SYSTEM AND METHOD FOR PREDICTING AND ASSESSING SEVERITY OF DISEASE USING DEEP LEARNING MODELS

Abstract

The present disclosure relates to a system for or predicting and assessing severity of disease using deep learning models. The system includes processors (202) that receives real-time data of user, real-time data pertains to input image obtained through Magnetic resonance imaging (MRI) scan. The processors pre-process real-time data by performing noise reduction, image enhancement, and normalization to obtain pre-processed real-time data. The processors (202) extract image features from pre-processed real-time data, image features include edges, corners, textures, and color. The processors (202) classify extracted image features and compare with pre-stored annotated dataset trained on pre-stored deep learning models, pre-stored annotated dataset includes labeled data that assist in distinguishing between healthy classes or disease classes. The processors (202) predict disease and assess severity of disease based on classified image features using Pufferfish Optimization Algorithm (POA)-based model.

Specification

Description:TECHNICAL FIELD
[0001] The present disclosure relates to a field of a deep-learning technology. More precisely, the present disclosure relates to a system and method for predicting and assessing severity of disease using deep-learning models.

BACKGROUND
[0002] The following description of the related art is intended to provide background information pertaining to the field of the present disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it should be appreciated that this section is used only to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of the prior art.
[0003] Medical imaging has become a crucial tool in modern healthcare, particularly for disease classification and diagnosis, allowing clinicians to visualize, assess, and understand complex internal structures. In the context of respiratory illnesses, especially COVID-19 and pneumonia, imaging modalities like X-rays and CT scans offer invaluable insights. These methods enable medical professionals to detect anomalies and gauge disease severity, playing a significant role in clinical decision-making. However, traditional diagnostic methodologies often encounter challenges when it comes to accurately categorizing diseases and assessing their severity, particularly with novel pathogens like COVID-19. Consequently, researchers are increasingly turning to advanced technologies, such as artificial intelligence (AI) and deep learning, to enhance diagnostic precision and reliability.
[0004] Current diagnostic systems face various limitations, from data quality issues to technological constraints. For instance, while AI-powered fungal detection presents a promising frontier, concerns about its accuracy due to similar disease presentations and data limitations must be addressed. Technical challenges, including infrastructure demands and privacy concerns, also present significant barriers, necessitating careful planning, cost considerations, and rigorous privacy safeguards before broad deployment.
[0005] Similarly, the application of deep learning to X-ray scans for COVID-19 detection has shown promising results; however, these systems often struggle with data quality issues and lack external validation, which is crucial for clinical adoption. Without further research and external validation, deploying these models in real-world settings may carry risks, as their readiness for clinical use remains uncertain.
[0006] Cloud-based services integrated with IoT for remote healthcare access offer another solution with great potential. These technologies could enable continuous patient monitoring and facilitate remote diagnostics, which is particularly beneficial in managing infectious diseases like COVID-19. Yet, these systems are not without limitations: data breaches pose privacy risks, technical challenges may limit accessibility, and the risk of algorithmic bias could lead to inequitable healthcare delivery. Additionally, cost burdens and the digital divide continue to challenge widespread implementation, making patient well-being and equity primary considerations in the responsible deployment of these systems.
[0007] To enhance diagnostic capabilities, recent methodologies have applied machine-learned networks in the form of deep three-dimensional convolutional neural networks (3D CNNs) for image-to-image processing. These advanced models are particularly effective in handling high-dimensional medical imaging data, capturing nuanced details in disease progression. This innovation in image processing through 3D CNNs supports more accurate and robust categorization, aiding clinicians in making informed decisions based on precise visual information.
[0008] Moreover, while AI and deep learning technologies offer transformative potential for disease detection and classification in medical imaging, existing challenges underscore the need for continuous refinement. Future developments in these domains must prioritize patient privacy, equity, and robust validation to support effective, responsible implementation in healthcare.
[0009] There is, therefore, a need in the art to provide a system and method that can overcome the shortcomings of the existing prior arts.


OBJECTS OF THE PRESENT DISCLOSURE
[00010] Some of the objects of the present disclosure, which at least one embodiment herein satisfies are as listed herein below.
