Making the results of a powerful heart imaging technique more reliable

Tensor-mapping to resolve the Achilles heel of echocardiographic strain imaging

Echocardiographic strain imaging is a new technique that uses ultrasound to look at the heart and measures changes in its shape and size (referred to as strain) as it beats. This technique can provide a more comprehensive and accurate assessment of heart function than other commonly used methods, and is useful for detecting early signs of heart disease. However, doctors do not routinely use strain imaging because scanning machines from different manufacturers give different results. This problem is difficult to solve because the algorithms used in the data analysis are kept secret. Strain is typically calculated via a series of steps, with each step only involving the result of the previous step, not the original data. Therefore, much data is lost in processing, often making results inaccurate, and conflicting between machines. A new way to calculate strain, called Tensor-Mapping, has been developed by Professor Darrel Francis’ team. They created a mathematical framework (a tensor) that allows all processing steps to happen simultaneously without discarding any data. This provides more reliable strain measurements, even when picture quality is below average. Professor Francis will develop Tensor-Mapping for routine use in heart clinics. The researchers plan to make their algorithms free to use and open for inspection, comparison and improvement. This research could lead to strain imaging becoming routine clinical practice for detecting a variety of heart problems.