ARTIFICIAL NEURAL NETWORK BASED BRAIN DISORDER DIAGNOSTIC SYSTEM

• Main Authors: Chandiran. I, Sarath, Raman, S.G, Bora, Ashim, Soubache, I. D, Dixit, Chandra Kumar, Sharma, Mukta, Bhattacharya, Sumanta, Sneha, Yerram, Bindu, P, Swamy, B Venkata, Mangal, Adarsh, Padmanayaki, S
• Format: Patent
• Language: English
• Published: 09.09.2021
• Subjects: DIAGNOSIS; HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATIONTECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING ORPROCESSING OF MEDICAL OR HEALTHCARE DATA; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

ARTIFICIAL NEURAL NETWORK BASED BRAIN DISORDER DIAGNOSTIC SYSTEM This invention relates to a method for empirical risk assessment of brain disorder using a neural network diagnostic system. The said method comprises the steps of magneto-encephalography (MEG) measurement of resting-state MEG signals using a MEG scanner, data pre-processing of the MEG signals measured, developing network architecture, made of a convolution neural network, model training and testing, by evaluating the performance of the developed network architecture based on the sample data and the open-access databases, nested cross validation, for classifying 10 patients with Alzheimer's disease, Epilepsy, and Parkinson's disease, decoding the disease labels, by using relative power of the MEG signals using a support vector machines (SVM) model. MEG Signals measurement. Data Pre-processing of the measured MEG Signals. Developing network architecture model. [ Model training and testing. [ Nested cross-validation of the trained and tested model, Decoding the disease labels. Figure 1. Flow Diagram of Artificial Neural Network Based Brain Disorder Diagnostic System