[00011] It is an object of the present disclosure to provide a system and method for predicting and assessing severity of disease using deep-learning models.
[00012] It is another object of the present disclosure to provide a system and method for predicting and assessing severity of disease using deep-learning models, which classifies and assess severity of the disease using Magnetic resonance imaging (MRI) images.
[00013] It is another object of the present disclosure to provide a system and method for predicting and assessing severity of disease using deep-learning models, which enhances diagnostic precision and efficiency, contributing significantly to informed clinical decision-making during the global health crisis.

SUMMARY
[00014] This summary is provided to introduce certain objects and aspects of the present disclosure in a simplified form that are further described below in the detailed description. This summary is not intended to identify the key features or the scope of the claimed subject matter.
[00015] An aspect of the present disclosure relates to a system for predicting and assessing severity of disease using deep learning models. The system can include processors, and memory coupled to the processors, said memory having instructions executable by the processors to receive real-time data of at least one user by a data acquisition module, where the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan. The processors can pre-process the real-time data and perform noise reduction, image enhancement, and normalization to obtain pre-processed real-time data. The processors can extract image features from the pre-processed real-time data, where the image features can include edges, corners, textures, and color pertaining to the input image. The processors can classify the extracted image features and compare the extracted image features with a pre-stored annotated dataset trained on pre-stored deep learning models, where the pre-stored annotated dataset can include labeled data that assist in distinguishing between healthy classes or disease classes. The processors can predict disease and assess severity of the disease based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.
[00016] In an aspect, a method for predicting and assessing severity of disease based on Magnetic resonance imaging (MRI) scans, the method includes the steps of receiving, by one or more processors, real-time data of at least one user, where the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan. The method includes the steps of pre-processing, by the one or more processors, the real-time data to reduce noise, enhance image, and normalize image to obtain pre-processed real-time data. The method includes the steps of extracting, by the one or more processors, a plurality of image features from the pre-processed real-time data, where the plurality of image features can include edges, corners, textures, and color pertains to the input image. The method includes the steps of classifying, by the one or more processors, the plurality of extracted image features and comparing with pre-stored annotated dataset trained on pre-stored deep learning models, where the pre-stored annotated dataset can include labeled data that assist in distinguishing between healthy classes or disease classes. The method includes the steps of predicting disease and assessing severity of the disease by the one or more processors based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.
[00017] Various objects, features, aspects, and advantages of the present disclosure will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which numerals represent like features.
[00018] Within the scope of this application, it is expressly envisaged that the various aspects, embodiments, examples, and alternatives set out in the preceding paragraphs, in the claims and/or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. Features described in connection with one embodiment are applicable to all embodiments, unless such features are incompatible.

BRIEF DESCRIPTION OF THE DRAWINGS
[00019] In the figures, similar components and/or features may have the same reference label. Further, various components of the same type may be distinguished by following the reference label with a second label that distinguishes among the similar components. If only the first reference label is used in the specification, the description is applicable to any one of the similar components having the same first reference label irrespective of the second reference label.
[00020] FIG. 1 illustrates an exemplary network architecture of the proposed system for predicting and assessing severity of disease using deep-learning models, by an embodiment of the present disclosure.
[00021] FIG. 2 illustrates an exemplary representation of system, in accordance with an embodiment of the present disclosure.
[00022] FIG. 3 illustrates a flow diagram illustrating a method for predicting and assessing severity of disease using deep-learning models, in accordance with an embodiment of the present disclosure.
[00023] FIG. 4 illustrates an exemplary computer system to implement the proposed system in accordance with embodiments of the present disclosure.

DETAILED DESCRIPTION
[00024] 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 scope of the present disclosure as defined by the appended claims.
[00025] In the following description, numerous specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent to one skilled in the art that embodiments of the present disclosure may be practiced without some of these specific details.
[00026] The present disclosure relates to a field of a deep-learning technology. More precisely, the present disclosure relates to a system and method for predicting and assessing severity of disease using deep-learning models.
