AU2021102593A4

AU2021102593A4 - A Method for Detection of a Disease

Info

Publication number: AU2021102593A4
Application number: AU2021102593A
Authority: AU
Inventors: Sandeep Kumar Bhakhar; Jeewan Bhatia; Kamlesh Gautam; Ankita Gupta; Abhishek Singh Kilak; Rakesh Kumar; Rotash Kumar; Seema Rani; Kiran Deep Singh; Prabh Deep Singh
Original assignee: Individual
Current assignee: Individual
Priority date: 2021-05-15
Filing date: 2021-05-15
Publication date: 2021-07-01
Anticipated expiration: 2029-05-15

Abstract

The present disclosure seeks to provide a method for detection of a disease. Developing particle swarm optimization assisted support vector machine based diagnostic system for lung cancer prediction at the early stage. In this disclosure, the integration of Particle swarm optimization (PSO) with a support vector machine (SVM) is proposed to detect lung cancer at a very early stage. The proposed technique is verified by using the various standard lung cancer classification data sets. The comparison is drawn among the proposed and the existing technique based upon the various standard quality metrics of the data mining. Experimental results indicate that the proposed algorithm is more efficient than existing techniques. In the future, the suggested method would help recommend diagnosing cancer in various human organs. 15 100 O btaining a plurality of information of the disease, wherein pre-processing the plurality of informationtoenhancethe quality of information usirg pre-processing module. 102 Extracting a plurality of features from the processed information using a feature extraction 104 module,whereinapluralityofattributes relatedtothediseaseisobtainedduringextraction. Creating a population of the extracted attributes acqui red fro the feature extraction module using a K-rneanrs technique.10 Evaluatinga position of each of the extracted attribute inthe population usinganobjectye 108 function, wherei narn inirn urnrootrneanrsqua re error is used as theobjective function. Determi ni rg a best pos ition of each of the exracted attri bute anrd pl atirig each of the extracted attributeateachoftheestposition,whereinabestattributeisselectedforea chbestpositionon n 110 order to opfti mize the extracted attri butes for disease detectiort 14 Figure 1 Steps Desenpoon Algoihm Pouanonotagents parameters p Posionofagenta in thesoluiionspace f Objecne function 1 Velocityofagentsa Pa o Neighborboodofagent a(fixed) The neighborhood concept i PSO is not the samine as the one used in another metaheunsncs sech sance nPSO cachparticle'sneighborhoodnever Chan ges (is fixed) Required Number ofparticles usually between 10 and 50 attributes C, is the importance ofpesenal bes value with C: is the unportance ofsaghboboodbest value values Usually C. - C: 4 (empirically chosenvalue) If velocity is too low- algorithm itoo slow If velocity is too high -algorithm too unable Particle P = P - t' wvrh update v =-cl * rand •(pets- p) rule - c2 - rand • tgqest - p) whert parocle-s position, path direction c; the weight of local information. :the weight ofglobal information,pBesr the best position of the artcle.gBer the best position ofthe swuarm.and rand random variable Algofithms P = Parrclejntrah:artron(e For - I to msrt For each partre p ia P do Ip - 1(p) If fp is better than fpaest) pBest p; end endyffest - best p in P. For enA ;prtree p i n P do r- r• cl rand • p~est-p) - c2 • rand • (gest - p). p = p - r.%endend PSD based I Createa'popaoen'ofatunbutes(paetdes)fremdevelopedclustersof K-means data set X usingthe K-reansalgoithm algorithm 2 Evaluat each particle's position according to the obedse fiMctionse for minimum root mean squared errer (RMSE) training 3 If a particle's curri position is better than its previous bes position, update it 4 Determine the best particle accedingg to the partide's preios bt positions) 5 Rerurnopmizedtrained model Figure2

Description

O btaining a plurality of information of the disease, wherein pre-processing the plurality of informationtoenhancethe quality of information usirg pre-processing module. 102

Extracting a plurality of features from the processed informationusing a feature extraction 104 module,whereinapluralityofattributes relatedtothediseaseisobtainedduringextraction.

