Roy Perlis

Dr. Tanaka and colleagues have published an interesting article in our journal: "Clarifying Causal Effects of Interest and Underlying Assumptions in Randomized and Nonrandomized Clinical Trials in Oncology Using Directed Acyclic Graphs and Single-World Intervention Graphs" https://lnkd.in/etTmTqJ5 Check it out!

16

1 Comment

🚨 Recommended article 🚨 Swarnali Goswami of the Journal of Medical Economics Early Career Panel has selected "𝐕𝐚𝐥𝐮𝐞 𝐨𝐟 𝐧𝐞𝐱𝐭 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐬𝐞𝐪𝐮𝐞𝐧𝐜𝐢𝐧𝐠 (𝐍𝐆𝐒) 𝐭𝐞𝐬𝐭𝐢𝐧𝐠 𝐢𝐧 𝐚𝐝𝐯𝐚𝐧𝐜𝐞𝐝 𝐜𝐚𝐧𝐜𝐞𝐫 𝐩𝐚𝐭𝐢𝐞𝐧𝐭𝐬" as her recommended read, to find out why, see her post below. Big thanks to the authors for publishing their work in JME. Jesse Ortendahl, Gebra Cuyun Carter, Snehal Thakkar, Katalin Bognar, David Hall, Yara Abdou #healtheconomics #oncology #pharmacoeconomics #cancer #medicaleconomics #HEOR

24

1 Comment

Our second session covers: Addressing the Gap Between Clinical Trials and Implementation into Clinical Practice and Population Health. Nandini Raghuraman from WashU OBGYN presented "De-Implementation of Low-Value Care". The panel was moderated by Bill Powderly, MD, the Associate Dean for Clinical and Translational Science. Learn more about how the ICTS can support your clinical and translational research> www.icts.wustl.edu

1

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

AI meets Rheumatology: ChatGPT and patient response optimization Large language models like ChatGPT, trained on vast text data, are revolutionizing healthcare by understanding and generating human-like language.

1

After her interesting (and fun!) presentation at Stanford University entitled “Harnessing AI to Enrich Patient Care”, Yaa Kumah-Crystal, MD MPH FAMIA and I discussed whether MDs/NPs are going to be more liable for using AI to make decisions (and making a mistake) or NOT using AI to make decisions (and making a mistake). Thoughts? Stanford Department of Medicine Functional Medicine Research and Technology Center #AI #Healthcare #Ethics #DecisionMaking Grayson Dimick

2

Dr Eric J. Feldman, director at Montefiore Einstein Comprehensive Cancer Center, explains how they are bringing early phase clinical trials to their community. Learn more about this process and how he is addressing these challenges in our interview below. Read the article here: https://lnkd.in/eT8yJNTB #ClinicalResearch #ClinicalTrials #Trials

11

‘Harvard Thinking’: New frontiers in cancer care In podcast, experts discuss breakthroughs in treatment, from genomic sequencing to AI, and how close we are to personalized vaccines https://lnkd.in/eUQHpm9U

19

1 Comment

#aidrugdiscovery, #Insilico Medicine, #Alex Zhavoronkov Is Generative AI in Drug Discovery Overhyped? https://lnkd.in/dnFTcG3H ...I came up with a list of simple questions to help evaluate the companies in this reasonably mature field of AI-powered drug discovery and estimate their hype-to-results ratios: * How many of your internally discovered drugs are in Phase I, Phase II, and Phase III? * Did any of your pharma partnerships progress into Phase I, II, and III, and how long did it take? * How many preclinical candidates (PCCs) did you nominate in a single year and in total? * How long, on average, does it take you to nominate a PCC, and how much does it cost? * What was the level of novelty of the targets and what is the level of novelty of the molecules? * How long, on average, does it take you to go into clinical trials with an AI-discovered target? * Did you in-license any of your drugs from other companies, repurpose them, or discover them internally? * How many pharma companies use your AI software to discover drugs, and how many users do you have? How many renew every year? * What was your revenue last year, and what was the revenue growth year over year? * How much money have you raised, in how many rounds, and what is the valuation of your last round?