[00027] An aspect of the present disclosure relates to a system can include one or more processors; a memory coupled to the one or more processors, where said memory stores instructions which when executed by the one or more processors cause the system to receive real-time data of the at least one user, the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan. The one or more processors can be configured to pre-process the real-time data, where the one or more processors may perform noise reduction, image enhancement, and normalization to obtain pre-processed real-time data. The one or more processors can be configured to extract a plurality of image features from the pre-processed real-time data, the plurality of image features can include edges, corners, textures, and color pertains to the input image. The one or more processors can be configured to classify the plurality of extracted image features and compare with pre-stored annotated dataset trained on pre-stored deep learning models, the pre-stored annotated dataset can include labeled data that assist in distinguishing between healthy classes or disease classes. The pre-stored deep learning models can include at least one of a deep learning Convolutional Neural Networks (CNN) model, and Recurrent Neural Networks (RNN) model, the deep learning Convolutional Neural Networks (CNN) model pertains to a DenseNet-121 CNN model. The pre-stored deep learning models are configured to identify and isolate infected regions of the body based on the plurality of classified image features. The one or more processors can be configured to predict disease and assess severity of the disease based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.
[00028] FIG. 1 illustrates an exemplary network architecture (100) of the proposed system for predicting and assessing severity of disease using deep-learning models, in accordance with an embodiment of the present disclosure.
[00029] In an embodiment, referring to FIG. 1, the network architecture (100) can include the system (102) which may be configured connect to a network (104), which is further connected to at least one computing device (108-1), (108-2), … (108-N) (collectively referred as computing device 108, herein) associated with one or more users (106-1), (106-2), … (106-N) (collectively referred as user 106, herein). In an implementation, the system (102) may facilitates an accurate and timely disease classification, particularly in medical imaging. The disease may include, but not limited to pneumonia, tuberculosis, chronic obstructive pulmonary disease (COPD), and other infections (COVID-19 disease) that impact the lungs. Similarly, the detection of multiple cancer types, such as lung, breast, brain, liver, and pancreatic cancers, using modalities like CT, MRI, and PET scans. Additionally, cardiovascular diseases such as arrhythmias, coronary artery disease, and heart failure, using imaging techniques such as echocardiograms and cardiac MRIs. For musculoskeletal diseases, imaging can help classify arthritis, osteoporosis, fractures, and soft tissue injuries with tools like X-rays, MRI, and ultrasound.
[00030] In an exemplary embodiment, the computing device (108) may include, but not be limited to, a computer-enabled device, a mobile phone, a smartphone, a tablet, or some combination thereof. A person of ordinary skill in the art will understand that the at least one computing device (108) may be individually referred to as a computing device and collectively referred to as a computing devices (108). The computing device (108) may be associated with at least one user (106). At least one user may include, but not limited to an individual, a doctor, a physician, a surgeon, a specialist, a patient, and the like.
[00031] In an exemplary embodiment, the network (104) may include, but not be limited to, at least a portion of one or more networks having one or more nodes that transmit, receive, forward, generate, buffer, store, route, switch, process, or a combination thereof, etc. one or more messages, packets, signals, waves, voltage or current levels, some combination thereof, or so forth. In an exemplary embodiment, the network (104) may include, but not be limited to, a wireless network, a wired network, an internet, an intranet, a public network, a private network, a packet-switched network, a circuit-switched network, an ad hoc network, an infrastructure network, a Public-Switched Telephone Network (PSTN), a cable network, a cellular network, a satellite network, a fiber optic network, or some combination thereof.
[00032] In another exemplary embodiment, the centralized server (110) may include or comprise, by way of example but not limitation, one or more of: a stand-alone server, a server blade, a server rack, a bank of servers, a server farm, hardware supporting a part of a cloud service or system, a home server, hardware running a virtualized server, one or more processors executing code to function as a server, one or more machines performing server-side functionality as described herein, at least a portion of any of the above, some combination thereof. In an embodiment, the system (102) may be coupled to the centralized server (110). In another embodiment, the centralized server (110) may also be operatively coupled to the computing devices (108). In some implementations, the system (102) may be associated with the centralized server (110).
