Researchers from Middle Technical University (MTU) in Baghdad and the University of South Australia (UniSA) have recently published a conference paper reviewing advancements in artificial intelligence techniques for detecting Parkinson's disease (PD).
Image Credit: MMD Creative/Shutterstock.com
Researchers from Iraq and Australia say algorithms capable of detecting subtle changes in a person's voice are emerging as a promising new diagnostic tool for Parkinson's disease.
Speech impairments are frequently among the first signs of Parkinson's disease, the fastest-growing neurological disorder globally, affecting over 8.5 million people. However, traditional diagnostic methods tend to be complex and slow, leading to delays in early detection.
MTU Associate Professor Ali Al-Naji, a Medical Instrumentation Engineer and Adjunct at UniSA, asserts that the evidence strongly indicates AI-powered voice analysis could revolutionize early Parkinson's disease (PD) diagnosis and facilitate remote monitoring of the neurodegenerative disorder.
Vocal changes are early indicators of Parkinson’s disease, including small variations in pitch, articulation, and rhythm, due to diminished control over vocal muscles. By analyzing these acoustic features, AI models can detect subtle, disease-related vocal patterns long before visible symptoms appear.
Ali Al-Naji, Associate Professor, Middle Technical University
AI techniques primarily employ machine learning and deep learning algorithms trained on extensive datasets from simple voice recordings of Parkinson's patients and healthy individuals.
These algorithms extract key features, such as pitch, speech distortions, and vowel changes, and then classify the voice recordings with impressive accuracy—up to 99% in one study.
Researchers emphasize that, although there is no cure for Parkinson's, early diagnosis and intervention can enhance quality of life and help slow the progression of symptoms.
As well as detecting Parkinson’s early, AI could also help monitor patients from a distance, reducing the need for in-person visits.
Ali Al-Naji, Associate Professor, Middle Technical University
However, researchers recognize the need for further studies involving larger and more diverse populations.