Automated detection of mild cognitive impairment and dementia from voice recordings: A natural language processing approach

Abstract

Introduction: Automated computational assessment of neuropsychological tests would enable widespread, cost-effective screening for dementia. Methods: A novel natural language processing approach is developed and validated to identify different stages of dementia based on automated transcription of digital voice recordings of subjects' neuropsychological tests conducted by the Framingham Heart Study (n = 1084). Transcribed sentences from the test were encoded into quantitative data and several models were trained and tested using these data and the participants' demographic characteristics. Results: Average area under the curve (AUC) on the held-out test data reached 92.6%, 88.0%, and 74.4% for differentiating Normal cognition from Dementia, Normal or Mild Cognitive Impairment (MCI) from Dementia, and Normal from MCI, respectively.Discussion: The proposed approach offers a fully automated identification of MCI and dementia based on a recorded neuropsychological test, providing an opportunity to develop a remote screening tool that could be adapted easily to any language.

Large Language Models in Neurology Research and Future Practice

“…Our own group previously showed that frameworks such as BERT and neural network–based sentence encoding can be used to automatically transcribe digital voice recordings and differentiate cognitively impaired persons from those with normal cognition. 22 Agbavor and Liang 23 similarly leveraged GPT-3 to develop a model to predict dementia in persons using their spontaneous speech.…”

Automatic classification of AD pathology in FTD phenotypes using natural speech

“…15 A more recent study used a large dataset and a natural language processing (NLP) approach to differentiate individuals with dementia from HC with 87.1% accuracy and to distinguish individuals with mild cognitive impairment (MCI) from HC with 71.2% accuracy. 16 Other reports based on language and speech features ranged from 70% to 89.6% accuracy. [17][18][19][20][21][22][23][24][25] While previous studies have shown promising results with speech features classifying the amnestic AD presentation versus HC, a clinically meaningful challenge that has not been addressed is distinguishing underlying ADNC from FTLD pathology in individuals presenting with non-amnestic phenotypes.…”

AI‐based assessments of speech and language impairments in dementia

“…It helps them to detect linguistic and acoustic impairments associated with dementia 17 and make accurate prognoses about the subsequent development of dementia, in particular AD 11,17,18 . Furthermore, some studies have suggested that AI‐powered systems developed based on the speech and language of individuals can detect cognitive decline associated with MCI, 19,20 dementia, and AD. Thus, these systems not only can be used as speech and language assessments but also be used to develop various digital platforms such as tablets, smartphones, and external hardware 20,21 to assess language impairments associated with dementia or MCI in older adults who live in rural, isolated areas, or medical deserts with limited access to clinical services and dementia care settings 22 .…”

“…Furthermore, some studies have suggested that AI‐powered systems developed based on the speech and language of individuals can detect cognitive decline associated with MCI, 19,20 dementia, and AD. Thus, these systems not only can be used as speech and language assessments but also be used to develop various digital platforms such as tablets, smartphones, and external hardware 20,21 to assess language impairments associated with dementia or MCI in older adults who live in rural, isolated areas, or medical deserts with limited access to clinical services and dementia care settings 22 . Moreover, clinicians can use them to differentiate patients with AD from MCI and monitor the progression of cognitive decline in older adults and PwD 21,23 …”