AI literacy in dementia:benefits and challenges of clinical applications of AI

Kaoru Sakatani^1）2）

^1）Institute of Gerontology
^2）Department of Human and Engineered Environmental Studies, Graduate School of Frontier Sciences The University of Tokyo

In this review, the focus is on the application of AI in the medical field, particularly in the domain of dementia.　The review examines the current state and potential of AI in this context and elucidates the following points:（1） Deep learning is the most advanced technique in machine learning used in AI, evolving the analytical method known as neural networks and enabling high-precision analysis and predictions.　（2） Learning methods encompass supervised learning, unsupervised learning, and semi-supervised learning.　（3） Deep learning includes feedforward, convolutional, and recurrent neural networks.　（4） Convolutional neural networks are well-suited for image diagnostics, while recurrent neural networks are ideal for predicting time series data.　（5） Challenges exist in the application of deep learning in the medical field, such as data scarcity and bias.
Finally, it is emphasized that the enhancement of AI literacy among medical professionals is pivotal in unlocking the benefits of AI adoption in healthcare.　Simultaneously, the ethical considerations and the significance of privacy associated with AI implementation are underscored.　The application of AI in the realm of dementia continues to evolve, making AI literacy increasingly crucial in the future.

Address correspondence to Dr. Kaoru Sakatani, Department of Human and Engineered Environmental Studies, Graduate School of Frontier Sciences The University of Tokyo （Kashiwa Research Complex 2. Room 206, 5-1-5 Kashiwanoha, Kashiwa-shi, Chiba 277-8561, Japan）