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 radiology. Here are ten key takeaways from today’s AIMed Clinician Series/Radiology webinar in partnership with the American College of Radiology:
My top ten takeaways:
- The Data Science Institute of American College of Radiology had put together a COVID response with AI-enabled projects to increase the capabilities of radiologists during this historic time for health care to decrease likelihood of burnout.
- The pandemic has helped to accelerate the process of data integration with breaking down of traditional silos of data and lack of collaboration amongst centers; this forced disruption is also seen with efforts in innovation as well.
- Burnout is a real current issue with a decrease of medical image volume (up to 70%) and fear of job loss. A good intervention for mitigating professional burnout (Geraldine McGinty) is to spend more time with your patients (even with your PPE).
- The biggest advance by far in AI projects in radiology has been the availability of medical image data and not always necessarily advances in algorithms or AI techniques (although both together have pushed the AI agenda).
- Medical data are people (Matt Lungren) so the medical image data need to be in a position for social good via open collaborations and not always used as commodity for enterprises (the latter not always with social good as top priority).
- The predictive analytics workflow for radiology needs checkpoints for security (as many as 40 steps warrant security measures) to ensure that medical image data security is enabled throughout the entire process.
- We need to continue to build a data scientist-clinician interface as the AI tools are more available and mature in radiology; clinical cognition as input for these projects are more important than ever before.
- AI tools are already very useful for automatic detection and segmentation as well as volume measurements for a myriad of lesions on medical images so this support can mitigate the burden and stress for radiologists.
- The tool of natural language processing (NLP) hybridized with machine and even deep learning (CNN) is vastly under-leveraged in radiology; this is an essential tool in the annotation of medical image data especially with complicated and unstructured data as well as with radiology workflow.
- Part of the recovery process in radiology and other subspecialties should involve a new paradigm of using AI tools for more efficiency in workflow, better allocation of resources, and improved patient outcomes so we do not go back to pre-COVID19 state.
- AI in medicine will be perceived as successful when we no longer talk about AI in medicine as if it is a separate part of medicine but more that AI is embedded in clinical medicine and healthcare in many ways.
Thank you faculty and attendees for your knowledge and expertise as we all learned a great deal at AIMed Radiology !
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