1

Numerous studies conducted by Dana-Farber researchers show promising results for patients with endometrial, lung, breast, prostate, and colorectal cancers. The results of these studies, along with dozens of others led by Dana-Farber faculty, were presented at this year’s American Association for Cancer Research (AACR) Annual Meeting last month in California.   Notable research from Dana-Farber researchers included: Novel ADC and Immunotherapy Combo Shows Promise in Endometrial Cancer Subtype  Rebecca Porter, MD, PhD, a medical oncologist in the Gynecological Oncology Program of the Susan F. Smith Center for Women’s Cancers, presented results of a small, investigator-initiated phase 2 study of a novel combination of an antibody-drug conjugate and an immune checkpoint inhibitor. Blood Test May Identify Patients with Non-Small Cell Lung Cancer Subtype Who Can Benefit from Combination Therapy  Pasi Jänne, MD, PhD, senior vice president for Translational Medicine, presented an update to his practice-changing FLAURA2 clinical trial published last year, revealing that patients with non-small cell lung cancer (NSCLC) marked by a mutation in the EGFR gene fared significantly better when treated with the drug osimertinib plus platinum-based chemotherapy than with osimertinib alone. Exercise Program Can Improve Heart Function in Cancer Survivors  Cami Christopher, MPH, a research scientist in the Department of Medical Oncology, Population Sciences, presented findings suggesting that a moderate to vigorous aerobic and resistance exercise program can improve a key aspect of heart function in survivors of breast, prostate, and colorectal cancer. Study Provides First Large-Scale Look at Genetic Architecture of Prostate Cancer in Sub-Saharan Africa  Timothy Rebbeck, PhD, the Vincent L. Gregory Professor of Cancer Prevention, led the first large-scale genomic analysis of prostate cancer in sub-Saharan Africa which revealed that the genetic architecture of the disease – its basic underpinnings in abnormalities in DNA – is unique to Africa and varies across the continent. In addition, the AACR recognized Rebbeck, Christopher D.M. Fletcher, MD, FRCPath, and Gordon J. Freeman, PhD, with 2024 Scientific Achievement Awards during the Annual Meeting.  More: https://lnkd.in/etk8fwUp

58

1 Comment

Structure and Function of Somatostatin and Its Receptors in Endocrinology Zhang et al.: “In this review, we discuss the recent findings related to the structural analysis of somatostatin receptors and somatostatin receptor ligands, the relationships between the structural data and clinical findings, and the future development of novel structure-based therapies” Open access in Endocrine Society’s Endocrine Reviews: https://lnkd.in/dWMZzRhP

31

1 Comment

#PatientsUseAI Here is a great simple example of how LLMs can help laypersons understand a fundamental, and yet poorly understood concept in medicine and statistics (sensitivity vs specificity): Me: "Explain: Snoring has a sensitivity of 80 to 90 percent for diagnosing OSA, but its specificity is below 50 percent." ChatGPT: **Sensitivity and Specificity in Diagnosing OSA with Snoring** 𝐒𝐞𝐧𝐬𝐢𝐭𝐢𝐯𝐢𝐭𝐲: - 𝗗𝗲𝗳𝗶𝗻𝗶𝘁𝗶𝗼𝗻: Sensitivity measures how often a test correctly identifies a condition. In this case, it refers to the percentage of people with obstructive sleep apnea (OSA) who snore. - 𝗛𝗶𝗴𝗵 𝗦𝗲𝗻𝘀𝗶𝘁𝗶𝘃𝗶𝘁𝘆 (𝟴𝟬-𝟵𝟬%): Snoring is highly sensitive for diagnosing OSA because 80-90% of people who have OSA also snore. This means if someone has OSA, there's an 80-90% chance they will snore. Hence, snoring is a common symptom among those with OSA. 𝐒𝐩𝐞𝐜𝐢𝐟𝐢𝐜𝐢𝐭𝐲: - 𝗗𝗲𝗳𝗶𝗻𝗶𝘁𝗶𝗼𝗻: Specificity measures how often a test correctly identifies people who do not have a condition. Here, it indicates the percentage of people without OSA who do not snore. - 𝗟𝗼𝘄 𝗦𝗽𝗲𝗰𝗶𝗳𝗶𝗰𝗶𝘁𝘆 (𝗯𝗲𝗹𝗼𝘄 𝟱𝟬%): Snoring has low specificity for diagnosing OSA because less than 50% of people who snore actually have OSA. This means many people who snore do not have OSA, and thus, snoring alone is not a reliable indicator of the condition. 𝐈𝐦𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬: - 𝗛𝗶𝗴𝗵 𝗦𝗲𝗻𝘀𝗶𝘁𝗶𝘃𝗶𝘁𝘆: Snoring is a good initial indicator to suggest further testing for OSA. If a patient snores, there's a high chance they could have OSA, warranting additional diagnostic tests. - 𝗟𝗼𝘄 𝗦𝗽𝗲𝗰𝗶𝗳𝗶𝗰𝗶𝘁𝘆: Snoring alone should not be used to diagnose OSA. Many people snore without having OSA, so other symptoms and tests must be considered to confirm the diagnosis. In summary, while snoring is common in individuals with OSA (𝐡𝐢𝐠𝐡 𝐬𝐞𝐧𝐬𝐢𝐭𝐢𝐯𝐢𝐭𝐲), not everyone who snores has OSA (𝐥𝐨𝐰 𝐬𝐩𝐞𝐜𝐢𝐟𝐢𝐜𝐢𝐭𝐲). Therefore, snoring is a useful sign for identifying potential OSA but not definitive without further testing.

