lung are demarcated and analyzed.
[0081] Then, the lung masks extraction from the CT images using a filter based on the pixel intensity takes place. This filter keeps the pixel intensity between 1 and 190 and thus isolates the lung regions from other image anatomical structures. The filtered image provides lung masks called M1, which are two largest blobs corresponding to lungs. This step is important because it defines the main area of interest for the succeeding process, which should narrow down to the lung area only.
[0082] The CT image continues its processing by the vessel and GGO masks extraction module in this phase of processing. A second filtering on pixel intensity retains the value from 1 to 120 to include vessels and GGOs. This filtering resulted in vessel masks M2. For the specific isolation of GGOs, vessel masks are subtracted from lung masks. This would yield the GGO masks, M3, by subtraction from M1. This subtracting keeps only the GGOs and removes other structures so that the major analyses may focus on these critical abnormalities within the lungs.
[0083] The final component is the lacunarity-based fractal analysis, which presents an advanced examination of the segmented GGO patterns. These masks are used to have M3 compute fractal images that preserve the original texture of GGOs. Subsequently, fractal dimension is calculated as an indicator of the complexity in GGO structures, while lacunarity analysis is applied to quantify their texture and distribution. This would greatly help one to understand the severity and extent of lung pathology in depth, thus enriching knowledge of the progression of lung disease.
[0084] The present invention provides an integrated approach to the detection and fractal analysis of GGOs in chest CT images by seamlessly integrating all these components. In such a synergy between image acquisition, lung and vessel mask extraction, and fractal analysis, comprehensive and accurate assessment of lung abnormalities is assured. It is, therefore, very befitting for diagnosis and follow-up study of lung diseases that are marked by GGOs, hence assisting clinicians to make informed decisions on the care needed by the patients.
[0085] FIGs. 4(i)-(iii) illustrate an adaptive pixel threshold segmentation for masked images at various threshold levels (a) input CT image (b) 71-80, (c) 81-90, (d) 91-100, (e) 101-110, (f) 111-120, (g) 121-130, (h) 131-140, (i) 141-150, (j) 151-160, (k) 161-170, (l) 171-180, (m) 181-190, (n) 191-200, and (o) 201-210, in accordance with an exemplary embodiment of the present disclosure.
[0086] As illustrated, the process of adaptive pixel threshold segmentation is applied to a CT image, which is essential for differentiating between various tissue densities and structures. The first panel (a) shows the original CT image, serving as the baseline for comparison. Subsequent panels (b) to (o) display masks created using different pixel intensity thresholds ranging from 71-80 to 201-210. Each mask highlights regions of the CT image that fall within the specified pixel value ranges. The lower thresholds (e.g., (b) 71-80) capture broader areas, including both lung tissue and potential GGOs, while higher thresholds (e.g., (o) 201-210) become more selective, isolating denser structures. The progression of masks demonstrates how increasing threshold levels refine the segmentation, making it possible to differentiate and analyze specific features within the CT image more effectively.
[0087] FIGs. 5(i)-(ii) illustrate a sequential processing of lung and Ground Glass Opacity (GGO) Masks (a) input CT image (b) lung mask (1-190) (c) overlaid lung mask (blue) (d) GGO and vessels mask (1-120) (e) overlaid GGO and vessels mask (f) GGO mask (g) overlaid GGO mask (h) final overlaid (lung, GGO and vessels), in accordance with an exemplary embodiment of the present disclosure.
[0088] As illustrated, the step-by-step process of segmentating and analyzing the lung and GGO structures from a CT image is described. The first panel (a) shows the original CT image. Panel (b) presents the lung mask generated by thresholding pixel values between 1-190, which isolates the lung regions from the rest of the image. In (c), the lung mask is overlaid in blue to provide a clear visual representation of the lung areas.
[0089] Panel (d) displays the combined mask of GGOs and vessels with pixel values ranging from 1120. Panel (e) shows the overlay of this mask with the previously obtained lung mask to highlight the regions where GGOs and vessels are present within the lung. Panel (f) focuses on the GGO mask alone, while (g) overlays this GGO mask to further distinguish GGO regions from other structures. Finally, panel (h) presents the comprehensive final overlay, combining the lung, GGO, and vessel masks to provide a complete view of the segmented features.
