Therapixel Case Study: AI Helps to Avoid False Positive Recall

Ferrum Health partners with best-in-class AI providers offering solutions designed to reduce radiology workload and improve patient care.

Therapixel Case Study: AI Helps to Avoid False Positive Recall

Introduction

A clinical study, reviewing a benign diagnosis, compared the performance of radiologists reading with and without MammoScreen® 3D. 

Clinical Study

Clustered microcalcifications in the upper outer quadrant at the junction of the middle and posterior third of the breast were identified on a baseline mammogram by the original reader. A follow-up exam recommended a stereotactic biopsy of the microcalcifications. Results diagnosed the area of microcalcifications and surrounding tissue as benign.

MammoScreen correctly flagged this cluster on the mammogram with a score of 4 (low suspicion). During a reader study, 5 radiologists did not recall this case unassisted against 12 utilizing MammoScreen. Moreover, utilizing MammoScreen reduced the reading time on average by 29%.