Stress Echocardiography - Isis Project No 6952

Segmentation of stress echocardiography sequences using rest-based patient-specific prior information.

Subjective Intepretations

In stress echocardiography, a heart (or typically the left ventricle) is imaged using ultrasound at rest and then after injection of a stress-inducing drug such as dobutamine, or exercise. Clinical analysis is typically 2D and focuses on visualising these 2 sequences (rest and peak stress) side by side. Importantly this means interpretation is patient-specific. A sick heart moves differently under stress than a healthy one and the clinician will grade this motion by eye and from this decide whether the heart is abnormal or not. This is a subjective interpretation.

Automating this task is difficult because of the lower quality of stress data relative to rest data meaning methods developed for rest data do not work well on stress data. This invention concerns automating stress echo data by using the result from tracking the higher quality rest data to constrain the segmentation of the lower quality stress echo data.

Efficient and Qualitative

Current stress echocardiographic analysis is by trained experts and opinions can vary between clinicians i.e. it is very subjective and qualitative. This new technique not only makes the process more efficient but also gives more reliable quantitative information such as confidence measures.

More importantly, it is repeatable with the same outcome on the same dataset each time and therefore less subjective.

As it has been implemented at the moment, it can be used in a clinical environment to help automate and quantify the process of stress echocardiography analysis, and this could be used as a screening tool to highlight abnormalities in motion which may require further investigation.

More generally, the invention could be used to relate any two sequences of segmentations for any organ or object whose motion may be thought to be linked, for instance diagnostic analysis of the lung, or fetal heart, or where motion assessment was done pre- and post- therapy, or where respiration motion compensation was required and a normal model was available The strength of the model lies in being able to derive a strong correlation from training data such that accurate prediction is possible when the model is applied to a new dataset.

Patent Status

The Oxford invention is the subject of a patent application. Isis would like to talk to companies interested in developing the commercial opportunity. Please contact the Isis Project Manager.

Request Further Information: Project Number 6952 - Stress Echocardiography