Richard Chen

This research from NEJM AI shows the potential of large language models (LLMs) like GPT-4—the technology behind ChatGPT—in addressing disparities in oncology expertise. The study uses 2,044 oncology questions to demonstrate GPT-4’s superior performance, marking progress toward effective AI integration in clinical settings. The research reveals a key improvement area: all models, including GPT-4 and Claude-v1, struggled with questions on female-predominant malignancies, indicating a need for targeted model training. Nicholas Rydzewski and his team introduced methods using self-rated confidence scores and repeating prompts, which forms a feedback loop, enabling medical professionals to refine AI reliability and enhance decision-making. This study establishes standards for AI in healthcare and starts conversations about what AI can and can't do in medicine. #AIinHealthcare #ClinicalOncology #MachineLearning #HealthAI

28

2 Comments

If you are involved in developing LLM applications, you may find this resource useful. It highlights key strategies for addressing the challenges in LLM application development. #LLMDevelopment #DataQuality #AIApplications 

3

In this week's For Your Innovation, founding chair of the Department of Machine Learning at Moffitt Cancer Center Dr. El Naqa shares how Artificial Intelligence technologies support and enhance diagnostic and treatment processes, especially in the oncology field. Watch to learn about the practical and potential applications of AI in transforming patient care and treatment efficiency. https://arkinv.st/3UFw2nB

6

2 Comments

Stanford University School of Medicine magazine showcases the rich history of AI at Stanford and highlights many of our faculty members who are at the helm of melding powerful advancements in AI with medicine and health care. The recent buzz around AI isn't new, but excitement is building around use of AI to reveal insights and connections we haven't seen before. "AI can be, in some ways, superhuman because of its ability to link disparate data sources," says AIMI Center Director Curtis Langlotz, commenting on the technology's capabilities. "It can take genomic information and imaging information and potentially find linkages that humans aren't able to make. Sylvia Plevritis, PhD, a professor of biomedical data science and of radiology, is also one of the leaders of the RAISE-Health initiative. She developed a systems biology cancer research program that bridges genomics, biocomputation, imaging and population sciences to decipher properties of cancer progression. "Today, AI is completely changing the way we connect the dots between basic science research and clinical care," said Plevritis, who is in the Stanford Medicine leadership group focusing on what this new path means for medical education. If you're new to #AI, large language models and machine learning, check out AIMI Center Co-Director Dr. Nigam Shah's "AI 101" list of definitions in the story. Visit https://lnkd.in/dHw-TcJt to read more from our Stanford Medicine and AIMI Center experts! Interested in diving deeper? Join us for the upcoming RAISE and AIMI symposiums, May 14 and 15, for in-depth discussions on the pioneering developments in AI that are shaping healthcare! RAISE Symposium - May 14 Time: 8:30 a.m. – 1 p.m. PDT Details and registration: https://lnkd.in/eZWBtMmk AIMI Symposium - May 15 Time: 9 a.m. - 4 p.m. PDT Details and registration: https://lnkd.in/gtjPfiAE #AIMI24 #ResponsibleAI #AIInHealthcare #MachineLearning #AIMIsymposium #RAISEhealth #RAISEhealthSymposium

198

1 Comment

We're in a Nature publication 🎉 "Human-in-the-loop avatar chatbot shows promise in supporting hematopoietic stem cell transplantation patients": https://lnkd.in/gxDJgZmi So proud of the team at Harvard University / Dana-Farber Cancer Institute (Amar Kelkar, Nicholas Groblewski, Dillon Clancy, Sara Close, Lauren Sullivan, Corey Cutler, Hajime Uno & Dr. Greg Abel) & care.coach (Chantal Kerssens, PhD, Danielle Sherman-Manhard, and many folks behind the scenes)! It took a lot of doing, but across 3 cohorts, we were able to beat each of our targets: >90% study completion, >66% interventional retention, <20% dissatisfaction, and >75% completion of educational modules. I want to point out the leadership of the #Harvard+#DFCI team in taking the traditionally inpatient transplant process and transforming it into outpatient treatment, allowing patients undergoing this scary situation (which involves wiping out your entire immune system and replacing it) to be in the comfort of their own homes, freeing up hospital beds so more patients can get treatment faster, and helping to control healthcare costs that would otherwise be unsustainably increasing every year. Every time I tell a #BMT / #HCT expert about what this team is pulling off, they are surprised and impressed. It's an honor to have the opportunity to work alongside such pioneers, and to contribute in our own way to supporting the success of this paradigm shift. Next step: #RCT. Please share with anybody interested in innovations in BMT/HCT for people with blood/lymphatic cancers, and more generally, hospital-at-home models. StartUp Health Blueprint Health LLC Nex Cubed AgeTech Collaborative™ from AARP National PACE® Association Patti Weber Greg Evans Sharon Inouye Stephanie Rogers Sharon Wexler Springer Nature Group CalPACE MassPACE Association PACE Association of Michigan Matt Hudson Rick Robinson Amelia Hay Michael Hughes Caitlin Farrell Bill Lotter Gary Robinson Arthur Kuan Monica Schmiede Neelkanth Bardhan Nick Bott Rick Lee Thomas Osborne, MD San Agarawal Nurse Engineer Scientist(RN, comp sci, Ayurveddoc) Special thanks to National Cancer Institute (NCI) for supporting this project. #bonemarrowtransplant #BMT #hematopoieticstemcelltransplantation #HCT #AI #avatar #chatbot #humanintheloop #digitalhealth #cancer #lymphoma #leukemia #hospitalathome #healthsystem #healthcare