[00033] In an embodiment, the system (102) can include one or more processors (refer FIG. 2); a memory (refer FIG. 2) coupled to the one or more processors, where said memory stores instructions which when executed by the one or more processors cause the system (102) to receive real-time data of the at least one user (106), the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan. The one or more processors can be configured to pre-process the real-time data, where the one or more processors may perform noise reduction, image enhancement, and normalization to obtain pre-processed real-time data. The one or more processors can be configured to extract a plurality of image features from the pre-processed real-time data, the plurality of image features can include edges, corners, textures, and color pertains to the input image. The one or more processors can be configured to classify the plurality of extracted image features and compare with pre-stored annotated dataset trained on pre-stored deep learning models, the pre-stored annotated dataset can include labeled data that assist in distinguishing between healthy classes or disease classes. The pre-stored deep learning models can include at least one of a deep learning Convolutional Neural Networks (CNN) model, and Recurrent Neural Networks (RNN) model, the deep learning Convolutional Neural Networks (CNN) model pertains to a DenseNet-121 CNN model. The pre-stored deep learning models are configured to identify and isolate infected regions of the body based on the plurality of classified image features. The one or more processors can be configured to predict disease and assess severity of the disease based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.
[00034] In an embodiment, the Pufferfish Optimization Algorithm (POA)-based model is configured to adjust weights, bias terms, and other hyperparameters of the pre-stored deep learning models, the Pufferfish Optimization Algorithm (POA)-based model can be configured to utilize the plurality of extracted image features to fine-tune the hyperparameters until the pre-stored deep learning models achieve optimal prediction accuracy during the training process.
[00035] In an embodiment, the system (102) can be configured to generate visual indicators based on the predicted disease, and integrate diagnostic findings with patient's existing medical data to deliver a comprehensive view of the patient's health, the visual indicators can include heat maps, highlighting the infected regions in the plurality of classified image features most likely to be affected by the disease.
[00036] In an embodiment, the system (102) can be configured to assess performance metrics of the pre-stored deep learning models using cross-validation techniques, the system (102) can be configured to compare the predicted disease with the pre-stored annotated dataset to determine the performance metrics of the pre-stored deep learning model, the performance metrics pertain to sensitivity and specificity of the pre-stored deep learning model which assists in identifying the disease classes versus the healthy classes. The cross-validation techniques can be configured to divide the pre-stored annotated dataset into multiple subsets to ensure the model's robustness and generalization across different patient populations. The one or more processors can be configured to optimize the pre-stored deep learning models using a feedback loop based on the evaluation results, the optimization can include adjusting model parameters or retraining the pre-stored deep learning models with additional annotated data to improve accuracy and reduce false positives/negatives.
[00037] FIG. 2 illustrates an exemplary representation of the system, in accordance with an embodiment of the present disclosure.
[00038] In an aspect, referring to FIG. 2, the system (102) may include one or more processor(s) (202). The one or more processor(s) (202) may be implemented as one or more microprocessors, microcomputers, microcontrollers, edge or fog microcontrollers, digital signal processors, central processing units, logic circuitries, and/or any devices that process data based on operational instructions. Among other capabilities, one or more processor(s) (202) may be configured to fetch and execute computer-readable instructions stored in the memory (204) of the system (102). The memory (204) may be configured to store one or more computer-readable instructions or routines in a non-transitory computer-readable storage medium, which may be fetched and executed to create or share data packets over a network service. The memory (204) may include any non-transitory storage device including, for example, volatile memory such as Random Access Memory (RAM), or non-volatile memory such as Erasable Programmable Read-Only Memory (EPROM), flash memory, and the like.
[00039] Referring to FIG. 2, the system (102) may include an interface(s) (206). The interface(s) (206) may include a variety of interfaces, for example, interfaces for data input and output devices, referred to as I/O devices, storage devices, and the like. The interface(s) (206) may facilitate communication to/from the system (102). The interface(s) (206) may also provide a communication pathway for one or more components of the system (102). Examples of such components include but are not limited to, processing unit/engine(s) (208) and a local database (210).
