MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641095060 A) filed by Adhiparasakthi Engineering College, Melmaruvathur on August 05, 2026, for Advanced Speech Signal Processing Framework For Early Detection Of Parkinson’s Disease Using Hybrid Ann–mlp Architecture.
Inventors include Dr. C. Dhaya; Mrs. T. M Padmapriya; Muthumadathi P; Dr. R. Srivel; Mrs. Devika M; and Mr. G. Sekar.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: The increasing prevalence of Parkinson’s disease and the diagnostic limitations of conventional motor-symptom-based clinical assessments necessitate the development of intelligent, non-invasive early detection systems. This invention proposes an Advanced Speech Signal Processing Framework for the Early Detection of Parkinson’s Disease using a Hybrid Artificial Neural Network (ANN) – Multi-Layer Perceptron (MLP) Architecture augmented with an Attention-Based Bidirectional Long Short- Term Memory (BiLSTM) network. The system collects raw speech data, applies a comprehensive multi-stage preprocessing pipeline (denoising, normalization, silence removal, resampling, and signal smoothing), extracts a 19-dimensional acoustic feature vector encompassing MFCC coefficients, jitter, shimmer, HNR, RPDE, DFA, and PPE values, and classifies speech signals as either Parkinson’s-affected or healthy using the hybrid deep learning model. The custom attention mechanism focuses the network on the most diagnostically informative t
Disclaimer: Curated by HT Syndication.
