Chapter 19 - Artificial intelligence-based early detection of neurological disease using noninvasive method based on speech analysis

Abstract

Several studies have been carried out that predict there will be a global surge in the number of patients with neurological diseases by 2050. Many of these neurological diseases such as Alzheimer's disease, acute spinal cord injury, amyotrophic lateral sclerosis, ataxia, Bell's palsy, brain tumors, cerebral aneurysm, epilepsy, and seizures are often diagnosed late resulting in several complications and irreversible effects. Recently, physiological signals, patient data, artificial intelligence and machine learning techniques, and medical images are being utilized for advanced signal processing and analysis in clinical decisions and diagnosis of neurological diseases. Several speech monitoring and recording devices are aimed at rapidly detecting and analyzing difficulties in speaking to enhance early diagnosis. These techniques are noninvasive approaches used for analysis and interpretation of complex pathways and biomarkers of neurological diseases. Therefore, application of smart algorithms could be utilized to diagnose, detect, and analyze early dysfunctions in a patient's neurological status. Thus, artificial intelligence-based, noninvasive approaches to speech analysis could serve as inexpensive, easy, and quick methods for overcoming the challenge of neurological diseases. This chapter describes the latest noninvasive techniques such as emotion recognition/intelligence, virtual environments, and behavioral analysis in the diagnosis, screening, evaluation, and early detection of common neurological diseases.