ProFound Detection (V4.0)

iCAD, Inc.'s ProFound Detection V4.0 is a Class II radiological computer-assisted detection and diagnosis software designed to aid interpreting physicians in identifying breast abnormalities within digital breast tomosynthesis (DBT) exams. The device leverages AI to provide "Certainty of Finding" and "Case Scores," incorporating prior exam data and demonstrating performance improvements over its predecessor, ultimately allowing for more efficient and accurate diagnostic support.

iCAD, Inc. has obtained a Class II classification and marketing permission for its ProFound Detection (V4.0) software, effective November 8, 2024. This radiological computer-assisted detection and diagnosis device was deemed substantially equivalent to existing legally marketed predicate devices, confirming its compliance with general controls provisions including registration, listing, and good manufacturing practices. The ProFound Detection V4.0 software is engineered to assist interpreting physicians during digital breast tomosynthesis (DBT) exams by identifying soft tissue densities and calcifications within 3D slices. It provides diagnostic aids such as "Certainty of Finding" and "Case Scores," integrates the processing of prior exams, and features a Predetermined Change Control Plan for future system extensions. Non-clinical assessments validated its performance as improved or non-inferior to its predecessor, ProFound AI V3.0, substantiating its equivalence.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What specific clinical benefits, such as reduction in recall rates or improvement in cancer detection, does ProFound Detection V4.0 offer over unaided reading or previous versions?; How were the 'Certainty of Finding' and 'Case Scores' validated clinically, and what specific thresholds or interpretations are recommended for their use in diagnosis?; What are the known limitations of ProFound Detection V4.0 concerning specific breast densities, types of lesions, or patient populations?; Can the Predetermined Change Control Plan (PCCP) facilitate integration with AI models from other vendors or future advancements in imaging technology beyond iCAD's ecosystem?; What were the precise non-clinical performance assessment metrics that demonstrated improved or non-inferior performance compared to ProFound AI V3.0, and how significant were these differences?