After AIMed Cardiology and Radiology which took place this June in Chicago, Illinois. The latest round of AIMed sub-specialty events: AIMed Surgery and ICU are happening today (7 October) and tomorrow (8 October) respectively in Miami, Florida. These fields of work are chosen for a reason. Some of them, as we believe, will be ahead in the coming years of artificial intelligence (AI).
Primarily because many of the innovations and new technologies are relevant to their work and they are likely to receive the greatest benefits. As such, while AIMed flagship events are
Understand the lessons from man landing on moon
As AIMed Founder, Pediatric Cardiologist and Chief Intelligence and Innovation Officer, Children’s Hospital of Orange County (CHOC) Dr. Anthony Chang said in the recent AIMed Europe, humans are doing a great job in complex decision making; coming up with new solutions for problem solving but we are not good at aggregating a large volume of data. Besides, most medical professionals rely heavily on their experiences and gut feeling to perform their job. On the other hand, machines decide based on logic and evidence.
A human-machine synergy, will not only allow human wisdom to be seated upon new insights generated by
On top of which, Armstrong was also faced with an alarm upon landing, something which did not happen during his training and led him to think it might be a false alarm. Fortunately, the ground team quickly discovered it was a sign of system overload. The machine was receiving too much data and prevented subsequent errors as a result of cognitive oversight. With that, Dr. Chang urged for more people, especially those who are forwarding thinking or those who can understand either or both AI and medical language, to come forward and together, learn from prior lessons, and participate actively in the advancement of AI in medicine and healthcare.
Knowing AI beyond AI
Dr. Chang also criticized the over-emphasis on deep learning (DL) which may aggravate the Blackbox problem (i.e., the failure to comprehend how and why an algorithm derives at a given solution). Often, basic technology with clinical benefits is more productive than DL with little clinical advance. Hence, building new infrastructure or boosting existing ones to support such technology is as important as getting more data to develop complex algorithms. After all, a more accurate prediction does not necessarily lead to a better patient
Furthermore, Dr. Chang suggested looking into different aspects of AI such as Natural Language Processing (NLP) as there have been attempts to embed it into the workflow for report reading and extraction of information. The trend is likely to persist within healthcare and medicine. Sometimes, it is crucial to be fair to machines as some of them are indeed more superior in certain tasks. For example, chatbots are relatively more accepted than a human clinician in some instances, because they are deemed as less judgmental.
Dr. Chang said it is never too old to learn about AI, he went back to school in his 50s to do it and he believes other senior professionals are capable to do so too. He hopes AIMed will be a good stepping stone; if not the starter platform for some to know more about what is going on in the field and the opportunities that one can tap into. That’s why AIMed initiated the many events, webinars, and seminars throughout the year. Ultimately, it is the knowledge, not just the data, should be included in AI.
A science writer with data background and an interest in the current affair, culture,