What happened
Hangzhou Yitu Healthcare Technology built Yitu_AICare_Bone_Age, software that locates ossification centers on hand and wrist X-rays with deep learning and calculates bone age by the TW3 method. China's Center for Medical Device Evaluation (CMDE) reviewed the company's application for Class III registration as a priority device. The trial was the maker's. It was prospective and enrolled 1,000 subjects at four clinical institutions. An expert panel's TW3-RUS and TW3-Carpal bone ages served as the gold standard. The report says that the software's mean absolute error against the panel reached the preset target. It gives no figures. CMDE judged that the product “能够有效提高医师阅片效率” (can effectively improve doctors' reading efficiency) when its clinical performance meets clinical needs. On 15 March 2021, it found that the benefits outweigh the risks and recommended registration. The trial did not time the reading. CMDE warns that accuracy may be lower for people aged 0.5 to 6 and 13 to 18, who were few in the trial. It adds that accuracy may be lower for ethnic minorities, because the algorithm work was mainly based on Han Chinese people.
Sources
-
Regulatory document, by authors with no tie to the system.
在临床性能满足临床需求的基础上,能够有效提高医师阅片效率
Details
AI system
- Reported configuration
-
Release 2 (full version 2.2.0)
- Workflow
-
The software located ossification centers on hand and wrist X-rays with deep learning. It calculated TW3-RUS and TW3-Carpal bone ages. An expert panel's TW3 bone ages served as the gold standard.
Follow-up
- Conditions for replication
-
Patients are aged 0.5 to 18 and do not have cone-shaped or deformed epiphyses. Trained doctors use the result and do not base a clinical decision on the software alone.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Hangzhou Yitu Healthcare Technology built Yitu_AICare_Bone_Age to help trained doctors assess bone age by the TW3 method from hand and wrist X-rays of people aged 0.5 to 18.
30 more questions not yet checked
model, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). Yitu_AICare_Bone_Age met its bone-age error target in China's review (AIBD-1077). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1077.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)