Rayvolve PTX-PE

Rayvolve PTX-PE is a radiological computer-assisted triage and notification software that analyzes chest X-ray images (PA or AP views) of patients 18 years or older to detect pre-specified suspected critical findings: pleural effusion and pneumothorax. It uses artificial intelligence algorithms to automatically analyze images and provide study-level output to DICOM node servers (such as PACS) for worklist prioritization. As a passive notification tool, it does not send proactive alerts directly to medical specialists and is not intended for stand-alone clinical decision-making.

Rayvolve PTX-PE is a software-only device that accepts DICOM-format chest radiographs as input, employs a vendor-agnostic AI algorithm to detect pleural effusion and pneumothorax, and outputs results to PACS/workstation interfaces for worklist prioritization. It runs on cloud or on-premise platforms and integrates with existing DICOM node servers. The device filters radiographs by organ detection and provides passive notifications without markup of original images, with a performance time of approximately 19.5 seconds per analysis.

Not stated in this summary.

Rayvolve PTX-PE is substantially equivalent to the predicate Lunit INSIGHT CXR Triage (K211733) because both devices employ identical regulatory classification (Class II, 21 CFR 892.2080, product code QFM), identical intended purpose (computer-assisted triage and notification software for detecting pleural effusion and pneumothorax), identical technological principle (AI algorithms detecting critical findings in DICOM chest X-rays with passive notification output to PACS/workstation), and equivalent performance metrics. Minor differences in indication precision (specifying PA/AP views and age 18+), intended user nomenclature (healthcare professionals vs. medical specialists), and output platforms (DICOM viewers vs. PACS specifically) do not raise new safety or effectiveness concerns. Performance testing shows Rayvolve PTX-PE exceeds or meets the predicate's sensitivity, specificity, and AUC thresholds (all ROC AUC >0.95) across relevant subpopulations (age, gender, ethnicity, patient position).