Genius AI Detection 2.0
K243341Hologic, Inc. · cleared 2025-07-31 · product code QDQ · Radiology
Premarket evidence — what FDA accepted
source quote (p.3)
“Genius Al Detection is a computer-aided detection and diagnosis (CADe/CADx) software device intended to be used with compatible digital breast tomosynthesis (DBT) systems to identify and mark regions of interest including soft tissue densities (masses, architectural distortions and asymmetries) and calcifications in DBT exams from compatible DBT systems and provide confidence scores that offer assessment for Certainty of Findings and a Case Score.”
source quote (p.5)
“Genius Al Detection 2.0 analyzes each standard mammographic view in a digital breast tomosynthesis examination using deep learning networks.”
Validation studies (3)
Bench
sample size not stated
standards: IEC 62304: 2015 – Medical device software – Software Life Cycle Processes, ISO 14971: 2019 – Medical Devices – Application of Risk Management to Medical Devices
Retrospective clinical
n=1,475 patients · 15 site(s)
endpoints: fROC curves; ROC curves; sensitivity; specificity; false marker rate per view; accuracy (AUC) non-inferiority
standards: IEC 62304: 2015 – Medical device software – Software Life Cycle Processes, ISO 14971: 2019 – Medical Devices – Application of Risk Management to Medical Devices
Retrospective clinical
n=480 patients
endpoints: location specific cancer detection sensitivity; specificity
standards: IEC 62304: 2015 – Medical device software – Software Life Cycle Processes, ISO 14971: 2019 – Medical Devices – Application of Risk Management to Medical Devices
Reported performance (4 observations)
source quote (p.10)
“The detection performance of GAID 2.0 measured on a set of 132 cancer patients and 348 negative subjects with implant displaced images demonstrated location specific cancer detection sensitivity of 76% (CI 68%~84%)”
source quote (p.10)
“and specificity of 67% (CI 62%~72%).”
source quote (p.9)
“The CC-MLO Correlation algorithm accurately correlated the Genius Al Detection software 2.0 marks on 90% of the biopsy-proven malignant lesions.”
source quote (p.9)
“In addition, 73% of correlated pairs of marks on negative cases were considered as accurate by expert radiologists.”
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
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).