[00040] In an embodiment, the processing unit/engine(s) (208) may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine(s) (208). In the examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing engine(s) (208) may be processor-executable instructions stored on a non-transitory machine-readable storage medium and the hardware for the processing engine(s) (208) may include a processing resource (for example, one or more processors), to execute such instructions. In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine(s) (208). In such examples, the system (102) may include the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to the system (102) and the processing resource. In other examples, the processing engine(s) (208) may be implemented by electronic circuitry.
[00041] In an embodiment, the local database (210) may include data that may be either stored or generated as a result of functionalities implemented by any of the components of the processor (202) or the processing engines (208). In an embodiment, the local database (210) may be separate from the system (102).
[00042] In an exemplary embodiment, the processing engine (208) may include one or more engines selected from any of data acquisition module (212), a pre-processing module (214), an extraction module (216), a classification module (218), a prediction module (220), a generating module (222), an evaluation module (224), and other modules (224) having functions that may include but are not limited to testing, storage, and peripheral functions, such as wireless communication unit for remote operation, audio unit for alerts and the like.
[00043] In an embodiment, the system (102) can include the data acquisition module (212) which can be configured to receive real-time data of at least one user, the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan. The real-time data may be received from sensors, web scraping or public databases that needs pre-processing.
[00044] In an embodiment, the system (102) can include the pre-processing module (214) which can be configured to pre-process the real-time data, the pre-processing module (214) can be configured to perform noise reduction, image enhancement, and normalization to obtain pre-processed real-time data. The pre-processing may involve cleaning the data, removing errors, and formatting the data in a way that the deep learning model can understand. Noise reduction helps remove unwanted data from the scans, while image enhancement improves the quality of the images i.e., 225 ∗ 225 image resolution. Once pre-processed, the data enters the Customized DenseNet-121 that extracts image features.
[00045] In an embodiment, the system (102) can include the extraction module (216) which can be configured to extract a plurality of image features from the pre-processed real-time data, where the plurality of image features can include edges, corners, textures, and color pertains to the input image.
[00046] In an embodiment, the system (102) can include the classification module (218) which may be configured to classify the plurality of extracted image features and compare with pre-stored annotated dataset trained on pre-stored deep learning models, the pre-stored annotated dataset can include labeled data that assist in distinguishing between healthy classes or disease classes. The deep learning model is trained with annotated data to identify the patterns such as edges, and texture to predict the class. The pre-stored deep learning models are configured to identify and isolate infected regions of the body based on the plurality of classified image features.
[00047] In an embodiment, the system (102) can include the prediction module (220) which may be configured to predict disease and assess severity of the disease based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model. The Pufferfish Optimization Algorithm (POA)-based model is configured to adjust weights, bias terms, and other hyperparameters of the pre-stored deep learning models, the Pufferfish Optimization Algorithm (POA)-based model can be configured to utilize the plurality of extracted image features to fine-tune the hyperparameters until the pre-stored deep learning models achieve optimal prediction accuracy during the training process. The pufferfish optimizer is an algorithm that is used to adjust the weights in the CNN to improve its performance. Hyperparameter tuning is the process of adjusting the settings of the CNN in order to improve its performance. Here the POA is going to fine-tune the hyperparameters such as weights, bias term.
[00048] In an embodiment, the system (102) can include the generating module (222) can be configured to generate visual indicators based on the predicted disease, and integrate diagnostic findings with patient's existing medical data to deliver a comprehensive view of the patient's health, the visual indicators can include heat maps, highlighting the infected regions in the plurality of classified image features most likely to be affected by the disease.
[00049] The predicted disease or projections can also create heat maps showing the lung areas most likely to be affected by the disease. Finally, the results are integrated with the patient's data to provide a comprehensive view of their condition. Overall, the data processing system in the diagram is a complex but powerful tool for analyzing MRI scans. It can be used to diagnose a variety of diseases. It can also be used to track the progression of disease over time and to monitor the effectiveness of treatment. The performance of model may vary depending on the specific dataset used for training.