Creating a population of the extracted attributes acqui red fro the feature extraction module using a K-rneanrs technique.10

Evaluatinga position of each of the extracted attribute inthe population usinganobjectye 108 function, wherei narn inirn urnrootrneanrsqua re error is used as theobjective function.

Determi ni rg a best pos ition of each of the exracted attri bute anrd pl atirig each of the extracted attributeateachoftheestposition,whereinabestattributeisselectedforea chbestpositionon n 110 order to opfti mize the extracted attri butes for disease detectiort 14

Figure 1

Steps Desenpoon Algoihm Pouanonotagents parameters p Posionofagenta inthesoluiionspace f Objecne function 1 Velocityofagentsa Pa o Neighborboodofagent a(fixed) The neighborhood concept i PSO is not the samine as the one used in another metaheunsncs sechsance nPSO cachparticle'sneighborhoodnever Chan ges (is fixed) Required Number ofparticles usually between 10 and 50 attributes C, is the importance ofpesenal bes value with C: is the unportance ofsaghboboodbest value values Usually C. - C: 4 (empirically chosenvalue) If velocity is too low- algorithm itoo slow If velocity is too high -algorithm too unable Particle P = P - t' wvrh update v =-cl * rand •(pets- p) rule - c2 - rand • tgqest - p) whert parocle-s position, path direction c; the weight of local information. :the weight ofglobal information,pBesr the best position of the artcle.gBer the best position ofthe swuarm.and rand random variable

Algofithms P = Parrclejntrah:artron(e For - I to msrt For each partre p ia Pdo Ip - 1(p) If fp is better than fpaest) pBest p; end endyffest - best p in P. For enA ;prtreep i n P do r- r• cl rand • p~est-p) - c2 • rand • (gest- p). p= p- r.%endend PSD based I Createa'popaoen'ofatunbutes(paetdes)fremdevelopedclustersof K-means data set X usingthe K-reansalgoithm algorithm 2 Evaluat each particle's position according to the obedse fiMctionse for minimum root mean squared errer (RMSE) training 3 If a particle's curri position is better than its previous bes position, update it 4 Determine the best particle accedingg to the partide's preios bt positions) 5 Rerurnopmizedtrained model

Figure2

A Method for Detection of a Disease

FIELD OF THE INVENTION

The present disclosure relates to a method for detection of a disease. It further relates to developing particle swarm optimization assisted support vector machine based diagnostic system for lung cancer prediction at the early stage.

