Differentiating Alzheimer's disease from mild cognitive impairment: a quick screening tool based on machine learning

Area of Science:

• Neuroscience
• Gerontology
• Medical Diagnostics

Background:

• Alzheimer's disease (AD) is a neurodegenerative disorder causing cognitive decline, behavioral changes, and daily living impairments.
• Early differentiation between AD and mild cognitive impairment (MCI) is crucial for timely intervention.
• Existing diagnostic methods can be time-consuming and require specialized expertise.

Purpose of the Study:

• To develop and validate a rapid screening tool for differentiating AD from MCI.
• To identify key cognitive and functional measures that best distinguish between AD and MCI.
• To create a comprehensive scale assessing activities of daily living (ADL), behavioral and psychological symptoms of dementia (BPSD), and cognitive function (ABC).

Main Methods:

• Utilized machine learning algorithms (XGboost, CART, Bayes, SVM, logistic regression) to build and verify a diagnostic tool.
• Evaluated 458 newly diagnosed AD and MCI patients using eleven assessment batteries.
• Developed the ABC-Scale, incorporating ADL, BPSD, and cognitive assessments, with an estimated completion time of 18 minutes.

Main Results:

• The Alzheimer's Disease Assessment Scale-cognitive (ADAS-cog) word recognition task was the most significant predictor, followed by auditory verbal learning test recall and ADAS-cog orientation.
• The developed ABC-Scale demonstrated improved sensitivity across four machine learning models.
• The ABC-Scale efficiently covers ADL, BPSD, and cognitive domains.

Conclusions:

• The rapid screening ABC-Scale effectively differentiates Alzheimer's disease from mild cognitive impairment.
• The scale integrates assessments of daily living activities, behavioral symptoms, and cognitive function.
• This tool offers a promising approach for efficient and accurate early diagnosis of AD.