Long Phi Le

Researchers are leveraging AI to sift through decades of animal testing data, minimizing the need for new tests. Joseph Manuppello of the Physicians Committee for Responsible Medicine highlights AI's efficiency in extracting and synthesizing this data. Prof. Thomas Hartung of The Johns Hopkins University notes that AI matches or surpasses human accuracy in information extraction. Projects like AnimalGAN and Virtual Second Species are paving the way for AI-driven toxicity testing.  https://lnkd.in/g6t7i-pS #AI #Pharmaceuticals #MedTech #DeepTech #IP #VC #Patents

3

Join us on June 5, 2024, at 1 PM (ET) for a webinar on tackling the reproducibility and generalizability challenges in #MachineLearning ! Learn strategies for data accessibility, leakage prevention, preprocessing, algorithm selection, and more to ensure reliable ML outcomes. Farhad Maleki Society for Imaging Informatics in Medicine (SIIM) Register: ⬇ https://lnkd.in/dqrDUg2J Objectives: - Equip participants with strategies to facilitate data accessibility, prevent data leakage, and standardize data preprocessing to enhance the reproducibility of machine learning projects. - Provide insights into selecting appropriate algorithms, promoting code sharing, and managing dependencies effectively to achieve reliable machine learning outcomes. - Advocate for the adoption of comprehensive and transparent research documentation standards to improve the clarity and reproducibility of machine learning research. - Demonstrate how to utilize advanced tools and frameworks and adhere to community standards to significantly improve the reproducibility of machine learning systems. #ML #DataScience #Healthcare #Webinar #AI

14

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

🚀 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

Good Day, Medicine AI in Medicine Learning Objectives: Develop a framework about the history of Artificial Intelligence (AI) Provide an overview of types of AI in medicine Review current AI technologies in medicine Highlight important factors for the future of AI Evaluate and define ethical aspects in AI https://buff.ly/47iokDC

1

Dr. Shelby on the private capital dynamics for financing drug discovery today. Her newsletter is awesome if you aren't subscribed! (link below) "Ultimately, drug discovery is the deepest of deep techs, akin to space tech, in terms of capital requirement and timescales. However, the difference in making sick people healthy doesn’t quite capture the imagination of society (except in edge cases like pandemics). This means the pools of capital are smaller, and more traditional, with AI in drug discovery companies walking the line of traditional biotech in terms of data packages while selling a story of potential for outsized returns based on many shots on goal. The only problem is, the source of continuous, growth stage capital is much more tied to traditional biotech investors which caps the vision. We believe this is the short-term problem for AI in drug discovery and are still actively looking for investments in this field. As soon as one company shows increased clinical success, there will be another moment of everyone piling on again. Founders building in this space need to keep their heads down acknowledging that this is part of the innovation cycle which they can eventually capitalize upon if their companies stay alive." https://lnkd.in/e4u7DuJ3? #biotech #AI #VC #deeptech

28

3 Comments

Generative artificial intelligence is infiltrating peer review process 🤖 Navigating the New Frontiers: Generative AI in the Peer Review Process As the influx of manuscript submissions grows, the peer review process in scientific research faces increasing challenges. Recent advancements have seen generative AI tools like #ChatGPT being tested to support this crucial aspect of publication. However, their integration raises several ethical and practical concerns, including confidentiality and the preservation of critical human judgment. This recent study highlighted in Critical Care explores these dynamics, demonstrating both the potential efficiencies brought by AI and the complexities it introduces. What are your thoughts on this development? Can AI truly complement the human element in peer review without compromising the integrity and thoroughness required in scholarly publishing? #PeerReview #ArtificialIntelligence #ScientificPublishing #EthicsInAI Artificial IntelliGence in Anestesia e Terapia IntensivA Elena Giovanna Bignami Marco Cascella Jonathan Montomoli Valentina Bellini

1

In an analysis of 123 payers, MMIT’s Dinesh Kannan Kabaleeswaran and Julie Mohammed, MPH identify key trends in disease management and discuss future trends, including value-based contracting and biosimilar utilization. Read the paper published in the International Journal of Science and Research (IJSR): https://ow.ly/c5cp50SlhF2 #Healthcare #Biosimilars #PayerManagement

