Raviv Pryluk’s Post

We're pleased to introduce our novel Causal Responders Detection (CARD) method, detailed in our latest publication. CARD applies machine learning and causal tree techniques to identify individuals who benefit significantly from treatments, ensuring false discovery control in both RCTs and observational studies. This method refines our understanding of individual treatment responses, providing crucial insights for precision medicine development. Our simulations confirm CARD's effectiveness across various settings, enabling to apply this method across therapeutic areas and clinical phases. #PersonalizedMedicine #MachineLearning #ClinicalTrials #HealthcareInnovation Tzviel Frostig Oshri Machluf Elad Berkman Raviv Pryluk Amitay Kamber

Application of causal inference methods in individual-participant data meta-analyses in medicine: addressing data handling and reporting gaps with new proposed reporting guidelines https://lnkd.in/d4fwscss Observational data provide invaluable real-world information in medicine, but certain methodological considerations are required to derive causal estimates. In this systematic review, we evaluated the methodol...

The concept of semantic knowledge which alludes to long term memory and knowledge achieved by studying a complex topic eg. medicine. A framework to organize the enormous heterogeneous medical data as a semantic health knowledge graph by integrating conceptual and theoretical graphs with health data. The GraphRAG can enhance the accuracy and they have the potential to transform patient care and clinical research. #knowledgegraph #generativeai #digitalhealth #graphrag

An overlooked benefit of AI scribes: standardizing and enriching clinical documentation for research purposes. Today, patient stories are filtered heavily through the lens of the physician and are further warped by billing codes. Depending on the medical specialty, pertinent positive or negative findings may not be documented, especially if they are insignificant for billing purposes. Omitting this information has tremendous downstream implications for research on disease phenotyping, progression, and treatment, as well as clinical trial recruiting. If architected thoughtfully, AI scribes can preserve the integrity of the patient journey, producing standardized and consistent documentation to support breakthrough research across disease classification, diagnostics, natural history, and therapeutic development.

🔬✨ Can you imagine being able to personalise your medical treatment according to your DNA and lifestyle? With precision medicine, this is a reality. ⬇ Using #BigData, #ArtificialIntelligence and genomic data, we are improving medical outcomes and preventing diseases before they occur. 🤔 How? We tell you about it 👉🏻 https://lnkd.in/dMx2Xw3i

Estimating individualized treatment effects using an individual participant data meta-analysis https://lnkd.in/dkjG77iV One key aspect of personalized medicine is to identify individuals who benefit from an intervention. Some approaches have been developed to estimate individualized treatment effects (ITE) with a single randomi...

Application of #CausalInference methods in individual-participant data meta-analyses in medicine: Addressing data handling and reporting gaps with new proposed reporting guidelines https://lnkd.in/e4-P8s-A Richard Riley #IPDMA #MetaAnalysis #IPD

TileDB’s Jeremy Leipzig penned an article for HIT Consultant Media on precision medicine and the database it requires. Jeremy explains: “The future of precision medicine – where all types of data are housed and analyzed in aggregate; at scale; and safely shared among a growing number of constituents in order to cure and diagnose diseases through faster, more comprehensive insights – requires a completely new type of database. Ideally, it will be one that can handle multiple data modalities; is fast, unified and universally shareable.” Read the entire article here. https://lnkd.in/eNhqp75F #TileDB #PrecisionMedicine

A team of ADC Consultants, in collaboration with RadboudUMC hospital, have published a research paper on model-informed precision dosing using machine learning in patients with an underactive thyroid gland (Hashimoto’s thyroiditis). This study addresses the challenges patients and medical professionals face in finding the right dosage of Levothyroxine, a synthetic thyroid hormone. By leveraging patient characteristics such as blood values, weight, and age, our model predicts the optimal dose, potentially saving time and improving patient outcomes. The study included a simulation trial involving general practitioners, who responded positively to the use of the model. This demonstrates its clinical relevance, safety, and effectiveness in primary care settings. Read the full study here: https://lnkd.in/ep5wRe4f You can also find a summary of the early stages of the project on the ADC website: https://lnkd.in/e_C_kHmV #precisionmedicine #autoimmunedisease #thyroidhealth

Do you RISE To the challenge of cutting treatment delays in STEMI patients?? Check out this Prospective RCT of AI-Enhanced ECG for STEMI Activation NEJM AI: https://lnkd.in/ej3b_D_5 Our editorial written by Bill from Stanford University School of Medicine and I explores the implications of this great study of real-world AI implementation: https://lnkd.in/ex6vVxkK * AI-ECG reduced median door-to-balloon time by 14 minutes * ECG-to-balloon time decreased by 5.6 minutes for hospitalized patients * High predictive value with fewer unnecessary STEMI activations Limitations: * Single-center design * Short follow-up * No evaluation of care appropriateness or clinical safety endpoints