Implementation of a 3‐Dimensional Embodied Conversational Agent to Support Neurologists in Conducting Cognitive Assessment in Alzheimer Patients.

Autor: Sherimon, Vinu, Sherimon, P.C., Nair, Rahul V.
Zdroj: Alzheimer's & Dementia: The Journal of the Alzheimer's Association; Dec2023 Supplement 20, Vol. 19 Issue 20, p1-2, 2p
Abstrakt: Background: Alzheimer's disease is a universal medical and public health challenge. A timely identification of this ailment can enhance both the standard of care and the patient's quality of life. Alzheimer's diagnosis includes a critical component called cognitive assessment. These examinations have been carried out by neurologists, manually and digitally since the development of clinical neurological tests. But it's obvious that integrating AI technologies into such assessments has many advantages. This research paper describes the implementation of a 3‐Dimensional Embodied Conversational Agent to support Neurologists in conducting cognitive assessments in Alzheimer patients. Method: Twenty‐two questions are included in the assessment. The assessment checks different mental abilities of patients such as time orientation, place orientation, ability to recall things, concentration skills, language skills, ability to follow written and oral instructions, visual interpretation skills, etc. A score is given for each correct answer. A normal score must be higher than 24. A score of 24 or less is typically seen as abnormal and indicative of possible cognitive impairment further subject to detailed analysis of a neurologist as the result may occasionally be impacted by certain physical impairments, language or speech barriers, education level, or cultural differences. The 3D agent was created using software such as Blender, iclone7, Adobe Photoshop, JavaScript, and MIT App Inventor. Text‐to‐speech synthesis algorithms were used to convert the questions to speech. The answers given by the user will be recognized using Natural Language Processing (NLP) algorithms. Correspondingly, a score will be awarded. TinyDB is used for local storage and CloudDB for web storage. Result: The 3D agent receives the speech input, analyzes the input, and provides a vocal response. As per the total score, it will suggest suitable recommendations vocally. Also, an assessment report with the AI impression is generated. Conclusion: The implementation of a 3‐D Embodied Conversational Agent to support Neurologists in conducting cognitive assessment is portrayed in this research. Instead of trying to replace clinicians, Artificial Intelligence aims to help them and make their tasks simpler. [ABSTRACT FROM AUTHOR]
Databáze: Supplemental Index