Caption Guidance

Caption Guidance is a software accessory utilizing artificial intelligence to assist medical professionals in the acquisition of standard adult two-dimensional transthoracic echocardiography images by providing real-time, expert-emulating guidance. Its methodology incorporates features like a quality meter, prescriptive transducer directions, and automated clip capture to enhance image quality and usability, aiming to improve the consistency of cardiac ultrasound examinations.

The Caption Guidance software is an artificial intelligence-driven system designed to assist medical professionals in acquiring cardiac ultrasound images. It functions as an accessory to compatible general-purpose diagnostic ultrasound systems and is indicated for use in two-dimensional transthoracic echocardiography (2D-TTE) for adult patients, covering ten standard views such as Parasternal Long-Axis and Apical 4-Chamber. The software emulates the expertise of sonographers by providing real-time guidance to help users obtain anatomically correct images and achieve diagnostic quality. Key features of Caption Guidance include a Quality Meter for real-time feedback on image quality, Prescriptive Guidance that directs transducer manipulation, an Auto-Capture function to automatically record diagnostic clips, and a Save Best Clip feature for retrospective saving. This updated version incorporates enhancements for increased usability and performance, including additional prescriptive guidance, more user information via the UI, optimized quality meter and auto-capture, and customizable scanning protocols. The device underwent extensive testing, including software verification and validation, algorithm performance assessment for auto-capture and prescriptive guidance, and human factors testing with both inexperienced and experienced users, which indicated optimized usability and no use errors associated with critical tasks.

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 'Quality Meter' algorithm define and quantify 'diagnostic quality' for different echocardiographic views, and what are its sensitivity and specificity?; What were the quantitative performance improvements of the modified Auto-Capture feature compared to the predicate device or manual capture by experienced sonographers?; How does the prescriptive guidance adapt its recommendations for users with varying levels of prior sonography experience, as suggested by the human factors testing cohorts?; What specific 'additional prescriptive guidance' and 'information for users via the UI' were implemented in this version, and how did they contribute to the observed optimization of usability?; Beyond the listed standard views, what is the potential for the Caption Guidance AI to be trained for or assist with more specialized echocardiographic assessments or pediatric patients?