BioticsAI

BioticsAI is a software-as-a-service (SaaS) solution that uses artificial intelligence to automatically analyze fetal ultrasound images (DICOM instances) to detect anatomical planes, identify anatomical structures, and verify image quality criteria. It serves as a concurrent reading aid for healthcare professionals during and after fetal ultrasound examinations, typically for second-trimester anatomy scans at 18-39 weeks gestation.

BioticsAI uses AI-based computer vision with three primary components: high-level anatomy classification, per-class top-1 fetal plane classification, and fetal anatomical structure classification. Key differences from predicate (Sonio Detect): BioticsAI integrates via standard DIMSE connection provided by the client, while Sonio Detect ships separate edge software; BioticsAI is compatible with GE Medical ultrasound systems only, while Sonio Detect supports GE, Samsung, Canon, and Philips; both are cloud-based SaaS solutions using AI methodology.

Not stated in this summary.

BioticsAI has nearly identical indications for use as the predicate Sonio Detect (K240406)—both automatically detect views, identify anatomical structures, and verify quality criteria for fetal ultrasound as a concurrent reading aid. Although they differ in platform compatibility and integration approach, these differences do not raise new safety or effectiveness questions. Performance testing on 11,186 images across 296 patients demonstrated sensitivity and specificity rates comparable to safe and effective operation, supporting substantial equivalence.