As mentioned, AIMed founder and chairman, chief intelligence and innovation officer of Children’s Hospital of Orange County (CHOC) spoke at a panel discussion on integrated diagnostics at this year’s European Congress of Radiology

Dr. Chang was joined by Jörg Aumüller, vice president, global head of digitalizing healthcare marketing of Siemens Healthineers who acted as facilitator, Dr. Peter Chang, co-director at University of California Center for AI in diagnostic medicine and Dr. Daniel Rubin, professor of biomedical data science, radiology and medicine at Stanford University, in a session entitled “Radiologists or clinical data scientists? Expanding precision medicine with AI-powered integrated diagnostics”. 

Apart from reassuring the crowds that artificial intelligence (AI) will not replace radiologists. Dr. Peter Chang had highlighted some of the key opportunities which AI has to offer. 

What can deep learning do for radiology? 

According to Dr. Peter Chang, the primary advantage of deep learning lies in detection and triage, especially in urgent findings for prioritized interpretation. Often, radiologists do not read and access medical condition of one patient. In time of pressing needs, AI can provide a second opinion in the quickest time possible. 

Deep learning is also handy in quantification and reproducibility. AI quantifies a condition, for example potential strokes and blood in the brain. AI is able to inform physicians, with near human accuracy, the amount of blood in the patient’s brain and chances of a medical occurrence. Nevertheless, as Dr. Peter Chang noted, AI does not tease out a diagnosis. It is providing a recommendation or aspects which radiologists should pay attention to. At the end of the day, it is the radiologists and physicians themselves who make the final judgement. 

In post-image processing, deep learning is able to reconstruct case-based data without introducing noise. It leverages on existing data to create patterns, which can be turned into significant predictive tools. 

There is still a need to for transparency and results

The Center for the Governance of AI and Oxford University’s Future of Humanity Institute published a new study at the turn of the year, to highlight some of the concerns Americans have for artificial intelligence (AI). In general, Americans have a mixed feeling towards AI as they are unsure if it is something positive. Others believe advancement in will actually do more harm than good. Interestingly, Americans trust tech companies more than the federal government to manage AI. 

As the results were collated based on the feedbacks given by 2000 respondents, its representativeness is questionable. Likewise, we cannot assume the respondents probably are unaware of the Theranos scandal and confusion generated by Nvidia’s deepfake technology, to put their faith on tech companies. To be fair, the sense of unease could also be felt among radiologists, the “techiest” of all healthcare professionals

Some ECR delegates noted the importance of transparency, the need to inform healthcare professionals, the kind of data that have in place to develop the algorithm. Others emphasized on evidence and results. AIMed is taking the first step to get the conversation going, by hosting AIMed Radiology between 18 and 19 June at Ritz-Carlton, Chicago, Illinois. We aimed to provide those who are interested or those who are curious about what AI and deep learning are capable of, an avenue to understanding more. We also hope that a one of its kind event like this, will erase the hype and get people excited about the true potential of AI and its related technology. The preliminary conference agenda is now out, do keep an eye for the latest updates on the event page. We look forward to meeting you at AIMed Radiology between 18 and 19 June at Ritz-Carlton, Chicago, Illinois.  

Author Bio
synthetic gene empathy chinese artificial intelligence data medicine healthcare ai

Hazel Tang

A science writer with data background and an interest in current affair, culture and arts; a no-med from an (almost) all-med family. Follow on Twitter.