What happened
Beijing Deepwise built Dr.Wise AI Lung Cancer, software that applies deep learning to a nodule that a doctor selects on adult chest CT. It flags whether a nodule of 4 mm or more, other than a pure ground-glass nodule, may be malignant. 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 paired multi-reader multi-case study of 400 subjects at three sites, half of them positive. Imaging, thoracic surgery and pathology experts set the reference. Doctors' subject-level area under the ROC curve was 0.8397 with the software and 0.7325 without. Nodule-level sensitivity rose by 7.64% (95% CI 6.13% to 9.15%) and specificity by 12.58%. On 29 September 2024, CMDE concluded that “该产品上市带来的受益大于风险” (putting the product on the market brings benefits that outweigh the risks). It recommended approval. The product is not for lung cancer screening. The report does not say whether the readings guided any patient's care.
Sources
-
Regulatory document, by authors with no tie to the system.
AI+医生联合阅片和医生单独阅片的灵敏度组间差值及 95%CI 为 7.64%(6.13%,9.15%)
经综合评价,认为该产品上市带来的受益大于风险,综合剩余风险可接受。
Details
AI system
- Reported configuration
-
Release 1 (full version 1.1.0.1)
- Workflow
-
A doctor selected each nodule on the CT image. The software combined morphological features with deep learning to flag a suspected malignancy risk for the selected nodule.
Follow-up
- Conditions for replication
-
Nodules are 4 mm or larger on adult chest CT and are not pure ground-glass nodules. The software is not used for lung cancer screening. Trained doctors make the final judgment and do not base a clinical decision on the software alone.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Beijing Deepwise built Dr.Wise AI Lung Cancer to tell trained doctors whether a lung nodule of 4 mm or more on adult chest CT carries a suspected risk of malignancy.
29 more questions not yet checked
model, provider, role, configuration, inputs and outputs, autonomy, human oversight, 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). Dr.Wise AI Lung Cancer helped doctors rate lung nodule risk on CT (AIBD-1063). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1063.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)