What happened
Wuhan Haobo Technology built HB-AI-T1, software that uses deep learning on breast thermal tomography images to flag possible abnormal changes. Clinicians must read every image again, and the software is not meant to tell benign lesions from malignant ones. The Center for Medical Device Evaluation (CMDE) of China's National Medical Products Administration reviewed the company's application for Class III registration. The trial was the maker's own, a multicenter single-arm study against target values at three clinical institutions. It enrolled 743 women aged 18 to 75 who needed a breast examination. The reference standard was the diagnosis of a panel of senior sonographers. CMDE reports that the sensitivity and specificity of the software's triage reached the preset targets. The report gives neither the targets nor the results. On 21 November 2025, CMDE concluded that “该产品的上市为适用人群带来的受益大于风险” (putting the product on the market brings the intended population benefits that outweigh the risks). It recommended approval. The report does not say that the triage guided any woman's care.
Sources
-
Regulatory document, by authors with no tie to the system.
临床试验结论:软件辅助分诊相对于参考标准分析的灵敏度、特异度达到预期目标
根据申请人提供的申报资料,经综合评价,在目前认知水平上,认为该产品的上市为适用人群带来的受益大于风险。
Details
AI system
- Reported configuration
-
Release 1
- Workflow
-
Users imported breast thermal tomography images from the maker's specified imaging system into the software by USB drive. The software segmented the breast region and used deep learning on several indicators to give a positive or negative triage prompt. A panel of senior sonographers' diagnoses served as the reference standard.
Follow-up
- Conditions for replication
-
Patients are adult women who need a breast examination, without fever and not breastfeeding. Images come from the maker's specified thermal tomography system and probes. Clinicians read every image again, and the software does not replace ultrasound, mammography or MRI.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Wuhan Haobo Technology built HB-AI-T1 to give trained clinicians a triage prompt on whether breast thermal tomography images of adult women show abnormal changes.
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). HB-AI-T1 met its breast thermal triage targets in China's review (AIBD-1158). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1158.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)