[0090] This sequential processing approach effectively visualizes and distinguishes various anatomical structures and abnormalities in the CT images. The progression from initial lung masking to final comprehensive overlays enables detailed analysis and characterization of GGOs and associated vessels, which is critical for diagnostic and research purposes.
[0091] FIG. 6 illustrates a schematic diagram (a) chest CT image, (b) FD image, and (c) segmented GGO image, in accordance with an exemplary embodiment of the present disclosure.
[0092] As illustrated, the application of fractal dimension analysis enhances the visualization and segmentation of GGOs. The FD image provides insights into the spatial complexity of GGOs, while the segmented image offers a clear delineation of GGO regions, facilitating a more precise assessment of their characteristics.
[0093] Table 1 (as reproduced below) summarizes key statistical parameters that describe the texture and complexity of GGOs.
Table 1: Fractal Dimension (FD) and Lacunarity Analysis for Selected GGO Area
Average FD for selected area 1.8569
Standard deviation of FD for selected area 0.3773
Lacunarity for selected area 0.0413
The average FD indicates the overall complexity of the GGO patterns, while the standard deviation reflects variability in these patterns. Lacunarity measures the distribution and texture of the GGOs, providing additional context for understanding their nature and severity
[0094] FIG. 7 illustrates a schematic diagram (a) input image and (b) FD analysis of GGO in Region of Interest (ROI), in accordance with an exemplary embodiment of the present disclosure.
[0095] As illustrated, the application of fractal dimension analysis focuses region of interest within the CT image. The FD analysis provides insights into the structural complexity of GGOs within the ROI, aiding in the assessment and diagnosis of lung abnormalities.
[0096] Table 2 (as reproduced below) provides the average FD of 1.7731, a standard deviation of 0.3706, and a lacunarity of 0.0437 for the GGO area previously selected. It offers a detailed statistical analysis of the GGO area's fractal properties.
Table 2: Fractal Dimension (FD) and Lacunarity Analysis for the above selected GGO
Average FD for selected area 1.7731
Standard deviation of FD for selected area 0.3706
Lacunarity for selected area 0.0437
The FD and lacunarity metrics help quantify the texture and distribution of GGOs, contributing to a deeper understanding of their characteristics and potential implications for lung health.
[0097] FIG. 8 illustrates a schematic diagram (a) chest CT image, and (b) segmented GGO image, in accordance with an exemplary embodiment of the present disclosure.
[0098] As illustrated, the experimental results show the enhanced efficacy of the present invention in accurately detecting and analysing GGO patterns in chest CT images. Combining the adaptive pixel intensity filtering approach with fractal analysis improved the precision of GGO segmentation compared to traditional methods. Also, the subtraction of vessel masks from lung masks (M1M2) presented better isolation of GGOs, reducing false positives and improving the detection of clinically relevant regions.
[0099] Thus, the present invention offers a novel, efficient, and less resource-intensive solution, assisting radiologists in making informed diagnoses and improving patient outcomes. By automating the analysis process, the present invention reduces the risk of human error and enhances the reliability of GGO detection in clinical settings.
[00100] The present invention utilizes the adaptive pixel intensity filtering in the lung masks extraction stage. While traditional methods often cannot clearly outline lung regions in cases of overlapping anatomical structures, the present invention utilizes a selected range of pixel intensities between 1 and 190 with great efficiency for the isolation of lung tissues. By narrowing down to this range, the present invention reduces noise and enhances the sharpness of the lung mask by quality and therefore provides a better base for analysis.
[00101] The present invention filters the secondary pixel intensity within a narrow range from 1 to 120 designed exclusively to include vessel and GGO. This step is critical because of the subtraction of M2 masks from M1 masks for obtaining GGO masks, M3. This subtraction allows GGOs to be isolated from other structures in the lung, ensuring that only those areas of interest are analyzed. The approach will allow for far more detailed GGO detection with even fewer risks of misclassification than any previously developed techniques.
