What happened
iCAD of Nashua, New Hampshire, makes PowerLook Tomo Detection Software. It finds soft tissue densities in 3D tomosynthesis images and blends them into the 2D synthetic image that radiologists review. It uses classifiers trained on sample densities. In a retrospective reader study, 20 radiologists read 240 cases from five sites, two in the United States and three abroad. Each radiologist read each case with and without the software, at least four weeks apart. FDA reviewers report that reading time improved 29.2%, an average saving of 19.6 seconds per case. The average AUC rose from 0.841 to 0.850, which met the non-inferiority goal. Average specificity was 0.527 without and 0.509 with the software. FDA approved the software on 24 March 2017. The reviewers note limitations that delayed approval, including missing follow-up at one year for some cases and 50 excluded cases. Reading time and accuracy were analyzed separately, so fast wrong reads carried no penalty.
Sources
-
Regulatory document, by authors with no tie to the system.
U.S. Food and Drug Administration, Center for Devices and Radiological Health. Summary of Safety and Effectiveness Data (SSED), P160009: PowerLook Tomo Detection Software. Approved 2017-03-24.
Reading time improved 29.2% with CAD (95% CI: (21.1%, 36.5%); p < 0.01).
Details
AI system
- Workflow
-
The software enhanced up to five detections per breast volume on the 2D synthetic image. Readers had no prior images or patient history and read on a GE workstation.
Follow-up
- Conditions for replication
-
Exams come from GE Senoclaire tomosynthesis systems. Radiologists review the whole image set, because the software does not detect calcifications.
Entry details
- Sector
-
Healthcare.
- Purpose
-
ICAD built PowerLook Tomo Detection Software to help radiologists find soft tissue densities in tomosynthesis exams of the breast and review the exams more quickly.
Not yet checked: model, release, 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). PowerLook Tomo cut radiologists' reading time on 3D mammograms (AIBD-0792). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0792.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)