4

10 Comments

The AMIA Natural Language Processing Working Group will feature Dr. Yonghui Wu, the inventor of the largest clinical LLM GatorTron, with the talk titled “Prevails and Mirages of Large Language Models in Clinical NLP” on Tuesday August 13, 2024 at noon EST. Here’s the abstract: Large language models (LLMs) have profoundly changed clinical natural language processing (NLP) and brought substantial opportunities and challenges for artificial intelligence (AI) in healthcare. LLMs almost become a unified solution for clinical NLP, yet current findings are mixed with prevails and mirages. On one side, LLMs have been reported to be able to pass several professional exams, yet there is limited evidence supporting the benefit of LLMs to many classic clinical NLP tasks. This talk will introduce the technologies powering LLMs, overview the recent prevails, and examine the mirages in the hype of LLM magic. Based on the experience in developing two clinical LLMs in the clinical domain, including GatorTron and GatorTronGPT, this talk will provide insight into the potential application of LLMs for clinical NLP and healthcare. Please register at https://lnkd.in/e26rtx8Y

44

Read full paper in #Med

In this new video, Laura Dormer, Editor of The Evidence Base®, speaks with Qianyi Zhang, Senior Research Scientist, Flatiron Health about the poster, “Evaluation of real-world response rate in clinical trial-aligned cohorts of patients with lung, colorectal, and breast cancer using machine learning,” presented at ISPOR—The Professional Society for Health Economics and Outcomes Research 2024 (May 6–8, 2024; Atlanta, GA, USA). #heor #healtheconomics #outcomesresearch #rwe #rwd #realworldevidence #realworlddata #healthdata #AI #artificialintelligence #machinelearning #lungcancer #breastcancer #colorectalcancer #oncology #clinicaltrials

11

📢 New Review Published Online 📢 📚 CPPD—differential diagnostics and differential therapeutic challenges ✒ Authors: Maria L. Voulgari, Herbert Kellner ✅ This review provides a comprehensive overview of CPPD, focusing on its diagnosis, differential diagnosis, therapeutic challenges, and monitoring, with insights into the association between CPPD and cardiovascular risk. Diagnostics in CPPD rely on identifying CPP crystals in synovial fluid or joint tissues, with imaging modalities such as ultrasound and conventional radiography emerging as valuable tools...... 💡 Keywords: #CPPD, calcium pyrophosphate crystal deposition, #chondrocalcinosis 🔗 Read the full article: https://lnkd.in/gKRg4RWD 📑 PDF: https://lnkd.in/gqKS85Gd 🔎 This article belongs to our special issue “Calcium Pyrophosphate Deposition Disease”, edited by Prof. Jürgen Braun. (https://lnkd.in/gEe-T_9b)

3

Pleased to share new research out of Trilliant Health using the novel health plan price transparency data via JAMA Health Forum. The observed variation in negotiated rates both within and between payers is staggering. Same hospital and different insurance company, you can pay 2x for the same procedure. Same insurance company and different hospital in the same market, you can pay 2x for the same procedure. Add this one to the "it's the prices, stupid" pile.

50

2 Comments