The COVID-19 pandemic is running its apocalyptic course around the world. One thing is certain in the midst of total uncertainty and utter chaos: SARS-CoV-2 is a very daunting enemy and we remain totally subjugated by these viral overlords. It is easy, therefore, for we humans to lose patience and become careless as well as to decrease capacity for resilience. We need to regain our composure and to be defiant in order to for us to prevail this human vs virus struggle.
A major part of this resilience and defiance is continue our work in adoption and deployment of artificial intelligence concepts and projects in clinical cardiology and cardiac surgery.
Here are ten key takeaways from AIMed Cardiology 2020, part of our virtual Clinician Series in partnership with the American College of Cardiology:
My top ten takeaways:
- If one has to spend excessive time for the process of ETL (extract, transform, and lead) for data, it is difficult to scale AI projects in cardiology so this process needs to be more efficient.
- Use computer to work on data processing to reduce clinician cognitive burden for aspects of data collection and wrangling. Machine vs machines as the British mathematician Alan Turing always stated.
- EHR data is not analysis ready and one needs to first synthesize data into information and then go on to analysis. Data curation or munging can often take a lot of time and resource.
- “The less we need data garbage filter the better so we need to fix the garbage in problem without increasing human cognitive burden.” Registries is a stronger data repository.
- We need to have a faster and more agile way of doing clinical research (without always relying on publications even if these are faster than before) by being smarter about extracting insights from real-time data.
- It is essential for a mutual exchange of educational and learning: clinicians to invite data scientists to clinical rounds and activities and for clinicians to start some preliminary education for data (such as informatics) and data science (such as statistics and basics of machine learning).
- Wearable technology with some machine and deep learning will be a very powerful tool for outpatient cardiology especially when the coupled AI tools are even more accurate.
- We still need to work on explainability and trust for machine and deep learning in cardiology and this will mandate human cognition as part of AI in cardiology.
- Natural language processing (combined with other technologies like machine learning) is vastly under leveraged as an AI tool in cardiology in both assessment of chronic diseases as well as patient-to-physician communications.
- Bias in many types (statistical, social, etc) can be found in all aspects of the AI process in cardiology and can lead to inappropriate decisions in clinical management and/or resource allocation. This was observed in the pandemic.
Thank you faculty and attendees for your knowledge and expertise as we all learned a great deal at AIMed Cardiology !
Anthony Chang, MD, MBA, MPH, MS
Chief Intelligence and Innovation Officer
Medical Director, The Sharon Disney Lund
Medical Intelligence and Innovation Institute (mi3)
Children’s Hospital of Orange County