Abstract Millions suffer from neurodegenerative Parkinson's disease (PD). Effective PD treatment requires early and precise diagnosis. MRI offers brain structural information. Machine learning has improved diagnostic accuracy in medical imaging in recent years. This paper presents a novel method to categorize Parkinson's disease MRI images utilizing the Adaptive Neuro-Fuzzy Inference Systems (ANFIS) classification algorithm with optimization PSO feature Selection and image enhancement. Three main steps are proposed. First, MRI images are enhanced to increase quality and highlight significant features. Preprocessing includes noise removal, contrast improvement, and image sharpening. The next categorization phase uses improved photos. Second, this work presents illness diagnostic machine learning methods with optimization like PSO for feature extraction. Finally, ANFIS classifies MRI images as PD or non-PD. Parkinson's disease (PD) is a complex neurological ailment that needs early diagnosis and treatment. Machine learning can help diagnose PD by examining patient data attributes. This work provides an optimal hybrid model that classifies Parkinson's disease using numerous characteristics and multiclass diagnosis detection techniques. The hybrid model combines machine learning algorithms to boost classification accuracy. Clinical, demographic, and genetic data represent the disease. PD classification uses feature selection to find the most relevant and discriminative features. ANFIS fuzzy rules and parameters are designed for accurate classification. PD MRI scans are used to test the suggested method. Classification performance is measured by accuracy, sensitivity, specificity, and area under the curve. To prove its efficacy, the proposed classification method is compared to others. The findings show that ANFIS classification with image enhancement approaches can classify PD. The proposed MRI-based Parkinson's disease diagnostic method is accurate and sensitive. ANFIS's intelligent decision-making and MRI characteristics increase classification performance.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|