PeriCALM Patterns 3.0

PeriCALM Patterns 3.0 is a software device that analyzes electronic fetal heart rate and uterine contraction recordings to automatically detect and label fetal heart rate patterns (accelerations, decelerations, baseline) and contractions. It is intended as an adjunct to clinical decision-making during antepartum or intrapartum obstetrical monitoring at ≥32 weeks gestation, providing secondary information to support but not replace direct clinician evaluation.

The subject device uses Long and Short-Term Memory (LSTM) neural networks for pattern recognition, whereas the predicate uses signal processing and neural networks. Key differences include: neural network interference for combined accel-decel-baseline plus non-interpretable segment detection (new feature); analysis time intervals of 15 or 30 minutes versus 20 minutes; and handling of missing tracing segments (new capability). Contraction detection uses the same processes as the predicate.

Not stated in this summary. The document references FDA 2023 guidance documents on device software and cybersecurity but does not cite specific ISO, IEC, or ASTM consensus standards.

The device is substantially equivalent because: (1) intended use is identical—both serve as adjuncts to clinical decision-making during obstetrical monitoring; (2) the lower gestational age threshold (≥32 weeks vs. ≥36 weeks) does not raise different safety or effectiveness questions; (3) technological differences (neural network interference, modified time intervals, non-interpretable segment handling) do not raise different safety or effectiveness questions; and (4) clinical performance testing demonstrated non-inferiority to both ground truth expert consensus and clinician readers across both preterm and term gestational age groups on all 16 co-primary endpoints.