A Comprehensive AI-Based Detection and Differentiation Model for Neurological Disorders Using PSP Net and Fuzzy Logic-Enhanced Hilbert-Huang Transform

Abstract

Background information: Accurate early detection is necessary for effective intervention in neurological disorders, however conventional methods do not have enough sensitivity. This research presents a new AI model that utilizes the PSP Net structure for detailed feature extraction, the Hilbert-Huang Transform (HHT) for analysing complex signals, and fuzzy logic for managing uncertainties in the data. The model combines these technologies to improve diagnostic accuracy and distinguish between neurological disorders, offering a user-friendly interface designed for use in clinical settings. Methods: This system combines PSP Net for extracting image features, the Hilbert-Huang Transform (HHT) for analysing non-linear brain signals, and fuzzy logic for adjusting classification. PSP Net recognizes spatial characteristics, HHT breaks down signals into intrinsic mode functions for in-depth analysis, and fuzzy logic boosts decision-making by managing data uncertainties, leading to better classification accuracy and disorder differentiation. Objectives: The main goals are to enhance diagnostic precision, facilitate prompt identification, and distinguish between neurological conditions. By utilizing PSP Net's spatial analysis, HHT's signal processing, and fuzzy logic, the system aims to offer a strong, easy-to-use system for healthcare professionals to identify neurological disorders with great precision, aiding in improved clinical decision-making. Results: The suggested model outperforms traditional methods with 95% accuracy, 92% precision, 94% recall, 93% F1 score, and 95% specificity on various metrics. The experiment shows that combining PSP Net, HHT, and fuzzy logic results in the highest performance, emphasizing the importance of each component in enhancing detection abilities. Conclusion: This AI-powered system improves the precision of diagnoses and distinguishes between different neurological disorders. The combination of PSP Net, HHT, and fuzzy logic offers an effective tool for early and accurate detection of disorders, overcoming the drawbacks of traditional methods. This model is a major advancement in using AI for diagnosing neurological conditions.