In the domain of medical diagnostics, artificial intelligence (AI) has revolutionized the ability to detect diseases non-invasively, efficiently, and with remarkable accuracy. Recent developments in AI offer significant potential in diagnosing neuromuscular disorders like Amyotrophic Lateral Sclerosis (ALS), Parkinson's disease, and Myasthenia Gravis, as well as respiratory conditions such as Chronic Obstructive Pulmonary Disease (COPD), asthma, and pulmonary fibrosis. These conditions lead to distinct changes in speech and vocal patterns, articulation, and voice quality, primarily due to muscle weakening and neurological impairments. Furthermore, respiratory diseases alter breathing sounds, cough characteristics, and vocal resonance. Leveraging advanced AI techniques for acoustic and voice analysis stands out as a particularly promising method for early detection, continuous monitoring, and effective management of these ailments.

This Research Topic aims to address multiple challenges inherent in this interdisciplinary field. Key problems include developing AI algorithms that can discern pathological acoustic features from normal variability in speech caused by diverse factors like age, gender, language, and accent. The sparsity of large, well-annotated datasets is another significant barrier, limiting the training and validation of robust models. Furthermore, ensure these AI systems operate effectively in uncontrolled surroundings, amidst background noise and with various recording devices, represents another critical issue. Moreover, integrating explainable AI (XAI) into these systems to provide transparency about how decisions are made is crucial for gaining clinician trust and compliance with regulatory requirements.

To gather further insights in this transformative area of health innovation, we welcome articles addressing, but not limited to, the following themes:

• AI algorithms specifying disease identification through vocal biomarkers.
• Signal processing advances in clinical settings.
• Voice analysis for neuromuscular condition detection and monitoring.
• Acoustic diagnostics in respiratory disease analysis.
• Clinical trials, pilot studies, and real-world implementations of AI tools.
• Interpretability and trust in AI decisions through explainable AI.
• Non-invasive, remote monitoring technologies and their integration.
• Ethical considerations in AI applications including privacy, bias, and fairness.
• Multimodal diagnostic approaches and data integration.
• Standardization, benchmarking, and sharing of voice analysis datasets.

This article collection calls for original research, detailed reviews, and insightful case studies that collectively aim to push the boundaries of AI in acoustic analysis for improved patient outcomes in detecting and managing neuromuscular and respiratory conditions.

Topic Coordinator Roshan Sharma is an employee at Google Deepmind, and Topic Coordinator Ankit Parag Shah is an employee at Accenture. All other Topic Editors declare no conflicts of interest.

Are you interested in contributing a manuscript to this Research Topic?

Advances in AI for Acoustic Diagnostics of Neuromuscular and Respiratory Diseases