Wednesday 19 March 2025
Researchers have developed a new artificial intelligence (AI) system that can detect cognitive concerns in patients with dementia, using nothing but electronic health records and natural language processing.
The team, led by Hossein Estiri, used large language models to analyze thousands of clinical notes from Massachusetts General Hospital and Brigham and Women’s Hospital. They trained the AI on this data, teaching it to recognize patterns in language that indicate cognitive decline.
The system works by generating prompts based on patient records, which are then reviewed by an expert. This process is repeated several times until the AI achieves high accuracy in identifying cognitive concerns. The results were impressive: the agentic workflow achieved a sensitivity of 0.90 and specificity of 1.00, outperforming an expert-driven benchmark.
But what does this mean for patients? For those with dementia, early detection is crucial. The sooner cognitive decline is identified, the more effective treatment can be. This AI system could potentially revolutionize diagnosis and intervention, allowing doctors to identify concerns earlier than ever before.
The researchers also highlighted the potential of their technology in reducing the burden on healthcare professionals. With an automated system like this, experts can focus on higher-level tasks, rather than sifting through vast amounts of data.
One of the most significant benefits of this AI is its ability to analyze large volumes of data quickly and accurately. This could lead to faster diagnosis and treatment, which is particularly important for patients with dementia who often have limited time for diagnosis and intervention.
The team plans to continue refining their technology, incorporating more data and improving its accuracy. With further development, this AI system could become a vital tool in the fight against dementia.
In addition to its diagnostic potential, the research also sheds light on the importance of using electronic health records (EHRs) in healthcare. The widespread adoption of EHRs has led to a treasure trove of data that can be used to improve patient care. This study demonstrates the value of harnessing this data through AI-powered analysis.
The implications of this technology extend beyond dementia diagnosis, too. It could potentially be applied to other areas where cognitive decline is a concern, such as Alzheimer’s disease or traumatic brain injury.
As researchers continue to refine and develop this AI system, it’s clear that we’re on the cusp of a new era in healthcare diagnostics. With the potential to revolutionize diagnosis and treatment, this technology could have far-reaching consequences for patients and healthcare professionals alike.
Cite this article: “AI System Detects Cognitive Decline in Dementia Patients Using Electronic Health Records”, The Science Archive, 2025.
Artificial Intelligence, Dementia, Cognitive Decline, Electronic Health Records, Natural Language Processing, Diagnosis, Treatment, Alzheimer’S Disease, Traumatic Brain Injury, Healthcare Diagnostics
