Congenital Heart Disease

‘Tetralogy of Fallot’ is a relatively common syndrome of the heart, affecting 3 in every 10,000 babies. This syndrome affects the heart in several ways and can have various levels of impact. In severe cases, surgical treatment shortly after birth is required. However, later in life many patients require regular evaluation to monitor further progression of heart failure. 

Treatment options to prevent heart failure include pulmonary valve replacement. If this intervention is done too early in life it may need to be done again in several years’ time; or it may have not been required at all. On the other hand, if the valve is replaced too late in life, the heart does not recover leading to the progression of heart failure despite the attempt to prevent it. Medical imaging examinations such as magnetic resonance imaging (MRI), computed tomography and echocardiography, are frequently used to evaluate heart function and determine the best timing for interventions.

This project aims to provide doctors with more information on the progression of disease in patients with congenital heart disease and help identify the ideal time for intervention. To do this, Artificial Intelligence (AI) tools will automatically measure heart shape and function seen on medical images. 

In the future, the hope would be for this tool to generate a report for clinical staff detailing changes identified over time and between successive scans. The reporting would also identify and evaluate new markers of heart function that could then be studied to determine if they appear in relation to outcomes from interventions. 

Improved information for clinicians would allow them to better understand the health of the heart at that moment, identify the best time for an intervention if it’s needed and predict the success of the intervention; this will ultimately improve patient outcomes and personalise interventions to them. With a new method to predict the outcome of interventions, patients can receive intervention at the point of the greatest chance of success, avoid unnecessary interventions, and benefit from improved outcomes.

Published Academic Papers

1. Govil S, Crabb BT, Deng Y, Dal Toso L, Puyol-Anton E, Pushparajah K,  Hegde S Perry JC, Omens JH, Hsiao A, Young AA, McCulloch AD. (2023) A deep learning approach for fully automated cardiac shape modeling in tetralogy of Fallot. Journal of Cardiovascular Magnetic Resonance 25, 15.
2. Govil S, Mauger C, Hegde S Occleshaw C, Yu X, Perry PC, Young AA, Omens JH, McCulloch AD. (2023) Biventricular shape modes discriminate pulmonary valve replacement in tetralogy of Fallot better than imaging indices. Sci Rep 13, 2335.  
3. Mîra A, Lamata P, Pushparajah K, Abraham G, Mauger CA, McCulloch AD, Omens JH, Bissell MM, Blair Z, Huffaker T, Tandon A, Engelhardt S, Koehler S, Pickardt T, Beerbaum P, Sarikouch S, Latus H, Greil G, Young AA, Hussain T. (2022) Le Cœur en Sabot: shape associations with adverse events in repaired tetralogy of Fallot. J Cardiovasc Magn Reson. 24(1):46.
4. Elsayad A, Mauger CA, Ferdian E, Gilbert K, Scadeng M, Occleshaw CJ, Lowe BS, McCulloch AD, Omens JH, Govil S, Pushparajah K, Young AA. (2022) Right ventricular flow vorticity relationships with biventricular shape in adult tetralogy of Fallot. Front Cardiovasc Med 8:806107. 
5. Mauger CA, Govil S, Chabiniok R, Gilbert K, Hegde S, Hussain T, McCulloch AD, Occleshaw CJ, Omens J, Perry JC, Pushparajah K, Suinesiaputra A, Zhong L, Young AA. (2021) Right‑left ventricular shape variations in tetralogy of Fallot: associations with pulmonary regurgitation. JCMR 23:105