Healthcare AI Guy Weekly | 5/14

Interested in AI in healthcare? Join 5000+ readers (UHG, a16z, Google, etc)

Want to get in front of healthcare AI enthusiasts? Respond to this email!

Our Picks ✨

Highlights if you’ve only got 2 minutes…

1/

OpenAI new product release

OpenAI is in its 90’-93’ Michael Jordan era. The company continues to ship mind-blowing technology at a blistering pace. OpenAI just unveiled a new ChatGPT that listens, looks and talks called GPT-4o, the “o” in GPT-4o standing for “omni” referencing the model’s multimodal capabilities.

GPT-4o will effectively turn ChatGPT into a digital personal assistant that can engage in real-time, spoken conversations. If you’ve seen the movie Her, this is essentially that. OpenAI is betting that integrated text, audio, and visual will provide for a more natural human-computer interaction. This model, which also supports real-time translation and other interactive capabilities, is designed to function nearly as quickly as human response times in conversations.

Real-time voice and multimodal capabilities are shifting AI from a tool, to an intelligence we collaborate, learn, and grow with. We’ve already seen AI-powered companions take off and show effectiveness in improving mental health. With the advancements in GPT-4o, expect a wave of exciting and improved applications. (link)

2/

Google Deepmind releases Alphafold 3

Google DeepMind unveiled a new version of AlphaFold, a landmark tool for predicting protein structures, that will help scientists make breakthroughs in biology research. Alphafold sits within Google’s Isomorphic Labs, which is expected to be valued at over $100 billion following this release. Demis Hassabis, the CEO of both DeepMind and Isomorphic Labs, said, “I would be expecting, maybe in the next couple of years, the first AI-designed drugs in the clinic.” Here are some highlights of the model:

• AlphaFold 3 has a 50% improvement in predicting drug-like interactions compared to traditional methods.

• While AlphaFold 2 focused on protein structures, 3 can handle ‘all of life’s molecules’ and can model and predict complex interactions.

• The model is available freely for non-commercial use through the new AlphaFold Server, allowing scientists to generate predictions and accelerate research.

• Isomorphic Labs is already using AlphaFold 3 with pharmaceutical partners to design new drugs.

This new tool provides a more detailed and dynamic view of molecular interactions, marking a significant step toward modeling the dynamic systems of biology. AlphaFold 3 is expected to be particularly beneficial for drug discovery, as it can predict where a drug binds a protein. This more powerful and accurate iteration of Alphafold will supercharge drug discovery and our knowledge of the biological world! (link)

3/

Healthcare data remains untapped

Arcadia, a healthcare data and analytics platform, and HIMSS, Healthcare Information and Management Systems Society, analyzed how hospitals are using the 137 terabytes of data they produce everyday. They found that a vast amount of data remains entirely untapped despite the proliferation of AI.

Although four out of five healthcare leaders believe most of their data is accurate, 47% of healthcare data is underutilized when making clinical and business decisions. Healthcare leaders named enhancing data literacy (58%), using AI (47%), and addressing productivity challenges (34%) as key priorities to make data more usable. AI models are only as good as their inputs so ensuring proper data capture and the quality of data is foundational to the effectiveness of AI in healthcare. (link)

Tools & Partnerships 🔧

Latest on business, consumer, and clinical healthcare AI tools and partnerships…

TOOLS

• AI-discovered drugs have an 80%–90% success rate: AI-discovered molecules can have an 80%–90% success rate in Phase 1 clinical studies, a group of researchers from Boston Consulting Group said in a recent peer-reviewed article in the journal Drug Discovery Today. (link)

• Hospitals using AI to reduce violence against clinicians: Parkland Memorial Hospital in Dallas developed a predictive AI tool within its EHR to generate a risk assessment score that informs clinicians which patients are more likely to be violent. A test of the AI tool revealed that for every 1,000 patient-clinician interactions at Parkland, 7.1 violent events were correctly predicted, compared with 2.3 violent events missed. (link)

• AI eye-tracking tech to diagnose neurodegenerative conditions: French-Belgian medtech startup neuroClues is building eye-tracking tech that uses AI to watch how a patient’s eyes move. The goal here is to diagnose neurodegenerative conditions. neuroClues is starting with a focus on Parkinson’s. (link)

• Veda achieves AI precision validation: Veda, a leading health technology company specializing in provider data solutions, announced today that it has achieved third-party validation from the prestigious Erdős Institute, an independent organization of university PhDs advancing the fields of Data Science and Machine Learning. The independent audit finds Veda's AI precision exceeds 90%, solving inaccurate and outdated provider directory and payer network attestation challenges. (link)

PARTNERSHIPS

• Suki + Premier: Suki, a voice-assistive AI and scribe tool, inked a national group purchasing agreement with Premier Inc. The agreement could skyrocket Suki to the top of the healthcare AI assistant market, with Premier representing 4,350 hospitals and health systems. (link)

• Intermountain Health + Memora Health: Intermountain Health is partnering with San Francisco’s Memora Health on a care management platform designed to reduce the workload on nurses and give patients quick and seamless access to the resources they need. This platform uses AI to help patients with their care plan once they’ve left the hospital or doctor’s office, answering patient questions and guiding them to online resources. (link)

• Abridge + UChicago: Abridge announced that the University of Chicago Medicine will make Abridge's generative AI platform for clinical documentation available to groups of clinicians across their system. (link)

Deal Desk 💸 

Spotlight on latest capital raises, M&A, and investments…

FUNDING

• Rad AI, an AI-powered startup that helps radiologists save time on report generation, raised a $50M Series B from Khosla Ventures and others. (link)

• Sift Healthcare, an AI-powered healthcare payment solution, raised a $20M Series B funding round. Led by B Capital, with participation from others. (link)

• Triomics, a SF-based company using generative AI to streamline workflows for cancer centers, raised $15M in Series A funding from Lightspeed, Nexus Venture Partners, General Catalyst, and Y Combinator. (link)

• In-House Health, an AI-powered scheduling and management platform for modern nursing teams, raised $4M in seed funding. NEA and TMV led the round and were joined by existing investors Vine Ventures and Longevity Venture Partners. (link)

• OptMyCare, a clinically engineered AI platform in healthcare & risk management, raised $3M in Series A financing led by LiveOak Ventures. (link)

MERGERS & ACQUISITIONS

• Samsung’s Medison + Sonio, Samsung’s medical device unit, Medison, will pay $92.7M for French AI startup, Sonio. Sonio works with ultrasound software with a focus on pregnancy care. Crunchbase reports that Sonio raised €28 million while private. (link)

• Quest Diagnostics + PathAI Diagnostics, Quest Diagnostics (NYSE: DGX), a leader in diagnostic information services, and PathAI, a global provider of AI-powered technology for pathology, announced that Quest will acquire select assets of PathAI Diagnostics, the business of PathAI that provides anatomic and digital pathology laboratory services. (link)

Other Relevant News 🔍

News, podcasts, blogs, tweets, resources, etc…

• Is AI healthcare's new 'shiny object'? (link)

• Physicians cautiously starting to employ AI in medical practices (link)

• AI chatbots for mental health raises concerns (link)

• AI devices aid healthcare but fears linger about nurses’ role (link)

• Intermountain Health CEO discusses company investment in AI to reduce costs (link)

• Opinion: AI in healthcare will face legal liability challenges (link)

• Sam Altman on the All-in podcast (link)

Visuals of the Week 📸

Funny memes, cool pics, and interesting data from around the web…