13

AI MODELS LEVERAGING ROUTINELY COLLECTED CLINICAL DATA TO PREDICT PATIENT RESPONSE TO IMMUNOTHERAPY ⬇ This study explores the use of machine learning to predict the effectiveness and toxicity of immune checkpoint inhibitors (ICIs) using real-world electronic health record (EHR) data. The focus is on the prediction of hepatitis, colitis, pneumonitis, and 1-year overall survival post-ICI initiation. 1️⃣ A machine learning model using routine EHR data successfully predicts serious immunotoxicities and 1-year survival among patients treated with ICIs. 2️⃣ The study analyzed data from over 2,200 patients, creating predictive models with AUC values ranging from 0.729 to 0.755, demonstrating reasonably strong performance. 3️⃣ Detailed feature engineering was conducted, including the aggregation of laboratory measurements over specific time windows prior to ICI initiation, enabling the models to capture relevant changes in patient health status that predict outcomes. 4️⃣ The machine learning models utilize random forest classifiers, chosen for their robust handling of complex interactions between features, ensuring effective predictions of toxicity and survival outcomes. 5️⃣ Features used in the models include structured EHR data such as laboratory results, demographics, and medication history, without requiring additional data collection. 6️⃣ The predictive accuracy of the models varies by cancer type—melanoma, lung cancer, and genitourinary cancer—highlighting the potential need for cancer-specific models to enhance predictive precision. 7️⃣ Each model relies on an outcome-specific feature set, tailored to optimally predict specific outcomes like pneumonitis, hepatitis, or colitis, emphasizing the personalized approach of the study. ✍🏻 Levente Lippenszky, Kathleen F. Mittendorf, PhD, Zoltán Kiss, Michele Laura LeNoue-Newton, Pablo N., Protiva Rahman, Cheng Ye, Balázs Laczi, Eszter Csernai, Neha M. Jain, Marilyn (Holt) Hammer, PhD, Christina N. Maxwell, Madeleine B., Yufang Ma, Margaret Mitchell, MD, MS-HPEd, Douglas Johnson, David Smith, Ben Park, Christine Micheel, Daniel Fabbri, Jan Wolber, Travis Osterman. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clin Cancer Inform 8, e2300207(2024). DOI:10.1200/CCI.23.00207 💥 Join our community of 30.000 curious minds! ✅ Subscribe to our newsletter for your front-row seat to the latest groundbreaking studies. Get started here: https://lnkd.in/eR7qichj.

80

1 Comment

Vulnerability of White Matter Tracts in Alzheimer's Disease Alzheimer's disease (AD) disrupts neural communication by targeting specific white matter tracts (WMT). These fiber bundles act as highways, connecting various brain regions and enabling crucial information flow. Understanding the susceptibility and potential timeline of degeneration in these tracts can offer valuable insights into AD progression. Cascade of Disruption Research suggests a sequential vulnerability of WMTs in AD, here is a potential order of their degeneration: Fornix/Cingulum: key pathway connecting the hippocampus to other brain areas is often among the first to show damage. Frontal and Parietal White Matter: WMTs connect the frontal and parietal lobes, crucial for cognitive abilities that decline in AD. Transcallosal Prefrontal Tracts & Arcuate Fasciculus: Degeneration in these WMTs might occur around the same time, affecting executive function and potentially language abilities. Superior Longitudinal Fasciculus (SLF): This extensive tract's degeneration might occur later, contributing to the broader cognitive and motor decline seen in advanced AD. Diffusion Tensor Imaging (DTI) in AD Trials While this proposed order provides a framework, the precise timing can vary significantly between patients. This inherent heterogeneity presents a major challenge in designing effective clinical trials for AD therapies. Here's where Diffusion Tensor Imaging (DTI) emerges as a powerful tool: DTI is a non-invasive MRI technique that measures the diffusion of water molecules within tissues, providing insights into white matter microstructure. By incorporating DTI markers of white matter integrity into patient selection criteria for AD trials, researchers can achieve greater sample homogeneity. This reduces variability due to underlying white matter pathology, leading to more robust study results. Increased sample homogeneity translates to a higher probability of sufficient statistical power. In simpler terms, DTI helps ensure trials have the necessary number of participants to detect a true treatment effect, ultimately accelerating the path to effective AD therapies. Integrating DTI into AD clinical trials holds immense promise for improving sample quality, enhancing statistical power, and ultimately, accelerating the discovery of effective treatments for this devastating disease. At IAG, IMAGE ANALYSIS GROUP we understand the complexity of Dementias, and other forms of neurodegeneration. Reach out to me or imaging.experts@ia-grp.com to see how our neuroscience expertise can boost your clinical trial plans through medical imaging. #Alzheimers #ClinicalTrials #DTI #WhiteMatter

19

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

Machine learning analysis of retrosp EHR data (3/20-9/22) on 2+ million showed COVID (5.3%) outcomes in #IMID pts not linked to DMARD or IMID Dx; but rather to age, unvaccinated, & comorbidities (Afib, CAD, CHF, CKD, COPD, CA, liver Dz)

1

A recently published manuscript in JCO Clinical Cancer Informatics highlights how Tempus is offering providers the benefit of incorporating multimodal data, alongside genomic information, directly into the EHR to enhance treatment decision-making. Read the full article here: https://bit.ly/3QL9fo9

103

5 Comments

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

#𝐒𝐏𝐀𝐑𝐑𝐎𝐖: 𝐀𝐈 𝐡𝐞𝐥𝐩𝐬 𝐜𝐡𝐞𝐦𝐢𝐬𝐭𝐬 𝐟𝐢𝐧𝐝 𝐜𝐨𝐬𝐭-𝐞𝐟𝐟𝐞𝐜𝐭𝐢𝐯𝐞 𝐦𝐨𝐥𝐞𝐜𝐮𝐥𝐞𝐬 𝐟𝐨𝐫 𝐧𝐞𝐰 𝐝𝐫𝐮𝐠𝐬 Researchers at MIT have developed an AI framework named SPARROW, which streamlines the drug development process by evaluating potential drug molecules and their synthesis pathways. The Synthesis Planning and Rewards-based Route Optimization Workflow (SPARROW) aims to minimize synthetic costs while ensuring the molecules have the desired properties. https://lnkd.in/eszsMYNX

4