BACKGROUND OF THE INVENTION

Lung cancer, like other cancers, is unregulated irregular lung tissue development. During formation, cells migrate across the lung by metastasizing through surrounding tissue or other areas of the body. This cell development gradually creates a tumor mass. Three major forms of lung cancer include non-small cell lung cancer, small cell lung cancer, and lung cancer. NSCLC is the most prevalent form of lung cancer, around 85% of all lung cancers. Overall, a man's lifetime risk of contracting lung cancer is around 1 in 14; for a woman, it's around 1 in 17. Survival states the primary source of lung cancer (85%) is long term cigarette use. Data indicate a clear association between cigarette consumption and lung cancer. Lung cancer rises and declines are strongly parallel to tobacco use. As intake decreases, the occurrence of lung cancer rises. Tobacco use, passive smoking, radon radiation, genetic causes, and air quality also cause lung cancer. Healthcare services can never be isolated from machine learning technology growth. Machine learning technology plays a rising function in all stages of lung cancer such as lung field segmentation, bone suppression, and irregular tissue identification. Machine learning is a cross-discipline multi-domain. Machine learning's key objects are items, specific algorithms that boost their experience efficiency. Here characteristics are variables used to characterize entity trends. Extracting features requires identifying suitable variables to characterize artifacts. Feature classification is the method of making objective judgments based on factors used to classify artifacts. Machine learning is ideal for automated object detection. Machine learning technology learns trends from training data to detect artifacts, and the algorithm can continuously enhance its efficiency before convergence is accomplished through repetitive learning. Healthcare is an important aspect of human life and frequent analysis of crucial criteria since therapies are the fundamental feature of healthcare. These processes become much more important in addressing psychiatric illnesses that need qualified medical professionals to control their state. At distinct times, the current procedure for monitoring certain essential health metrics of healthcare is assessed by doctors or medical professionals. Furthermore, self-reported subjective health measurements are important to determine care efficacy. These methods also contribute to the lack of critical details on specific treatments during the day or night. These kinds of styles can be observed when a form of a summary of your insight information and facts, and also can be employed throughout even more investigation and also, to get a good example, throughout appliance discovering as well as predictive analytics. To give an example, the info mining phase could discover various teams through your data, which can be used to obtain more genuine prediction benefits utilizing a decision, help system. Or your current information selection, information and facts prep, neither outcome meaning as well as reporting will be the principle information pursuit phase but accomplish remain in the whole KDD method when even more steps. Machine Learning Process Details exploration is the period often utilized in pc science. Oahu is the approach used to acquire the actual helpful details via big data fixed making use of various techniques. A variety of data exploration algorithms are utilized to acquire details via the data fixed including Group, Clustering, Aggregation, and many more. The entire aim involving the data exploration approach is mostly to be able to acquire details via the data fixed as well as change the item into a superior easy to Understand variety which they can use further. Merely, getting helpful details via the data is called data mining. Abbreviations and Acronyms Data Mining In Healthcare The medical industry currently builds massive amounts of complex info regarding clients, medical centers resources, ailment a diagnosis, electric client files, healthcare devices, etc. In one of the solutions, they dealt with the particular exclusive top features of information mining by using healthcare data. They have discussed a variety of honest plus legalized areas linked to healthcare information mining just like information management, fear of legal cases, forecasted benefits, plus exclusive supervision issues. With this they stated which the numerical knowledge of approximation plus theory development in healthcare data are mainly not the same as all other information variety routines. In another solution, they came with market research on the obtainable materials on information mining utilizing gentle processing. A categorization offers been recently supplied predicated on various gentle computing devices and hybridizations put on, the data mining operate put on, plus the desire criterion identified by way of the unit. Inherited algorithms provide a successful look for algorithms to pick out one, via mixed advertising information, predicated on several choice criterion objective functions. Hard items are usually suitable to handle various types of concerns with data. Another prior art provides which the prevalent option of brand new computational techniques plus resources about information exam plus predictive modeling necessitates healthcare informatics gurus plus experts to help systematically select the most appropriate approach to cope with controlled forecast problems. An enormous assortment of these methods needs common plus essential pointers that can help experts within the appropriate variety of knowledge mining resources, progression plus agreement with predictive types, combined with distribution with predictive patterns within just clinical environment. In yet another solution, they have researched that the health care industry collects huge amounts of medical care information. These studies operate have created one Smart Soul Disease Prediction System (IHDPS) implementing information mining strategies, Choice Timber, Unaware Bayes PJAEE, 17 (9) (2020) 6206 plus Sensation problems Network. Success express that many approaches have exceptional muscle with spotting the particular ambitions on the acknowledged mining goals. Making use of healthcare single profiles just like grow older, gender, blood pressure level plus sugar levels perhaps it will predict the particular likelihood of people finding a coronary heart disease. An existing state of the art introduced an information exploration program made to support the particular quick development of data-derived NTCP models. Prestashop exploits the normal healthcare workflow and information encoded by using a normal ontology. Mcdougal stated which the system referred to is a piece of helpful information on the advance with irradiation oncology information exploration type's especially plus local-level LHS components with the general. Existing research offers consist of collaborative files mining procedure to offer multi level evaluation coming from health test out data. The aim would be to examine success by simply collaboratively implementing various files mining techniques such as classification, clustering, along with connection procedure mining. General, the method seeks from getting information coming from health test out files to be able to increase model of health checks by simply creating peace of mind inside thefinal results implementing multi-level evaluation Reported in one of the solutions that offers your story technique so that you can code health data files by simply collectively employing local mining along with worldwide studying approaches which are tightly hooked up along with mutually reinforced. Neighborhood mining efforts so that you can procedure the individual medical record by simply independently receiving the health strategies from the medical record itself and then maps these phones' authenticated terminologies. Existing research offers consist of the latest feature choice procedure which relies on a backward elimination treatment a lot like which put in place in assist vector equipment recursive feature elimination (SVM-RFE). As opposed to an SVM-RFE procedure, at each step, a consist of approach computes a feature ranking score coming from a mathematical analysis connected with bodyweight vectors connected with various straight line SVMs skilled with subsamples connected with the very first coaching data. One of the prior arts talked about a pair of methods of speeding connected with the anatomical encoding approach. First, an example may be the usage of a proficient algorithm that eliminates code. Subsequent one particular is a demotic approach to just about parallelizing the system one processor.GP efficiency with health classification troubles can be as opposed to coming from a benchmark data source having success received by simply nerve organ networks. Success demonstrates that GP functions equally in classification along with generalization. Another solution, offers used the approaches of data mining (also called Understanding Uncovering in databases) to find interactions inside of a large professional medical database. They will illustrate methods linked to mining your professional medical data source including file warehousing, file query& cleanup along file analysis. In order to improve the accuracy of the machine learning algorithm there is a need to develop particle swarm optimization assisted support vector machine based diagnostic system for lung cancer prediction at the early stage.

