01
Starting with an access challenge
Cardiac examination relies on both equipment and skilled people. Where specialists are limited, making useful assessment information available from acquired images is an important challenge.
Clinical and engineering teams came together around AI analysis of ultrasound video, beginning with tests of left ventricular function classification.
02
Connecting images and AI through a phone
We studied recording an ultrasound display with a phone and submitting the clip for AI classification, opening a way to evaluate use with existing clinical equipment.
The 2024 study achieved 177 correct results out of 184 clips, or 96.2%, for three categories, while Mildly Reduced EF was correct in 12 of 17 clips. Both findings inform product development.
References and further reading
03
Developing beyond the model
Practical use requires an app, data submission and processing services, reports, and connected system management. Platform development therefore sits alongside model and workflow evaluation.
A parallel task is preparing technical, risk, quality, and usability documentation for medical software. This is readiness work, not certification or authorization.
04
Fitting technology into a care team
Heart-failure care involves multiple professions, resources, and follow-up. An AI assessment is useful only when the team can read, understand, and consider it together.
We treat acquisition and reporting as part of service delivery, with effects on time, workload, and referrals still requiring evaluation.
05
Engaging networks in four regions
Activities in 2025 covered Chiang Rai, Chonburi, Khon Kaen, and Surat Thani, through presentations, exchange, and learning sessions with healthcare personnel.
Field engagement builds relationships and understanding of care settings. Multisite clinical evaluation is a separate part of the development plan.
06
Expanding shared learning
The next direction is to study use and image acquisition with more networks, including the training and support users need.
Expanded activity needs clear evaluation methods and goals, so each stage of collaboration answers useful development questions.
07
Extending acquisition and system readiness
Handheld ultrasound and integration with existing machines are directions we want to study, to understand which acquisition methods fit different services.
Each option needs testing for compatibility, image quality, and workflow before benefits or wider applicability can be established.
SOURCE NOTES
Sources and published project summaries
Project content is compiled and published by the EasyEF team from development records, presentations and study reports. Internal links lead to the team’s summaries, not additional independent evidence. Published research links to its original journal; study status and limitations are stated in the relevant articles.
Explore the source library →