Buzzworthy developments of the past few days.

• AI is on the cusp of transforming surgery. That’s the impression one comes away with after reading three articles on AI in the April Bulletin of the American College of Surgeons. “We’re going to be building our shared knowledge [with AI] to create what we call a shared surgical consciousness, one that holds more knowledge than any single surgeon can acquire,” says Ozanan Meireles, MD, inaugural vice-chair for innovation at Duke University’s surgery department, in one of the articles. “That collective surgical consciousness can guide us away from complications and truly improve patient care.” Access all three pieces here.
   
• AI is shaking up healthcare inside the Military Health System. At a conference of the MHS earlier this month, attendees learned the Department of Defense’s Health Affairs division is “working to identify the actions needed to enable advances and adoption of artificial intelligence and machine learning across military medicine.” Read DoD’s own coverage.
   
• When it comes to using AI for getting things done in real-world settings, no other industry has to be as careful as healthcare. That’s because no other industry has lives in the balance in so immediate a sense. Rajesh Viswanathan, chief technology officer for the healthcare software supplier Inovalon, considers the responsibility in a piece published by InsideBigData. “The highly regulated nature of our work, and the significant requirements around having supporting evidence for claims or decision-making, remind us that patient safety must always be top of mind,” he writes before running through several ways AI is already making a difference in healthcare, albeit in nonclinical settings. Read the rest.
   
• Europe is a major healthcare market but not just for any seller of healthcare software. This is to say that American digital-health startups hoping to sell their stuff in the Old Continent should know what regulatory hurdles they’ll need to clear under the EU’s still-new Artificial Intelligence Act, which passed just two months ago. The outlet MD+DI (for Medical Device and Diagnostic Industry) lays out the essentials in an up-to-date article published April 16. Look past the misplaced apostrophe in the headline and get edified by the rest.
   
• Be afraid. Be at least somewhat afraid. Researchers have shown that facial recognition technology can discern political beliefs—even in people doing their best poker face. One of the lead researchers, organizational behaviorist Michal Kosinski of Stanford, calls the demonstration of the technology’s prowess “inconvenient.” For ideological reasons, adds Kosinski, who is also a computational psychologist and a psychometrician, “scientists prefer to avoid discussing links between appearance and traits. However, companies and governments are keen to use facial recognition to identify intimate traits.” PsyPost has the story.
   
• Cedars-Sinai Technology Ventures is on a roll. The Los Angeles-based operation now manages upwards of 500 works-in-progress technologies, oversees more than 1,100 patents and, over the past five years, has generated more than $1 billion in tech-transfer revenues. The work includes the Cedars-Sinai Accelerator, which exists to “fast-track the development of the next generation of ideas that will transform patient care at Cedars-Sinai and beyond.” Learn more.
   
• It’s hard to think of an industry further afield of healthcare than earth-moving equipment. Maybe it’s me, but I think that’s a great reason to spend a couple of minutes boning up on how a major player in the latter field is using AI. “Using Gen AI, Caterpillar’s condition monitoring advisors are now presented with a concise report containing data [on equipment assets in the field] that has been automatically prepared and summarized,” says the manufacturer of construction and mining equipment, off-highway diesel and natural gas engines, industrial gas turbines and diesel-electric locomotives. “There’s still a human in the loop, but the time required to prepare data and create a recommendation [for the customer] is significantly reduced.” Fascinating.
   
• Recent research roundup: 
   
  • Tel Aviv University: GPT versus Resident Physicians — A Benchmark Based on Official Board Scores (NEJM AI) 
     
  • Sanford Burnham Prebys: AI tool predicts responses to cancer therapy using information from each cell of the tumor
     
  • Mass General Brigham: Study Finds ChatGPT Shows Promise as Medication Management Tool, Could Help Improve Geriatric Health Care
     
  • University of Cambridge: Artificial intelligence beats doctors in accurately assessing eye problems
     
  • Penn State: Novel robotic training program reduces physician errors placing central lines
     
• AI investments of note:
   
  • Microsoft invests $1.5 billion in Abu Dhabi’s G42 to accelerate AI development and global expansion 
     
  • HealthSage secures €3M (US ~$3.2M) seed investment for open source AI healthcare platform
     
  • German Center for Neurodegenerative Diseases: US $1M for mapping the springs of the spinal fluid 
     
• From AIinHealthcare’s news partners:
   
  • Health Imaging: Even AI struggles to work under stress
     
  • Radiology Business: GPT-4 can detect radiology report errors at the same rate as members of the specialty
     
  • Cardiovascular Business: New ACC President Cathie Biga wants to improve quality reporting in cardiology