SUMMARY OF THE INVENTION

The present disclosure seeks to provide a method for detection of a disease. Developing particle swarm optimization assisted support vector machine based diagnostic system for lung cancer prediction at the early stage. In this disclosure, the integration of Particle swarm optimization (PSO) with a support vector machine (SVM) is proposed to detect lung cancer at a very early stage. The proposed technique is verified by using the various standard lung cancer classification data sets. The comparison is drawn among the proposed and the existing technique based upon the various standard quality metrics of the data mining. Experimental results indicate that the proposed algorithm is more efficient than existing techniques. In the future, the suggested method would help recommend diagnosing cancer in various human organs.

Cancer is one of the countries' deadliest illnesses and it will cure if diagnosed early. Lung cancer is the main cause of death in humans, as the signs of lung cancer occur in advanced stages, so it is difficult to diagnose and contributes to high mortality among other forms of cancer. Therefore, early prediction of lung cancer is mandatory for the diagnostic process and offers better odds of effective therapy. Researchers focus on healthcare to diagnose and avoid lung cancer early. Medical data has achieved its full capacity by offering large data sets to researchers. . Machine learning is a division of artificial intelligence that utilizes a range of mathematical, probabilistic, and optimization techniques that allow computers to "learn" from past examples and identify correlations that are difficult to distinguish from big, noisy, or complex data sets. Machine Learning is commonly utilized in the detection and prognosis of lung cancer. In this paper, Particle Swarm Optimization assisted Support Vector Machine based Diagnostic System for Lung Cancer prediction at an early stage is proposed. The primary objective of this paper is to evaluate the effect of the PSO and SVM for mining the lung cancer PJAEE, 17 (9) (2020) 6203 dataset. The aim of this paper is to improve the accuracy of the machine learning algorithm. The proposed technique was also verified by using the various standard lung cancer classification data sets. The comparison is drawn among the proposed and the existing technique based upon the various standard qualities of service parameters. Experimental results indicate that the proposed algorithm is more efficient than existing techniques.

In an embodiment, the method 100 comprises of the following steps: At step 102, obtaining a plurality of information of the disease, wherein pre-processing the plurality of information to enhance the quality of information using pre-processing module; At step 104, extracting a plurality of features from the processed information using a feature extraction module, wherein a plurality of attributes related to the disease is obtained during extraction; optimizing the extracted attributes using an optimization module for analyzing the disease, wherein the optimization of the attributes comprises steps of: At step 106, creating a population of the extracted attributes acquired from the feature extraction module using a K means technique; At step 108, evaluating a position of each of the extracted attribute in the population using an objective function, wherein a minimum root mean square error is used as the objective function; and At step 110, determining a best position of each of the extracted attribute and placing each of the extracted attribute at each of the best position, wherein a best attribute is selected for each best position in order to optimize the extracted attributes for disease detection.

To further clarify advantages and features of the present disclosure, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.

BRIEF DESCRIPTION OF FIGURES

These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:

Figure 1 illustrates a method for detection of a disease in accordance with an embodiment of the present disclosure.

