| Unique ID issued by UMIN | UMIN000061842 |
|---|---|
| Receipt number | R000070767 |
| Scientific Title | Simulation-based evaluation of medical assistance using a physical AI-equipped robotic arm |
| Date of disclosure of the study information | 2026/06/10 |
| Last modified on | 2026/06/09 11:41:59 |
Simulation-based evaluation of medical assistance using a physical AI-equipped robotic arm
Robotic Arm Medical Assistance Study
Simulation-based evaluation of medical assistance using a physical AI-equipped robotic arm
Robotic Arm Medical Assistance Study
| Japan |
None
| Anesthesiology |
Others
NO
Applying a robotic arm equipped with physical AI (a technology that integrates AI-based recognition and judgment with physical movement), this study evaluates the feasibility, accuracy, and safety of the following three points in a simulated environment: (1) instrument handover during central venous puncture, (2) autonomous puncture assistance for spinal anesthesia using a lumbar puncture simulator, and (3) autonomous puncture assistance for procedures such as peripheral nerve block, epidural anesthesia, and central venous puncture based on ultrasound image recognition. Through this, the study examines the contribution to reducing the workload of medical staff and decreasing human error.
Efficacy
Verification 1: accuracy of instrument handover (whether the correct item was delivered in the correct order and orientation). Verification 2: puncture success rate. Verification 3: puncture success rate.
Verification 1: number of errors (misdelivery, dropping, etc.), task duration. Verification 2: puncture site accuracy, task duration, safety assessment by physicians. Verification 3: puncture site accuracy, task duration, safety assessment by physicians.
Observational
| 24 | years-old | < |
| 65 | years-old | >= |
Male and Female
Physicians affiliated with Yamagata University Hospital between May 1, 2026 and March 31, 2030 who provided consent to participate.
Those who did not provide consent to participate.
210
| 1st name | Hiroaki |
| Middle name | |
| Last name | Toyama |
Yamagata University Faculty of Medicine
Department of Anesthesiology
990-9585
2-2-2 Iida-nishi, Yamagata City, Yamagata
023-628-5400
hayasakatatsuya1101@gmail.com
| 1st name | Tatsuya |
| Middle name | |
| Last name | Hayasaka |
Yamagata University Faculty of Medicine
Department of Anesthesiology
990-9585
2-2-2 Iida-nishi, Yamagata city, Yamagata
023-628-5400
hayasakatatsuya1101@gmail.com
Yamagata University Faculty of Medicine
Takahashi Industry-Economy Research Foundation
Other
Japan
The Ethical Review Committee of Yamagata University Faculty of Medicine
2-2-2 Iida-nishi, Yamagata City, Yamagata
023-628-5015
ikekenkyu@jm.kj.yamagata-u.ac.jp
NO
| 2026 | Year | 06 | Month | 10 | Day |
Unpublished
Preinitiation
| 2026 | Year | 05 | Month | 01 | Day |
| 2026 | Year | 06 | Month | 10 | Day |
| 2030 | Year | 03 | Month | 31 | Day |
In Japan, especially in rural areas, a serious shortage of medical staff persists, making it difficult to secure personnel for peripheral tasks such as instrument handover and drug preparation during surgery, as well as for high-difficulty puncture procedures such as epidural anesthesia and central venous puncture. This study applies a robotic arm equipped with physical AI, a technology integrating AI-based recognition and judgment with physical movement that is already in practical use in industry, to medicine, and conducts three verifications in a simulated environment. In Verification 1, the robotic arm hands over items (syringe, guidewire, catheter, etc.) to the physician in a simulated central venous puncture setting, and the accuracy of handover, number of errors, and task duration are evaluated. In Verification 2, using a commercial lumbar puncture simulator, the robotic arm performs autonomous puncture based on AI image recognition and motion planning (the physician makes the final decision for each puncture); the puncture success rate, puncture site accuracy, task duration, and safety assessment by physicians are measured. In Verification 3, using a commercial simulator, the robotic arm automatically recognizes nerves and targets from ultrasound images and performs autonomous puncture assistance for peripheral nerve block, epidural anesthesia, central venous puncture, and similar procedures (the physician makes the final decision); the puncture site is not restricted, and the success rate, puncture site accuracy, task duration, and safety assessment by physicians are measured. Participants are physicians affiliated with Yamagata University Hospital who have consented to participate; no patients are directly involved. Statistical analysis uses the chi-square test, Fisher's exact test, Mann-Whitney U test, and t-test, with a significance level of P<0.05.
| 2026 | Year | 06 | Month | 09 | Day |
| 2026 | Year | 06 | Month | 09 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/icdr_e/ctr_view.cgi?recptno=R000070767