Prediction of Future Weight by Explainable AI
October 04, 2023
Prediction of Weight up to 3 Years could be Applied to the Field of Health Checkups and Patient Guidance Support
A research team consisting of Associate Professor FUJIHARA Kazuya and Professor SONE Hirohito in Department of Hematology, Endocrinology and Metabolism, Niigata University Faculty of Medicine and NEC Solution Innovators, Ltd. (hereinafter referred to as NEC Solution Innovators; Headquarters: Koto Ward, Tokyo; Representative Director, President and Executive Officer: ISHII Chikara) showed the possibility of predicting the change of body weight over the next three years by their diet and exercise habits. The research team evaluated the results of health checkups using the artificial intelligence (AI) developed by NEC Solution Innovators.
Furthermore, by using heterogeneous mixture learning technology, they are now able to address the “AI black box problem” that has conventionally been an issue with artificial intelligence.
The results were published in the international journal Frontiers in Public Health (IF: 5.2) on June 15, 2023.
Research Results
-
- Using explainable AI makes it possible to demonstrate the prediction of weight changes over a three-year period for individuals accompanied with the improvement/worseing of their diet and exercise habits.
- Weight prediction in this study has similar predictive accuracy to conventional regression analysis and is faster and more efficient.
Publication Details
Journal:?Frontiers in Public Health
Title: Machine learning approach to predict body weight in adults
Authors:?Kazuya Fujihara, Mayuko Yamada Harada, Chika Horikawa, Midori Iwanaga, Hirofumi Tanaka, Hitoshi Nomura, Yasuharu Sui, Kyouhei Tanabe, Takaho Yamada, Satoru Kodama, Kiminori Kato and Hirohito Sone
doi: 10.3389/fpubh.2023.1090146
More 篮球比分直播
-
Newly discovered role of amyloid precursor protein (APP) in nuclear waste disposal
Research results
-
iPatax: a Tablet-based Tool for Quantitative Assessment of Cerebellar Ataxia
Research results
-
Scg2 drives corticospinal circuit reorganization with spinal premotor interneurons and astrocytes for motor recovery after stroke in mice.
Research results
-
A newly discovered kofun (ancient Japanese burial mound) on a forested hillslope in the Kamigiri of Nagaoka City via topographical surveying using an uncrewed aerial vehicle (UAV)
Research results