A Comprehensive Survey on the Detection, Classification, and Challenges of Neurological Disorders.

Area of Science:

• Neuroscience and Computational Medicine
• Artificial Intelligence in Healthcare

Background:

• Neurological disorders (NDs) are a growing global health concern, affecting diverse populations including pregnant women, infants, and children.
• Advances in neuroimaging (MRI, MEG, PET) and computational tools have opened new avenues for understanding brain complexity and diagnosing NDs.

Purpose of the Study:

• To provide a comprehensive review of machine learning (ML) and deep learning (DL) approaches for the computer-aided diagnosis of neurological disorders.
• To critically compare the performance of existing ML and DL methods using various neuroimaging modalities and datasets.

Main Methods:

• The study outlines a computer-aided diagnosis methodology, including pre-processing and feature extraction techniques.
• It reviews and compares ML and DL algorithms applied to neuroimaging data (images, signals, speech) for ND detection.
• Standard evaluation metrics for result analysis and comparison are presented.

Main Results:

• The review critically assesses the performance of current ML and DL techniques in detecting neurological disorders.
• It highlights the potential of these computational methods in conjunction with neuroimaging for early and accurate diagnosis.
• The study identifies gaps and summarizes limited existing research in specific ND detection criteria.

Conclusions:

• Machine learning and deep learning show significant promise for the early detection and diagnosis of neurological disorders.
• Further research is needed to address open challenges and establish robust detection criteria for a wider range of NDs.
• This review provides a foundational workflow and insights for future research in AI-driven neurological disorder diagnosis.