114

15 Comments

🧬 A research team from University of California, Los Angeles (UCLA) Health have developed a machine learning model capable of identifying undiagnosed immune disorders by analyzing EHR data, according to a recent study published in Science Translational Medicine. The researchers focused specifically on common variable immunodeficiency (CVID) disease, a set of rare immune disorders. The research team indicated that CVID is a common type of inborn error of immunity (IEI) – a group of approximately 500 inherited conditions that cause immune system dysfunction. CVID is estimated to impact one in every 25,000 people, and the condition is characterized by antibody deficiencies and impaired immune response. #HealthTech #MachineLearning #Immunodeficiency #HealthcareInnovation #MedicalTechnology #DataScience #HealthcareAI #DigitalHealth #InnovativeHealthcare #MedicalResearch #HealthcareTech #Bioinformatics #HealthData #ArtificialIntelligence

1

📢 New Episode of the Nucleate Podcast! Listen to the latest episode with Sean McClain as he explains the synergy between generative AI and the development of novel therapies. He also shares valuable personal insights and advice you won't want to miss! #ai #entrepreneurship #biotech

2

UCSF to Develop AI Monitoring Platform for Clinical Care Platform aims to bridge the gap between the rapid evolution of AI technologies and the need for robust, ongoing assessment of their efficacy, safety, and equity

1

#ICYMI: Last month we announced that we raised $132.5 million in an oversubscribed #SeriesC financing to advance our pipeline of potential therapies for fibrotic lung diseases and oncology. We are thankful for the support of our investors, board, #biotech community and the entire Endeavor BioMedicines team.     In this BioSpace article, get more details about what this financing means for the future of our pipeline and our goal to restore hope for patients with life-threatening diseases: https://bit.ly/4aVqAmo    #biotechnews #fundraising #venturecapital #oncology #lungdiseases

52

Using LLMs to Annotate Complex Cases of SDoH in Longitudinal Clinical Records. A recent study showcased the robust capabilities of Large Language Models, in addressing complex health challenges like housing instability among pregnant women. 🏠🤰 The study leveraged #AI to mine unstructured data from 25,217 clinical notes, revealing that AI could match—and sometimes exceed—human annotators in identifying housing instability. With a recall rate of 0.924, GPT-4 outperformed other methods, including manual annotation which had a recall of 0.702. However, the precision was slightly lower at 0.850 compared to human annotators at 0.971. This leap in technology signifies a major step forward in utilizing AI to understand and intervene in social determinants of health. By integrating LLMs, healthcare providers can achieve greater efficiency and scale in identifying patients at risk, potentially transforming patient outreach and care strategies. Will this integration impact the future of personalized healthcare and preventive strategies? 🔗 Link to the study in the comments below. 👇 #DigitalHealth #PrecisionMedicine 👉 Follow xCures Read our LinkedIn Newsletter: https://lnkd.in/dnNJV2ti https://xcures.com/ 👀

2

1 Comment

🚀 Breaking Ground in ER Decision-Making! A recent study by Icahn School of Medicine at Mount Sinai reveals how generative AI is poised to revolutionize emergency room admissions. Surpassing traditional models, this AI can predict hospital admissions with minimal data training, potentially transforming high-pressure decision environments. Dive into the findings published in the Journal of the American Medical Informatics Association and see how AI is set to enhance, not replace, the human touch in healthcare. 🏥🤖 #HealthTech #AIinHealthcare #EmergencyMedicine