Figure 2 illustrates the proposed algorithm in accordance with an embodiment of the present disclosure.

Figure 3 illustrates (a) the accuracy and (b)the accuracy analysis in accordance with an embodiment of the present disclosure.

Figure 4 illustrates (a) mean square error and (b) mean square error analysis in accordance with an embodiment of the present disclosure.

Figure 5 illustrates (a) Sensitivity and (b) Sensitivity analysis in accordance with an embodiment of the present disclosure.

Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have been necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present disclosure. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein.

DETAILED DESCRIPTION

For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.

It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be restrictive thereof.

Reference throughout this specification to "an aspect", "another aspect" or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.

The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises...a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional sub-systems or additional elements or additional structures or additional components.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.

Embodiments of the present disclosure will be described below in detail with reference to the accompanying drawings.

Figure 1 illustrates a method for detection of a disease in accordance with an embodiment of the present disclosure. The method 100 comprises of the following steps: At step 102, obtaining a plurality of information of the disease, wherein pre-processing the plurality of information to enhance the quality of information using pre-processing module; At step 104, extracting a plurality of features from the processed information using a feature extraction module, wherein a plurality of attributes related to the disease is obtained during extraction; optimizing the extracted attributes using an optimization module for analyzing the disease, wherein the optimization of the attributes comprises steps of: At step 106, creating a population of the extracted attributes acquired from the feature extraction module using a K means technique; At step 108, evaluating a position of each of the extracted attribute in the population using an objective function, wherein a minimum root mean square error is used as the objective function; and At step 110, determining a best position of each of the extracted attribute and placing each of the extracted attribute at each of the best position, wherein a best attribute is selected for each best position in order to optimize the extracted attributes for disease detection. Figure 2 illustrates the proposed algorithm in accordance with an embodiment of the present disclosure. 1. Algorithm parameters:

A: Population of agents pi: Position of agent ai in the solution space f: Objective function vi: Velocity of agents ai V(ai): Neighborhood of agent ai (fixed). The neighborhood concept in PSO is not the same as the one used in another meta heuristics search, since in PSO each particle's neighborhood never changes (is fixed).

2. Required attributes with values:

Number of particles usually between 10 and 50 C1 is the importance of personal best value C2 is the importance of neighborhood best value Usually C1 + C2 = 4 (empirically chosen value) If velocity is too low --* algorithm too slow If velocity is too high --* algorithm too unstable

3. Particle update rule:

p = p +vwith v= v+ cl*rand*(pBest- p) + c2*rand*(gBest-p) Where: particle's position, v: path direction, cl: the weight of local information, c2: the weight of global information, pBest: the best position of the particle, gBest: the best position of the swarm, and rand: random variable

4. PSO Algorithm:

[x*] = PSO() P = ParticleInitializationO; Fori = 1 toitmax Foreach particlepinPdo

fp = f(p); Iffpisbetterthanf(pBest) pBest = p; end endgBest = bestpinP; Foreach particlepinPdo v= v+ cl*rand*(pBest- p) + c2 *rand* (gBest - p); p = p + v; endend

5. PSD based K-means algorithm for training:

1. Create a 'population' of attributes (particles) from developed clusters of data set X using the K-means algorithm. 2. Evaluate each particle's position according to the objective function i.e. minimum root mean squared error (RMSE). 3. If a particle's current position is better than its previous best position, update it. 4. Determine the best particle (according to the particle's previous best positions). 5. Return optimized trained model.

Figure 3 illustrates(a) the accuracy and (b)the accuracy analysis in accordance with an embodiment of the present disclosure. For research and implementation, the proposed technique is appraised using the WEKA tool. The evaluation of the proposed method is done on the origin of the following parameters as Accuracy, Mean square error, and Sensitivity. The subsequent data demonstrates the comparison regarding response to diverse parameters. The result demonstrates the proposed solution provides improvement over active approaches. After the results, we compared the proposed solution against the current procedures. Accuracy is one metric for assessing model classification. Classification quality is what we generally say by utilizing accuracy. It is the ratio of the number of accurate predictions to total input samples. It is calculated as TP + TN Accuracy = TP +TN + FP + FN

