| Issue |
ITM Web Conf.
Volume 80, 2025
2025 2nd International Conference on Advanced Computer Applications and Artificial Intelligence (ACAAI 2025)
|
|
|---|---|---|
| Article Number | 01037 | |
| Number of page(s) | 7 | |
| Section | Machine Learning & Deep Learning Algorithms | |
| DOI | https://doi.org/10.1051/itmconf/20258001037 | |
| Published online | 16 December 2025 | |
Hierarchical Frameworks for Embodied Medical AI
School of Artificial Intelligence, Shanghai Jiaotong University, Shanghai, 200240, China
* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.
Conventional disembodied models took up most of the medical artificial intelligence (AI) while serving a very limited use. Until recently, embodying artificial intelligence in physical robotics has expanded its capabilities and started to show promising clinical translation. This primarily includes automatic surgery platforms, diagnostic assistants, and rehabilitation exoskeletons that employ robust hierarchical control to safely deliver an integration of perception, decision, and action. This study synthesises methods for embodied medical artificial intelligence and conducts a systematic and quantitative analysis of 2025 publications across the aforementioned four domains. The hierarchical designs prove effective as data indicate an improved accuracy, which includes a ~10% higher macro-F1 or sub-millimetre surgical precision in specific tasks. This marks a milestone toward a more practical level of reliability required by clinical translation. These findings confirm that the layered decomposition and hierarchical control are one of the keys to further development of clinical- ready medical artificial intelligence.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.