Koios DS for Breast
K190442Koios Medical, Inc. · cleared 2019-07-03 · product code POK · Radiology
Premarket evidence — what FDA accepted
source quote (p.3)
“Koios Decision Support (DS) for Breast is a software application designed to assist trained interpreting physicians in analyzing the breast ultrasound images of patients with soft tissue breast lesions who are being referred for further diagnostic ultrasound examination. Koios DS for Breast is a machine learning-based decision support system, indicated as an adjunct to diagnostic ultrasound for breast cancer.”
source quote (p.3)
“Koios DS for Breast is a machine learning-based decision support system, indicated as an adjunct to diagnostic ultrasound for breast cancer. The engine uses computer vision and machine learning techniques embedded within a software capable of reading, interpreting, analyzing, and generating findings from ultrasound data. The underlying engine draws upon knowledge learned from a large database of known cases, tying image features to their eventual diagnosis, to form a predictive model.”
Validation studies (3)
Retrospective clinical
n=900 patients
endpoints: impact on Interpreting Physician (Reader) performance as defined by the area under the Receiver Operating Characteristic (ROC) Curve (AUC); inter-operator variability (Kendall Tau-B correlation coefficient); intra-reader variability (class switching rate)
Bench
n=900 cases
endpoints: degree of concordance with trained interpreting physicians; System performance on the 900 cases reported an AUC of 88.2%
Bench
n=1,300 cases
endpoints: degree of concordance with trained interpreting physicians; overall accuracy that fell within the 95% confidence interval of the radiologists' performance; categorical agreement between each pair of readers was compared to agreement between each reader and the system (Cohen's Kappa coefficient)
Reported performance (7 observations)
source quote (p.15)
“System performance on the 900 cases reported an AUC of 88.2%.”
source quote (p.14)
“This was found to be 0.0370 (0.030, 0.044) at a = .05, satisfying the success criteria for the primary endpoint.”
source quote (p.12)
“the first and second reads is considerably higher in the Koios DS for Breast study set than in the QuantX study with an AUC shift of .7090 to .7577 for QuantX and a shift from .836 to .873 for Koios DS for Breast.”
source quote (p.16)
“Shape 0.769 [0.711, 0.826]”
source quote (p.16)
“Orientation 0.728 [0.655, 0.801]”
source quote (p.16)
“Shape 0.738 [0.679, 0.797]”
source quote (p.16)
“Orientation 0.744 [0.675, 0.813]”
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) K242130 (decision 2024-11-15) from Koios Medical, Inc. for a matching device line ("Koios DS") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K242130
- re_clearance
The FDA AI/ML device list shows a newer 510(k) K212616 (decision 2021-12-16) from Koios Medical, Inc. for a matching device line ("Koios DS") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K212616
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).