Figure 4 illustrates (a) mean square error and (b) mean square error analysis in accordance with an embodiment of the present disclosure. For research and implementation, the proposed technique is appraised using the WEKA tool. The evaluation of the proposed method is done on the origin of the following parameters as Accuracy, Mean square error, and Sensitivity. The subsequent data demonstrates the comparison regarding response to diverse parameters. The result demonstrates the proposed solution provides improvement over active approaches. After the results, we compared the proposed solution against the current procedures. The mean squared error says the closeness of a regression line across points. It achieves this by taking and squaring distances from points on regression rows. The squaring must eradicate all derogatory indications. It also offers greater discrepancies in weight. The mean squared error is considered to find the average collection of errors.

2 MSE = m n(Xij - Yij) i=1 j=1

Where, Aij and Bij are the image pixel value of the reference image.

Figure 5 illustrates (a) Sensitivity and (b) Sensitivity analysis in accordance with an embodiment of the present disclosure. For research and implementation, the proposed technique is appraised using the WEKA tool. The evaluation of the proposed method is done on the origin of the following parameters as Accuracy, Mean square error, and Sensitivity. The subsequent data demonstrates the comparison regarding response to diverse parameters. The result demonstrates the proposed solution provides improvement over active approaches. After the results, we compared the proposed solution against the current procedures.

Sensitivity is a calculation of the proportion of real positive cases expected positive. This means that there would be another proportion of positive events, which will be wrongly predicted as negative. This may also be represented as a false negative rate. The addition of sensitivity and the false-negative rate is 1. It lies between 0-1. Value of sensitivity near 1 signifies efficient results.

TP Sensitivity =TP+FN

The drawings and the forgoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts necessarily need to be performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples. Numerous variations, whether explicitly given in the specification or not, such as differences in structure, dimension, and use of material, are possible. The scope of embodiments is at least as broad as given by the following claims.

Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any component(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or component of any or all the claims.

Claims

WE CLAIM

1. A method for detection of a disease, the method comprises of:

Obtaining a plurality of information of the disease, wherein pre-processing the plurality of information to enhance the quality of information using pre-processing module;

Extracting a plurality of features from the processed information using a feature extraction module, wherein a plurality of attributes related to the disease is obtained during extraction;

Optimizing the extracted attributes using an optimization module for analyzing the disease, wherein the optimization of the attributes comprises steps of:

Creating a population of the extracted attributes acquired from the feature extraction module using a K-means technique; Evaluating a position of each of the extracted attribute in the population using an objective function, wherein a minimum root mean square error is used as the objective function; and Determining a best position of each of the extracted attribute and placing each of the extracted attribute at each of the best position, wherein a best attribute is selected for each best position in order to optimize the extracted attributes for disease detection.

2. The method as claimed in claim 1, wherein a particle swam optimization technique is used for optimization of the attributes to detect the disease, wherein is a Fuzzy C-Mean clustering technique is used in particle swam optimization to label the attributes in the population, wherein Fuzzy c-means clustering technique segments background and fore ground from an image frame.

3. The method as claimed in claim 2, wherein a plurality of group is generated to track a plurality of record of the fore-ground information obtained from the image.

4. The method as claimed in claim 2, wherein a centroid is selected randomly during determination of best position for enhancing a quality of pixel in the foreground information.

5. The method as claimed in claim 1, wherein a database is used for storing a plurality of images obtained from an input module, wherein the input module comprises of an image capturing module.

6. The method as claimed in claim 1, wherein an image segmentation module is connected to the pre-processing module for dividing the pre-processed information into a plurality of the image frame, wherein each image frame is analysed for segmenting the fore-ground and the back-ground.

7. The method as claimed in claim 1, wherein the number of attributes is taken between 10 and 50, wherein the required attribute includes: personal best value, neighbourhood best value, empirically chosen value obtained by the summation of personal best value and neighbourhood best value and velocity.

8. The method as claimed in claim 1, wherein a particle update rule is utilized for updating the best attribute and the best position to position the best attribute at the best position.

9. The method as claimed in claim 1, wherein a support vector machine classification module is used for classification of disease in order to determine the disease at the early stage.