Genius AI Detection 2.0
K221449Hologic, Inc. · cleared 2022-10-06 · product code QDQ · Radiology
Premarket evidence — what FDA accepted
source quote (p.4)
“Genius AI Detection is a software device intended to identify potential abnormalities in breast tomosynthesis images. Genius AI Detection 2.0 is a software-only device.”
source quote (p.4)
“Genius AI Detection analyzes each standard mammographic view in a digital breast tomosynthesis examination using deep learning networks. The first change is an enlarged cancer data set used to train the deep learning AI models. The second change in Genius AI Detection 2.0 is the adoption of a more sophisticated Convolutional Neural Network (CNN) model for the key processing step of calcification cluster classification, replacing the previously used, simpler conventional Artificial Neural Network (ANN) model.”
source quote (p.8)
“Algorithm enhancements as well as the expanded training data set were used during development of the 2.0 version. The first change is an enlarged cancer data set used to train the deep learning AI models. The cancer database for training and evaluation is enlarged by two-fold. Subsequently, all deep learning AI models have been retrained on the enlarged dataset.”
source quote (p.5)
“"Content of Premarket Submissions for Management of Cybersecurity in Medical Devices” issued on October 2, 2014.”
Validation studies (2)
Bench
sample size not stated
standards: IEC 62304: 2015 – Medical device software – Software Life Cycle Processes (#13-79), ISO 14971: 2012 – Medical devices – Application of Risk Management to Medical Devices
Retrospective clinical
n=764 cases
endpoints: detection sensitivity; specificity; rate of false positive marks per view; fROC analysis
Reported performance (1 observation)
source quote (p.9)
“The specificity measured at the operating point of Genius AI Detection 2.0 demonstrated significant increase of 12% as compared to the original Genius AI Detection predicate device (McNemar's p< 0.001), while maintaining the sensitivity.”
Each value carries its own analysis unit and task — never compare or pool across devices. Source: 510(k) summary PDF.
Predicate network
Postmarket — what happened after clearance
- re_clearance
The FDA AI/ML device list shows a newer 510(k) K243341 (decision 2025-07-31) from Hologic, Inc. for a matching device line ("Genius AI Detection 2.0") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K243341
- re_clearance
The FDA AI/ML device list shows a newer 510(k) K230096 (decision 2023-05-23) from Hologic, Inc. for a matching device line ("Genius AI Detection 2.0 with CC-MLO Correlation") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K230096
Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device). Signals are descriptive observables with sources — never a judgment that the device is unsafe or drifting. Snapshot 2026-07-08.
Reimbursement — how devices like this got paid
Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).