[00102] The present invention provides an advanced approach for the analysis of complexity and distribution of GGOs by creating a fractal image that keeps the original texture of the GGO. The fractal dimensions and lacunarity are the quantitative measures for heterogeneity and variability in structure inside the GGO patterns. This level of analysis is particularly original, as it moves the observation from a simple visual inspection to a more complete and objective evaluation of the severity of lung pathology. Grading of texture and distribution of GGOs by lacunarity analysis is an important development that arms the clinician with a useful tool both to monitor disease evolution and to personalize the therapeutic intervention.
[00103] The present invention can be utilized in multiple ways/places. For example:
• The present invention is very much essential in clinical settings such as primary health care centres, where the system can be used for mass screening and early diagnosis.
• The present invention can be utilized in a wide range of areas that specializes in pulmonary, cardiac related healthcare centres, hospitals and out-patient clinics.
• The techniques of the present invention could easily be deployed in any setting that has minimum computer facilities as it involves resource-intensive processes.
• The present invention sets out to be used even in remote healthcare regions and intense military and naval zones for disease diagnosis.
[00104] A few of the major advantages/technical effects of the present invention over the conventional solutions:
[00105] The technical effects of the present invention include greatly improving the identification, extraction, analysis, and visualization of GGO patterns in CT images. Adaptive pixel thresholding allows for improvements in detecting GGOs through dynamic threshold adjustment in view of pixel intensity for the exact segregation of GGOs from their surrounding tissues. This is further combined with fractal dimension and lacunarity analysis, hence providing a high level of detail in characterizing GGO patterns. This could be used to clearly differentiate various types of GGOs and judge their severities more accurately.
[00106] The present invention further enhances segmentation and visualization by improving the image through multi-stage masking. It makes possible a clean and stepwise visualization of the various anatomical regions and abnormalities in sequence through the processing and overlay of the lung and GGO masks, respectively. This enables the correct delineation of the GGOs and their respective structures. The final overlaid images combine both the lung, GGO, and vessel masks to provide a full overview of the segmented regions, enhancing visualization and interpretation of the CT images.
[00107] The addition of fractal dimension and lacunarity metrics significantly enhances the capability of quantitative analysis. These objective metrics provide important information about the complexity and texture of GGOs, which are very valuable for assessing lung disease aggressiveness and temporal changes. Average measurements, including their standard deviations, of the FD and lacunarity measures support diagnostic and prognostic evaluations.
[00108] Moreover, the invention introduces automation and efficiency in the processing pipeline by offering a computer-aided diagnosis system that could automatically delineate and characterize the GGO structure. It reduces manual effort and time required for analysis, hence promoting efficiency in the pipeline. In its principle, the adaptive algorithms used here are robust to variations in CT image quality and intensity levels, making the system versatile and reliable.
[00109] Those of skill in the art would appreciate that the various illustrative blocks, modules, elements, components, methods, and algorithms described herein may be implemented as electronic hardware, computer software, or combinations of both. To illustrate this interchangeability of hardware and software, various illustrative blocks, modules, elements, components, methods, and algorithms have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application. Various components and blocks may be arranged differently (e.g., arranged in a different order, or partitioned in a different way) all without departing from the scope of the subject technology.
[00110] It is understood that any specific order or hierarchy of blocks in the processes disclosed is an illustration of example approaches. Based upon design preferences, it is understood that the specific order or hierarchy of blocks in the processes may be rearranged, or that all illustrated blocks be performed. Any of the blocks may be performed simultaneously. In one or more implementations, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the implementations described above should not be understood as requiring such separation in all implementations, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.
[00111] It should be noted that the description and figures merely illustrate the principles of the present subject matter. It should be appreciated by those skilled in the art that conception and specific embodiment disclosed may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present subject matter. It should also be appreciated by those skilled in the art by devising various systems that, although not explicitly described or shown herein, embody the principles of the present subject matter and are included within its spirit and scope.
