Hello, I’m Stephen M. Borstelmann MD. I am a Florida-based Board-Certified physician trained in Interventional Radiology and a father to three amazing children. My diverse professional journey spans clinical practice, administration, research, and most recently, the intersection of medicine and artificial intelligence (AI). Before I was a physician, I was an options trader on Wall Street – a ‘quant.’ Lately, I’ve been most well-known in the medical AI community (@drsxr across social media platforms) for my commitment to ethical and scientifically sound AI practices.
Leveraging my coding and quant stats skills from my trading days, I’ve ventured into machine and deep learning. This transition, driven by curiosity and a desire to enhance health and well-being through technology, has led me to advocate for algorithmically and statistically sound AI in medicine. I educate, lecture, and publish research to contribute to this rapidly evolving field always with an eye towards fairness and elimination of unwanted bias.
In the early days of the AI revolution, many physicians felt blindsided by predictions of AI replacing them within five years. A small group of us, familiar with these computing techniques, stepped forward to challenge these claims. We spoke to technologists and computer scientists in their language, advocating for a more balanced perspective. Today, physicians are integral to the AI conversation, designing AI algorithms for patient care improvement. This collaborative effort has been incredibly rewarding and underscores my belief in the power of diverse perspectives.
Selected publications include:
- Opportunistic detection of type 2 diabetes using deep learning from frontal chest radiographs in Nature Communications
- Validation of a deep learning value-based care model to predict mortality and comorbidities from chest radiographs in COVID-19 in PLOS Digital Health,
- Detecting racial/ethnic health disparities using deep learning from frontal chest radiography in the Journal of the American College of Radiology.
- Machine Learning Principles for Radiology Investigators in Academic Radiology
The current state of healthcare in the USA is challenging. Rising costs, administrative rigidity, and a growing knowledge gap among healthcare providers are pressing issues. Despite these challenges, advancements in pharmaceuticals have led to remarkable progress in treating previously untreatable diseases. However, the high costs of these treatments and accessibility issues persist. I believe AI has a significant role to play in addressing these challenges.
My mission remains clear: to advocate for scientifically sound AI algorithms in patient care. We have one chance to get this right, and any errors or biases we introduce into our AI systems now will impact us for decades to come. I’m committed to ensuring that the future of AI in healthcare is not only innovative but also ethical and beneficial to all members of society.
IMPORTANT: NOTHING IN THIS BLOG SHOULD BE CONSTRUED AS MEDICAL ADVICE, DIAGNOSIS OR TREATMENT, OR ESTABLISHING A PHYSICIAN-PATIENT RELATIONSHIP. IF THIS IS AN EMERGENCY, OR YOU ARE UNWELL OR SYMPTOMATIC, YOU ARE INSTRUCTED TO IMMEDIATELY CONTACT YOUR PHYSICIAN, OR CALL 911. I AM NOT A LAWYER, TAX EXPERT, OR FINANCIAL ADVISOR. NOTHING IN THIS BLOG SHOULD BE CONSTRUED AS MEDICAL, FINANCIAL OR LEGAL ADVICE. YOU ARE DIRECTED TO CONTACT YOUR OWN PHYSICIAN, ATTORNEY, TAX ACCOUNTANT, OR FINANCIAL ADVISOR FOR SPECIFIC ADVICE PERTAINING TO YOUR OWN CIRCUMSTANCES. BLOG OR SOCIAL MEDIA FOLLOWS, SHARES, RETWEETS, COMMENTS OR POSTS SHOULD NOT BE CONSTRUED AS ENDORSEMENT. POSTS MAY BE MADE FOR ENTERTAINMENT PURPOSES ONLY. I RESERVE ALL RIGHTS TO POSTED CODE WHICH MAY ONLY BE USED FOR NON-COMMERCIAL RESEARCH PURPOSES ON AN AS-IS BASIS WITHOUT ANY WARRANTIES.
DO NOT POST PROTECTED HEALTH INFORMATION ON THIS BLOG. PROTECTED HEALTH INFORMATION WILL BE DELETED.
PRIVACY POLICY: YOU SHOULD ASSUME THAT ALL VISITS TO, PAGE VIEWS, OR POSTS ON THIS BLOG ARE PUBLIC INFORMATION AND MAY SUBJECT YOUR DATA OR META-DATA TO PUBLIC OR THIRD PARTY USE. Seriously. I am not intentionally using cookies or data-mining you (I have much more interesting data to mine, frankly) but I cannot personally review wordpress code, so while I believe none of you information is being collected by ME, apart from whatever info you leave in a comment or by subscribing to the blog, I have no idea whether someone else is. Its the internet. Behave responsibly.
ALL WRITTEN MATERIAL COPYRIGHT© 2014,2015,2016,2017,2018, 2020, 2023