Written by Sajin Saju
For Dr Lizy Abraham, securing a grant of over half a million euro is not just a moment of elation but a critical step in bringing to fruition her research exploring the power of artificial intelligence to diagnose congenital heart diseases (CHDs).
“It was the result of a year-long dedicated effort,” emphasises Dr Lizy, a research associate at South East Technological University (SETU), after being awarded €556,070 in funding under the SFI-IRC Pathway programme for her research, titled ‘An Artificial Intelligence (AI)-based Automated Approach for the Classification of Pediatric Heart Murmurs and Disease Diagnosis using Wireless Phonocardiography’.
CHDs are malformations that occur due to abnormal development of the heart at the birth of a child. According to Heart Children Ireland, the condition affects one in every 100 babies born in Ireland per annum, which equates to approximately 500-600 children born each year.
“In Ireland, patients typically go to general practitioners when they are sick, but the technology to accurately diagnose complex conditions like heart murmurs, especially in children, is not always available,” the Waterford-based researcher points out in an interview with the Irish Indian Chronicle. “This gap in medical technology inspired me to focus on AI-based solutions that can provide accurate, reliable diagnoses and ultimately improve patient outcomes.”
While echocardiography is considered the gold standard for the diagnosis of CHD, the procedure has limitations when performed on children. It takes almost 30 to 45 minutes, and children should remain still during the process.
As per the project proposal, referring all children with a murmur for this expensive and lengthy diagnostic procedure is neither feasible nor cost-effective.
This is where Dr Lizy proposes a solution — an automated AI-based cardiac disorder detection system using the phonocardiogram (PCG) heart sound signals. PCG signals, obtained with digital stethoscopes, in combination with the power of AI, can provide an objective interpretation of heart sounds to complement traditional auscultation methods.
Dr Lizy believes that this system will significantly aid physicians and primary health care providers in the early diagnosis of CHDs, reducing the need for children to undergo expensive diagnostic procedures directly.
The project will commence in September 2024.
This is not Dr Lizy’s first time in the spotlight. She previously garnered attention as the principal investigator of the WESAT (Women Engineered Satellite), India’s first satellite payload developed by an all-female team. It was launched by the Indian Space Research Organisation (ISRO) in the 60th Polar Satellite Launch Vehicle mission on January 1, 2024.
Originally from the southern Indian state of Kerala, Dr Lizy holds a PhD in Electronics and Communication Engineering from the University of Kerala, India, and completed Post Doctoral Research in Wireless Sensor Networks from Tyndall National Institute, University College Cork. Currently, she heads the Emerging Networks Labs Division at Walton Institute, SETU.
Challenges ahead — Reliable AI model, parental consent, pediatric data
Dr Lizy acknowledges that carrying out a project of this magnitude comes with a multitude of challenges. “One significant challenge is integrating complex, heterogeneous data from various sources into a unified AI model that is both accurate and reliable. Ensuring the system’s performance across diverse patient demographics requires extensive validation and testing,” she says.
Obtaining pediatric data from hospitals would be difficult, especially when it involves sensitive data about children, she mentions, before adding: “Gaining personal consent is a necessary but challenging process, as it involves ensuring parents fully understand the scope of the research and implications.”
Securing collaboration with doctors, she says, has been fraught with complexities. “I spent considerable time reaching out to numerous doctors, navigating various administrative and logistical barriers. The process was particularly challenging because it involved integrating new technologies into their existing clinical workflows, often requiring convincing them of the potential benefits and feasibility of our AI system,” she explains.
Tips for early-career researchers
The path to securing funding remains challenging and competitive, Dr Lizy says, before sharing a few tips for early-career researchers.
“My advice is to build a strong foundation in both technical and domain-specific knowledge. Focus on addressing real-life problems through your research and seek practical solutions. Craft your proposals to clearly outline the significance of your research, how it addresses pressing issues and its potential impact on society,” she says.
“Embrace feedback from rejected proposals to improve and strengthen your future applications, and remain determined despite setbacks,” she continues.
She suggested that seeking mentorship from experienced researchers and actively engaging with the academic community through conferences, workshops, and seminars could also offer opportunities to present work and establish valuable connections.