Generative AI Use Cases in Medicine

OpenAI's work on synthetic voice generation could be immensely beneficial for patients. With this technology, my brilliant friend, Dr. Rohaid Ali, colleagues, and the Brown Neurosurgery Department have utilized OpenAI's voice engine to help a patient recover her voice after she lost her speech due to a vascular brain tumor. She was able to use a 15 second clip from a video for a class project to be the reference audio source for her synthetically reconstructed voice. For patients with neurodegenerative diseases or other chronic or as well as acute conditions that impact speech such as traumatic brain injury or stroke, this technology could help recover their voices using a simple voicemail they've left on a friend or family member's phone. A tremendous example of an application of AI in medicine that delivers real patient value and impact. This project also addresses what I've described as The Faster Horse Problem in AI (https://lnkd.in/eVv9WXqY). Instead of building a co-pilot, Dr. Ali and team are unlocking an entirely new class of tools for patients. With each new AI model and capability, we must work tirelessly on behalf of our patients to identify applications with such profound impact. Brown Neurosurgery Lifespan The Warren Alpert Medical School of Brown University #ai #artificialintelligence #generativeai #healthcare #medicine #neurosurgery https://lnkd.in/eB7hwwgd

Our new perspective piece in Nature introduces what we call "GenMI" –multimodal generative AI that transforms medical images into detailed reports. I'm particularly excited about our "AI resident" paradigm. Instead of replacing clinicians, these systems work alongside them – drafting initial reports, linking textual findings to relevant areas in images, and enabling interactive exploration. We envision three key benefits: - Reducing radiologist workload while preserving interpretation quality - Enhancing patient understanding through guided image exploration. Accelerating medical education with interactive feedback - Of course, challenges remain. We need better clinical benchmarks, transparency in model reasoning, and safeguards against over-reliance. The future of medical imaging isn't AI vs. humans – it's a thoughtful collaboration between both. A great collaboration with Eric Topol, MD & Vish Rao with Michael Hla Michael Moor Subathra Adithan Stephen Kwak Nature Portfolio. Harvard Medical School Department of Biomedical Informatics Harvard Medical School CC a2z Radiology AI, where we're working on building to solve the problems we lay out. #MedicalAI #Radiology #HealthTech https://rdcu.be/efcIG

🌟 Revolutionizing Clinical Trials with GenAI 🌟 This publication introduces a transformative framework for leveraging generative AI in clinical trials, addressing inefficiencies and biases to improve outcomes. 💡 The Challenge: Over 40% of clinical trials face significant flaws, wasting resources and delaying progress. Common issues include poor blinding, incomplete data, and inadequate diversity in participant selection. 🛠️ Proposed Solution: Develop Application-Specific Language Models (ASLMs) tailored for clinical trial design. These models, fine-tuned for the domain, can enhance protocol accuracy, reduce errors, and suggest best practices. 📋 Three-Phase Framework: 1️⃣ Regulatory Development: Agencies like the FDA create foundational ASLMs. 2️⃣ Customization: Health Technology Assessment bodies refine models for regional contexts. 3️⃣ Deployment: Researchers and trial designers access tools to improve protocols and submissions. 🌍 Key Benefits: ASLMs can address underrepresentation, predict safety issues, and ensure ethical, inclusive trials. They promise faster drug development, lower costs, and greater accuracy in trial outcomes. 🔗 Open Access and Collaboration: Advocates for open-source models to foster transparency, trust, and innovation, while maintaining rigorous oversight and validation. #GenerativeAI #ClinicalTrials #InnovationInMedicine #AIForGood #HealthcareTech #DiversityInTrials #MedicalInnovation #DrugDevelopment #EthicalAI #DigitalHealth