[00050] In an embodiment, the system (102) can include the evaluation module (224) can be configured to assess performance metrics of the pre-stored deep learning models using cross-validation techniques, the evaluation module (224) can be configured to compare the predicted disease with the pre-stored annotated dataset to determine the performance metrics of the pre-stored deep learning model, the performance metrics pertain to sensitivity and specificity of the pre-stored deep learning model which assists in identifying the disease classes versus the healthy classes. The cross-validation techniques configured to divide the pre-stored annotated dataset into multiple subsets to ensure the model's robustness and generalization across different patient populations. The one or more processors (202) which may be configured to optimize the pre-stored deep learning models using a feedback loop based on the evaluation results, the optimization can include adjusting model parameters or retraining the pre-stored deep learning models with additional annotated data to improve accuracy and reduce false positives/negatives.
[00051] In an exemplary embodiment, the system pertains to a diagnostic system for detecting COVID-19 from MRI scans, utilizing a deep-learning model based on a customized DenseNet-121 architecture. The system (102) receives MRI scans of patients as input, which are processed through multiple layers in the DenseNet-121 model, where each layer builds upon the previous to enhance information flow and extract intricate features from the imaging data. These extracted features are then passed to a classifier within the system (102), which labels each image as either positive (indicating the presence of COVID-19) or negative.
[00052] The system (102) leverages the DenseNet-121 model's unique connectivity pattern, which ensures efficient feature propagation, resulting in an accurate and robust diagnostic tool. Despite requiring a substantial amount of training data, this architecture achieves a high level of performance in identifying COVID-19 markers within MRI scans. However, recognizing that interpretability may be challenging due to the model's complexity, this embodiment is intended to complement other diagnostic methods rather than serve as a standalone diagnostic tool. This multi-modal approach enhances diagnostic reliability in clinical settings, providing a comprehensive assessment that supports informed medical decisions.
[00053] FIG. 3 illustrates a flow diagram illustrating a method for predicting and assessing severity of disease using deep-learning models, in accordance with an embodiment of the present disclosure.
[00054] As illustrated, method (300) includes, at block (302), receiving, by one or more processors, real-time data of at least one user, wherein the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan.
[00055] Continuing further, method (300) includes, at block (304), pre-processing, by the one or more processors, the real-time data to reduce noise, enhance image, and normalize image to obtain pre-processed real-time data.
[00056] Continuing further, method (300) includes, at block (306), extracting, by the one or more processors, a plurality of image features from the pre-processed real-time data, where the plurality of image features can include edges, corners, textures, and color pertains to the input image.
[00057] Continuing further, method (300) includes, at block (308), classifying, by the one or more processors, the plurality of extracted image features and comparing with pre-stored annotated dataset trained on pre-stored deep learning models, where the pre-stored annotated dataset can include labeled data that assist in distinguishing between healthy classes or disease classes.
[00058] Continuing further, method (300) includes, at block (310), predicting disease and assessing severity of the disease by the one or more processors based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.
[00059] FIG. 4 illustrates an exemplary computer system to implement the proposed system in accordance with embodiments of the present disclosure.
[00060] As illustrated in FIG. 4, a computer system (400) can include an external storage device (410), a bus (420), a main memory (430), a read only memory (440), a mass storage device (450), communication port (460), and a processor (470). A person skilled in the art will appreciate that the computer system may include more than one processor and communication ports. Examples of processor (470) include, but are not limited to, an Intel® Itanium® or Itanium 2 processor(s), or AMD® Opteron® or Athlon MP® processor(s), Motorola® lines of processors, FortiSOC™ system on chip processors or other future processors. Processor (470) may include various modules associated with embodiments of the present disclosure. Communication port (460) can be any of an RS-232 port for use with a modem based dialup connection, a 10/100 Ethernet port, a Gigabit or 10 Gigabit port using copper or fiber, a serial port, a parallel port, or other existing or future ports. Communication port (460) may be chosen depending on a network, such a Local Area Network (LAN), Wide Area Network (WAN), or any network to which computer system connects.
