Further Improvements in AI Technologies to Expand the Clinical Utilisation of 3D Echocardiography

Yoshihiro Seo, MD

Global longitudinal strain (GLS) has matured into a clinically applicable evaluation method based on research into conditions such as cancer therapeutics-related cardiac dysfunction (CTRCD). Because GLS values may change by only a few percent as we follow them over time, good reproducibility must be ensured. It’s also essential to minimise variation in the measurement results obtained by different examiners and to maximise reproducibility by employing automatic measurement methods developed by using AI. The number of examinations has been rising in recent years, and the number of measurement items may also increase substantially depending on the objectives of the examination. This means it’s more important than ever to perform examinations quickly and efficiently. In this regard, the automatic measurement functions developed by using AI are ideal tools. They not only reduce examination times but also ensure high data accuracy. This trend is expected to continue into the future. The left atrium is currently attracting a great deal of attention, as it’s been shown that its reservoir function strongly affects exercise tolerance and the prognosis of patients with heart failure^1,2. Examinations of the left atrium tend to be extremely time-consuming because the chamber has a very complicated shape, and variation in the measurement results obtained by different examiners is an important issue. If these challenges can be overcome by employing automatic measurement functions developed by using AI, we will have cleared a major hurdle in the introduction and widespread acceptance of such diagnostic imaging systems. It’s my belief that the left atrial reservoir function will be recognised as a useful index for assessing left ventricular diastolic function in the future. I expect that Canon will continue to introduce a wide variety of technological innovations to meet the needs of medical professionals in clinical practice, including automatic measurement functions with good reproducibility and high accuracy. //

Please note that the comments presented above are the views of the speakers.