Intellectual Property India Publishes Patent Application for 'Multi-Scale Feature Extraction And Classification For Early Detection Of Neurological Disorders' Filed by V. Kavitha; Janani Annur Thiruvengadam; Dhanya G S; Sumesh Joe Cherian; Filix Asa Jyothi; G. Mahalakshmi; G. Deepa; Sherin Paul P; S. Radhika; and M. Nabeela

MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641019391 A) filed by V. Kavitha; Janani Annur Thiruvengadam; Dhanya G S; Sumesh Joe Cherian; Filix Asa Jyothi; G. Mahalakshmi; G. Deepa; Sherin Paul P; S. Radhika; and M. Nabeela, Coimbatore, Tamil Nadu, on Feb. 19, for 'multi-scale feature extraction and classification for early detection of neurological disorders.'

Inventor(s) include V. Kavitha; Janani Annur Thiruvengadam; Dhanya G S; Sumesh Joe Cherian; Filix Asa Jyothi; G. Mahalakshmi; G. Deepa; Sherin Paul P; S. Radhika; and M. Nabeela.

The application for the patent was published on Feb. 27, under issue no. 09/2026.

According to the abstract released by the Intellectual Property India: "The multi-scale feature extraction and classification system presented in the proposed invention is intended to aid in the early diagnosis of neurological conditions such as multiple sclerosis, Parkinson's disease, and Alzheimer's. The technology increases the accuracy of diagnosis and automates the detection process by fusing ML models with sophisticated signal processing techniques. The methodology begins with the acquisition of brain signal data via EEG and MRI scans. This data undergoes pre-processing using a Bilateral Filter, which effectively eliminates noise while retaining crucial details. Subsequently, Edge Detection is employed to extract vital patterns and structural features, improving the signal-to-noise ratio. The system further applies a GLCM to extract texture-based features, enhancing its ability to identify subtle yet significant data patterns. For classification, the invention leverages a Hybrid CNN-LSTM model, combining CNN's spatial feature detection with LSTM's temporal sequence analysis to improve prediction accuracy. The system is designed to continuously adapt and refine its performance by learning from new data. The output delivers comprehensive diagnostic insights, aiding medical professionals in confirming diagnoses and creating effective treatment plans."

Disclaimer: Curated by HT Syndication.