[00061] Memory (430) can be Random Access Memory (RAM), or any other dynamic storage device commonly known in the art. Read-only memory (440) can be any static storage device(s) e.g., but not limited to, a Programmable Read Only Memory (PROM) chips for storing static information e.g., start-up or BIOS instructions for processor (470). Mass storage (450) may be any current or future mass storage solution, which can be used to store information and/or instructions. Exemplary mass storage solutions include, but are not limited to, Parallel Advanced Technology Attachment (PATA) or Serial Advanced Technology Attachment (SATA) hard disk drives or solid-state drives (internal or external, e.g., having Universal Serial Bus (USB) and/or Firewire interfaces), e.g. those available from Seagate (e.g., the Seagate Barracuda 7102 family) or Hitachi (e.g., the Hitachi Deskstar 7K1000), one or more optical discs, Redundant Array of Independent Disks (RAID) storage, e.g. an array of disks (e.g., SATA arrays), available from various vendors including Dot Hill Systems Corp., LaCie, Nexsan Technologies, Inc. and Enhance Technology, Inc.
[00062] Bus (420) communicatively couple processor(s) (470) with the other memory, storage and communication blocks. Bus 420 can be, e.g. a Peripheral Component Interconnect (PCI) / PCI Extended (PCI-X) bus, Small Computer System Interface (SCSI), USB or the like, for connecting expansion cards, drives and other subsystems as well as other buses, such a front side bus (FSB), which connects processor (470) to software system.
[00063] Optionally, operator and administrative interfaces, e.g. a display, keyboard, and a cursor control device, may also be coupled to bus (420) to support direct operator interaction with a computer system. Other operator and administrative interfaces can be provided through network connections connected through communication port (460). The external storage device (410) can be any kind of external hard-drives, floppy drives, IOMEGA® Zip Drives, Compact Disc - Read Only Memory (CD-ROM), Compact Disc-Re-Writable (CD-RW), Digital Video Disk-Read Only Memory (DVD-ROM). Components described above are meant only to exemplify various possibilities. In no way should the aforementioned exemplary computer system limit the scope of the present disclosure.
[00064] If the specification states a component or feature "may", "can", "could", or "might" be included or have a characteristic, that particular component or feature is not required to be included or have the characteristic.
[00065] As used in the description herein and throughout the claims that follow, the meaning of "a," "an," and "the" includes plural reference unless the context clearly dictates otherwise. Also, as used in the description herein, the meaning of "in" includes "in" and "on" unless the context clearly dictates otherwise.
[00066] Moreover, in interpreting the specification, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms "comprises" and "comprising" should be interpreted as referring to elements, components, or steps in a non-exclusive manner, indicating that the referenced elements, components, or steps may be present, or utilized, or combined with other elements, components, or steps that are not expressly referenced. Where the specification claims refer to at least one of something selected from the group consisting of A, B, C ….and N, the text should be interpreted as requiring only one element from the group, not A plus N, or B plus N, etc.
[00067] While the foregoing describes various embodiments of the proposed disclosure, other and further embodiments of the proposed disclosure may be devised without departing from the basic scope thereof. The scope of the proposed disclosure is determined by the claims that follow. The proposed disclosure is not limited to the described embodiments, versions or examples, which are included to enable a person having ordinary skill in the art to make and use the invention when combined with information and knowledge available to the person having ordinary skill in the art.

ADVANTAGES OF THE PRESENT DISCLOSURE
[00068] The present disclosure provides a system and method for predicting and assessing severity of disease using deep learning models.
[00069] The present disclosure provides a system and method that achieves high accuracy in complex prediction tasks, such as identifying disease markers in images.
[00070] The present disclosure provides a system and method that automates the data acquisition, cleaning, and processing stages, which saves time and reduces the potential for human error. Automated pre-processing makes the approach scalable across large datasets from varied sources.
[00071] The present disclosure provides a system and method that identifies intricate patterns (such as edges and textures in images) effectively that are not easily recognizable by simpler algorithms, making them suitable for tasks like image analysis or natural language processing.
[00072] The present disclosure provides a system and method that utilizes K-fold cross-validation and provides a robust way to test model performance on different subsets of data, which helps in assessing its generalization capabilities and minimizes overfitting.
[00073] The present disclosure provides a system and method that visualizes results in charts or graphs makes it easier for users and stakeholders to understand and interpret model outcomes, facilitating data-driven decisions.
