Contribution of application of deep learning approaches on biomedical data in the diagnosis of neurological disorders: A review on recent findings

Abstract

Recent developments in Artificial intelligence and their applications in Medical health diagnosis have shown promising results especially in the diagnosis of brain disorders, thereby paving the way for better healthcare and early detection of Neural disorders. Several diagnostic tests are routinely performed on humans to evaluate the nature of brain signals and images. These diagnostic tests typically rely on the physiological signals and neuroimaging analysis commonly performed by an expert manually. Large databases of clinical data of recorded brain signals and images are available. Secondary use of these medical records for precision medicine and predictive analysis involves machine learning which has fueled the excitement of medical experts and data scientists equally. For an effective medical diagnosis, intelligent behavior is the prerequisite for learning, and there has been a deep insight in this particular area for small diagnostic problems. Symbolic learning, statistical methods and neural networks in a broader perspective have contributed to the development of machine learning in general and medical diagnosis in particular. Deep learning, a recent trend in machine learning has been able to produce exceptionally accurate results in the diagnosis and treatment of various diseases. This review article covers the exciting contributions of the many researchers involved in this field and highlights the challenges considered critical in this area of research.