Lumify Diagnostic Ultrasound System

The Lumify Diagnostic Ultrasound System is a versatile, mobile imaging platform now incorporating an AI-driven Auto EF Quantification feature for enhanced cardiac assessment. This technology leverages a neural network based on the LVivo EF algorithm to automatically measure left ventricular ejection fraction, end-diastolic volume, and end-systolic volume from apical four-chamber views, with clinical studies confirming its strong correlation to manual methods.

The Philips Lumify Diagnostic Ultrasound System is a mobile, software-controlled medical device designed for acquiring and displaying high-resolution ultrasound data in B (2D), Pulsed Wave Doppler, Color Doppler, Combined (B+Color), and M-modes. It utilizes commercial off-the-shelf (COTS) Android or iOS mobile devices, Philips Ultrasound Lumify software, and specific Philips USB transducers (C5-2 Curved array, L12-4 Linear array, S4-1 Sector array). The system is intended for diagnostic ultrasound imaging and fluid flow analysis across a wide range of applications, including fetal/obstetric, abdominal, pediatric, cardiac, and lung imaging, operated by healthcare professionals in various healthcare environments. A key enhancement to the Lumify system is the Auto EF Quantification feature, an artificial intelligence (AI) based technology available on version 5.0 (Android) specifically for cardiac imaging with the S4-1 transducer. Derived from DiA's LVivo EF algorithm, this feature automates the assessment of left ventricular ejection fraction (EF), end-diastolic volume (EDV), and end-systolic volume (ESV) from apical four-chamber (A4CH) views. The software automatically traces and tracks LV borders, provides an editing functionality for user review and adjustment, and includes a mechanism for rejecting false results. Clinical performance data, based on a study of 80 patients, demonstrated strong correlations between the automated LVivo EF measurements and manual tracing results for EF (r=0.82), EDV (r=0.95), and ESV (r=0.94), with the algorithm successfully processing 95% of the clips.

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. How does the Auto EF algorithm's performance vary across different ultrasound image qualities or in patients with complex cardiac pathologies not extensively covered in the 80-patient study?; What specific criteria trigger the automated rejection of false results by the Auto EF feature, and what is its documented false positive/negative rate for this rejection mechanism?; What was the reported inter-rater and intra-rater reliability for the manual tracings used as the reference standard in the clinical performance study for Ejection Fraction, End Diastolic Volume, and End Systolic Volume?; How does the processing time for Auto EF Quantification compare to manual assessment in a typical clinical workflow, and what is the impact on overall examination time?; Are there plans to expand the Auto EF Quantification feature to other cardiac views beyond the apical four-chamber view or to other ultrasound transducers within the Lumify system?