 
, Claims:1. A system (102) for predicting and assessing severity of disease using deep learning models, the system (102) comprising:
one or more processors (202); and
at least one memory (204) coupled to the one or more processors (202), said memory (204) having instructions executable by the one or more processors (202) to:
receive real-time data of at least one user by a data acquisition module (212), wherein the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan;
pre-process the real-time data by a pre-processing module (214), wherein the pre-processing module (214) is configured to perform noise reduction, image enhancement, and normalization to obtain pre-processed real-time data;
extract a plurality of image features from the pre-processed real-time data by an extraction module (216), wherein the plurality of image features comprising edges, corners, textures, and color pertains to the input image;
classify the plurality of extracted image features by a classification module (218) and compare with pre-stored annotated dataset trained on pre-stored deep learning models, wherein the pre-stored annotated dataset comprising labeled data that assist in distinguishing between healthy classes or disease classes; and
predict disease and assess severity of the disease by prediction module (220) based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.
2. The system as claimed in claim 1, wherein the pre-stored deep learning models comprising at least one of a deep learning Convolutional Neural Networks (CNN) model, and Recurrent Neural Networks (RNN) model,
wherein the deep learning Convolutional Neural Networks (CNN) model pertains to a DenseNet-121 CNN model.
3. The system as claimed in claim 1, wherein the pre-stored deep learning models are configured to identify and isolate infected regions of the body based on the plurality of classified image features.
4. The system as claimed in claim 1, wherein the Pufferfish Optimization Algorithm (POA)-based model is configured to adjust weights, bias terms, and other hyperparameters of the pre-stored deep learning models,
wherein the Pufferfish Optimization Algorithm (POA)-based model configured to utilize the plurality of extracted image features to fine-tune the hyperparameters until the pre-stored deep learning models achieve optimal prediction accuracy during the training process.
5. The system as claimed in claim 1, wherein the system (102) comprising a generating module (222) configured to generate visual indicators based on the predicted disease, and integrate diagnostic findings with patient's existing medical data to deliver a comprehensive view of the patient's health,
wherein the visual indicators comprising heat maps, highlighting the infected regions in the plurality of classified image features most likely to be affected by the disease.
6. The system as claimed in claim 1, wherein the system (102) comprising an evaluation module (224) configured to assess performance metrics of the pre-stored deep learning models using cross-validation techniques,
wherein the evaluation module (224) configured to compare the predicted disease with the pre-stored annotated dataset to determine the performance metrics of the pre-stored deep learning model,
wherein the performance metrics pertain to sensitivity and specificity of the pre-stored deep learning model which assists in identifying the disease classes versus the healthy classes.
7. The system as claimed in claim 6, wherein the cross-validation techniques configured to divide the pre-stored annotated dataset into multiple subsets to ensure the model's robustness and generalization across different patient populations.

8. The system as claimed in claim 1, wherein the one or more processors (202) configured to optimize the pre-stored deep learning models using a feedback loop based on the evaluation results,
wherein the optimization comprising adjusting model parameters or retraining the pre-stored deep learning models with additional annotated data to improve accuracy and reduce false positives/negatives.
9. A method for predicting and assessing severity of disease based on Magnetic resonance imaging (MRI) scans, the method (102) comprising:
receiving, by one or more processors, real-time data of at least one user, wherein the real-time data pertains to an input image obtained through a Magnetic resonance imaging (MRI) scan;
pre-processing, by the one or more processors, the real-time data to reduce noise, enhance image , and normalize image to obtain pre-processed real-time data;
extracting, by the one or more processors, a plurality of image features from the pre-processed real-time data, wherein the plurality of image features comprising edges, corners, textures, and color pertains to the input image;
classifying, by the one or more processors, the plurality of extracted image features and comparing with pre-stored annotated dataset trained on pre-stored deep learning models, wherein the pre-stored annotated dataset comprising labeled data that assist in distinguishing between healthy classes or disease classes; and
predicting disease and assessing severity of the disease by the one or more processors based on the plurality of classified image features using a Pufferfish Optimization Algorithm (POA)-based model.

Documents