[00112] Furthermore, all examples recited herein are principally intended expressly to be for pedagogical purposes to aid the reader in understanding the principles of the present subject matter and the concepts contributed by the inventor(s) to further the art and are to be construed as being without limitation to such specifically recited examples and conditions. The novel features which are believed to be characteristic of the present subject matter, both as to its organization and method of operation, together with further objects and advantages will be better understood from the following description when considered in connection with the accompanying figures.
[00113] Although embodiments for the present subject matter have been described in language specific to package features, it is to be understood that the present subject matter is not necessarily limited to the specific features described. Rather, the specific features and methods are disclosed as embodiments for the present subject matter. Numerous modifications and adaptations of the system/device of the present invention will be apparent to those skilled in the art, and thus it is intended by the appended claims to cover all such modifications and adaptations which fall within the scope of the present subject matter. 
, Claims:We claim:
1. An improved method (100) of assessment of Ground Glass Opacity, GGO, in a medical computed tomography, CT, image, the method comprising:
a) acquiring (102) the CT image;
b) performing (104) a first filtering of pixels of the CT image, with pixel intensities from a first value to a second value, to extract lung masks comprising two largest blobs in the CT image;
c) performing (106) a second filtering of pixels of the CT image, with pixel intensities from the first value to a third value, to extract vessel and GGO masks, wherein the second value is greater than the third value;
d) eliminating (108) the vessel masks from the lung masks, to segment out only the GGO masks;
e) creating (110) a fractal image of the segmented GGO masks, while preserving an original texture of the GGO masks;
f) determining (112) fractal dimensions based on the fractal image of the segmented GGO masks; and
g) performing (114) lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, to quantify texture and distribution of the GGO patterns in the CT image.
2. The method (100) as claimed in claim 1, wherein the first, second, and third values are 1, 190, and 120, respectively.
3. The method (100) as claimed in claim 1, wherein the lacunarity analysis grades the texture and distribution of the GGOs in the CT image, to determine an extent of pathology in lungs.
4. The method (100) as claimed in claim 1, comprising determining an average fractal dimension for a selected GGO area in the segmented GGO masks, and a standard deviation of the fractal dimension of the selected area.
5. The method (100) as claimed in claim 4, wherein the average fractal dimension indicates the overall complexity of the GGO patterns in the CT image.
6. The method (100) as claimed in claim 4, wherein the standard deviation reflects variability in the GGO patterns in the CT image.
7. The method (100) as claimed in claim 1, wherein the lung masks, vessel and GGO masks, and GGO masks are overlaid in a predetermined first, second, and third colors, respectively, to distinguish one of the other masks.
8. The method (100) as claimed in claim 1, wherein the fractal image provides insights into the spatial complexity of GGO patterns in the CT image.
9. A system (200) for assessment of Ground Glass Opacity, GGO, in a medical computed tomography, CT, image, the system comprising:
a) a CT module (202) configured for capturing the CT image;
b) a filtering module (204) coupled to the CT module and configured for receiving the CT image subsequently performing (i) a first filtering of pixels of the CT image, with pixel intensities from a first value to a second value, to extract lung masks comprising two largest blobs in the CT image, and (ii) a second filtering of pixels of the CT image, with pixel intensities from the first value to a third value, to extract vessel and GGO masks, wherein the second value is greater than the third value;
c) an eliminating module (206) coupled to the filtering module and configured for eliminating the vessel masks from the lung masks, to segment out only the GGO masks;
d) a fractal image module (208) coupled to the eliminating module and configured for creating a fractal image of the segmented GGO masks, while preserving an original texture of the GGO masks;
e) a fractal dimension (210) module coupled to the fractal image module and configured for determining fractal dimensions based on the fractal image of the segmented GGO masks; and
f) a lacunarity analysis module (212) coupled to the fractal dimension module and configured for performing lacunarity analysis of the segmented GGO masks, based on the fractal dimension and the fractal image, to quantify texture and distribution of the GGO patterns in the CT image.
10. The system (200) as claimed in claim 9, wherein the first, second, and third values are 1, 190, and 120, respectively.
11. The system (200) as claimed in claim 9, wherein the lung masks, vessel and GGO masks, and GGO masks are overlaid in a predetermined first, second, and third colors, respectively, to distinguish one of the other masks.

Documents