Special webinar with Dr Sandeep Reddy | Translating AI in healthcare

Webinar recap: Integrating AI into healthcare 

Dr Reddy identified some key opportunities and challenges for future research to help address some of the issues currently facing the integration of AI into clinical settings.

 “On balance, I am an optimistic person, so I think we do have many solutions for the challenges of implementing AI in healthcare”, Prof. Reddy said.  He went on to highlight the potential of AI in medical imaging, clinical decision support, and multimodal applications.

However, he was also careful to explain the limitations and key challenges for both clinicians and researchers. Touching on AI’s current opaque decision-making issues, i.e. when AI cannot explain where or why it came to its output, goes rogue, or proliferates misinformation, as well as concerns around coded bias. While many of these problems have potential solutions like introducing Explainable AI, the biggest challenge he sees is the environmental impact and the increasing footprint of large language models.

Dr Reddy wrapped things up by looking towards the future with some potential solutions and areas of research that could help close the gap between AI research and real-world application. He called for more research, more access and opportunities for data gathering to train AI models, investigating small language models that consume less energy, and a review of clinical guidelines to identify opportunities for incorporating AI into clinical settings just to name a few.

Overall, Dr Reddy was positive and confident about the integration of AI in healthcare, especially its potential to reduce stress on clinicians.