1

1 Comment

Imagine AI solutions that enable physicians to return home to their families earlier and engage more meaningfully with their patients. In this episode of "The Future of AI in Health" podcast series, co-hosts Dr. Jenny Yu, MD FACS and Saul Marquez interview Dr. Shivdev Rao to discuss how AI technologies are revolutionizing the healthcare industry by enhancing the doctor-patient relationship and streamlining complex workflows while addressing future challenges in integrating AI into existing healthcare systems. Tune in and explore these transformative developments and envision the future of healthcare with AI. Click this link to listen to the entire episode and to check the resources: https://lnkd.in/g_Ufb8VT #outcomesrocket #network #healthlinemedia #healthcare #AI #future

8

1 Comment

🌟 BREAKTHROUGH DISCOVERY IN CANCER GENETICS 🌟 UCLA researchers have published a groundbreaking study in Nature Communications that revolutionizes our understanding of how noncoding DNA influences cancer progression. Their discovery of mutations in these "silent" regions significantly impacting mRNA levels, leading to abnormal protein production which can accelerate or hinder cell proliferation, challenges previous assumptions about noncoding DNA's role. The study also introduces a new prognostic tool, the untranslated tumor mutation burden (uTMB), which could predict patient survival in certain cancers. This pioneering research paves the way for developing targeted treatments and personalized medicine strategies, offering new hope in the fight against cancer. #CancerResearch #Genetics #InnovativeScience

16

So amazing to discuss how data and AI changes our (as doctors/pathologists) and eventually the patients' life. Multimodal data, LLM use in healthcare #Data_new_oil #LLM #LargeLanguageModel #ai #Data_in_patient_care

25

3 Comments

Weekly Tech+Bio Highlights: • Building Foundation Models for Cancer Biology: NOETIK is developing self-supervised world models that learn cancer biology from primary data to discover new therapeutic insights, co-founded by Ron Alfa, MD, PhD. • Meet nach0: a New General Purpose LLM for Chemical and Biomedical Applications: Insilico Medicine and NVIDIA introduced nach0, a large language model for biological and chemical tasks, detailed in the Chemical Science Journal. Alex Zhavoronkov, CEO of Insilico Medicine, highlights the model's capabilities in integrating diverse datasets for biological and chemical applications. • Owkin, co-founded by Thomas Clozel Unveils AI-Driven Oncology and Immunology Pipeline: Owkin has unveiled an AI-driven pipeline, featuring OKN4395, a dual EP2/4 antagonist, set for Phase 1 trials in 2025. • Novel Cell Analytics Technology: Cytomos will present its advanced cell analytics technology, Celledonia™, at ISCT 2024, enhancing cell and gene therapy processes. • Quantum Computing in Biopharma: - Key companies and initiatives; - Quantum advantage in biopharma; - Challenges and future outlook. • New Discovery Potentially to Re-write Biology Textbooks: Researchers have discovered a unique bacterial enzyme that creates new genes by reading RNA templates and converting them into DNA, challenging the traditional understanding of genetic information flow. For a more detailed overview and more, check out the full article: https://lnkd.in/dxPJ5MCe

23

The fusion of cutting-edge AI/ML technologies with a humanistic research approach is revolutionizing the pharma industry! By combining the power of advanced data analytics with a deep commitment to patient-centric care, we are accelerating the journey from lab discoveries to bedside treatments. At Bristol Myers Squibb, we believe that technology should amplify the human touch in healthcare. Through AI and ML, we can uncover insights that drive innovative treatments, while our dedicated researchers ensure these advancements resonate with the unique needs of every patient. Effective science communication is key to this process. By clearly conveying complex scientific concepts and AI/ML-driven breakthroughs, we empower patients, healthcare providers, and the broader community to understand and trust in the advancements being made. Together, we are making strides towards a future where science, technology, and communication go hand in hand. #BMS_Employee #PharmaInnovation #AIinHealthcare #PatientCentric #ScienceCommunication #MedicalResearch #FutureofHealthcare

2

🚀 Unleashing #AI's potential for rare disease treatment: Dr. David Fajgenbaum, MD, MBA, MSc reveals the innovative work of Every Cure, a pioneering nonprofit harnessing the vast realm of medical knowledge to unveil potential treatments for #RareDisease patients with artificial intelligence.   🎙️ During the UCSF Rare Disease Symposium, our President Dr. James Levine interviewed Dr. Fajgenbaum about leveraging AI algorithms for drug repurposing and evaluating the efficacy of every FDA-approved drug against every rare disease. Watch the full interview 👉 https://lnkd.in/egXRCkSe

5