AI Detects Alzheimer’s with 78.5% Accuracy by Analyzing Speech Patterns

A new artificial intelligence (AI) tool shows promise in the early detection of Alzheimer’s disease, a condition that remains one of the most significant medical challenges of our time. While the exact causes of Alzheimer’s are still unknown, its devastating effects on cognitive function and quality of life are well-documented. Early and accurate diagnosis is crucial for managing the disease and potentially slowing its progression.

AI Aids in Early Alzheimer’s Detection

Researchers at Boston University have developed an algorithm that analyzes speech patterns to predict the development of Alzheimer’s disease with remarkable accuracy. The AI model, trained on voice recordings from individuals with mild cognitive impairment (MCI), can predict which patients will develop Alzheimer’s within six years with 78.5% accuracy. This builds on previous work where a model was trained on over 1,000 voice recordings to detect cognitive decline.

The algorithm was trained by analyzing audio transcriptions from 166 individuals between the ages of 63 and 97, all diagnosed with MCI. Using machine learning, the AI identified specific vocal markers in 90 participants who later progressed to Alzheimer’s disease. This non-invasive approach offers a potentially simpler and more accessible method for early detection.

The method offers several potential advantages:

• It is non-invasive and doesn’t require specialized equipment.
• It is quick and inexpensive.
• It could potentially be used at home.
• There is a possibility of future integration into a smartphone application.

Alzheimer’s disease is the most common cause of dementia and is accompanied by warning signs such as subtle changes, including in the way a person speaks. © AquaArtStudio, iStock

A Window of Opportunity for Early Intervention

While there is currently no cure for Alzheimer’s, early detection offers significant benefits. Ioannis Paschalidis, a computer scientist at Boston University, emphasizes: “If you can predict what is going to happen, you have a greater opportunity and a window of time to intervene with medications, and at least try to maintain the stability of the condition and prevent the transition to more severe forms of dementia.”

• Faster implementation of existing treatments.
• A better understanding of the disease’s progression.
• Increased opportunities for clinical research.
• The potential development of more effective treatments.

Potential for Improvement and Future Implications

The study, published in Alzheimer’s & Dementia, suggests significant possibilities for improvement. The recordings used in the study were of relatively poor quality. With higher-quality audio data, the algorithm’s accuracy could increase further, providing a more nuanced understanding of the early stages of Alzheimer’s.

This advancement could also shed light on the crucial question of why MCI sometimes progresses to Alzheimer’s disease, and sometimes does not. Dr. Paschalidis expressed optimism: “We hope, like everyone else, that more treatments for Alzheimer’s” will become available.