Artificial intelligence (AI) tools have proven to be more effective than the standard clinical method in predicting heart transplant rejection, according to a recent study conducted by researchers from Emory University, Case Western Reserve University, and the University of Pennsylvania. The study, published in Circulation: Heart Failure, highlights the potential of AI in improving patient outcomes and treatment decisions.
In the United States alone, more than 4,500 heart transplants were performed in 2023. While this procedure offers a new lease of life for many recipients, the risk of organ rejection remains a significant concern, with up to 32% of patients experiencing acute rejection within the first year.
Currently, clinicians rely on histologic grading of cardiac biopsies to diagnose acute rejection. However, this method has limitations, as it assigns subjective grades that lack diagnostic accuracy. The researchers recognized the need for a more precise and informative method to predict rejection outcomes and developed an AI tool called the Cardiac Allograft Rejection Evaluator (CARE).
CARE utilizes AI algorithms to analyze cardiac biopsy images and extract features related to the shape, texture, and spatial architecture of muscle cells, immune cells, and stromal fiber in heart tissue specimens. By leveraging these features, the tool can accurately predict rejection outcomes for heart transplant patients, empowering clinicians to make informed treatment decisions.
The study, which included 2,900 patients, demonstrated that the CARE model outperformed the standard clinical method in predicting cardiac rejection severity. Moreover, the AI model used transparent and intuitive image features, making it more easily interpretable for clinicians compared to other opaque “black box” AI models.
The ultimate aim is to equip pathologists and cardiologists with tools that enable them to make precise and well-informed decisions regarding heart rejection cases, says Dr. Sara Arabyarmohammadi, one of the researchers involved. By providing access to more accurate predictions, aggressive treatments can be applied when necessary, leading to better prevention of heart transplant failure.
This breakthrough in AI-driven prediction of heart transplant rejection holds great promise for the field of transplantation medicine. It enables healthcare professionals to intervene proactively and provide tailored treatments, ultimately improving patient care and post-transplant outcomes. With further advancements in AI technology, we can expect even better accuracy and reliability in predicting rejection outcomes and enhancing the success rates of heart transplants.
*Note:
1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it