Evidence for an anti-AI bias for getting medical advice nature.com/articles/s4159… Nature Medicine Moritz Reis

In two preregistered studies involving 2,280 participants using simulated clinical scenarios, advice labelled as involving AI was evaluated as significantly less reliable and less empathetic compared to “human”-labelled advice Moritz Reis Universität Würzburg #UniWürzburg nature.com/articles/s4159…

In two preregistered studies involving 2,280 participants using simulated clinical scenarios, advice labelled as involving AI was evaluated as significantly less reliable and less empathetic compared to “human”-labelled advice Moritz Reis Universität Würzburg #UniWürzburg nature.com/articles/s4159…

#LLMs show promise in controlled settings, but a new study exposes their limitations in real-world clinical settings and points towards robust evaluation & benchmarking before clinical use. News & Views from Nigam Shah and colleagues Nigam Shah Stanford Medicine nature.com/articles/s4159…

Hager et al. assessed LLM performance in clinical decision-making (nature.com/articles/s4159…). In Nature Medicine, Suhana Bedi, Nigam Shah, and I discuss the main takeaways from this excellent paper and share a framework for evaluating LLMs for clinical use (nature.com/articles/s4159…)

Anxiety, depression, headaches — is political polarization bad for your health? News feature by anita makri nature.com/articles/s4159…

An open source and computing-friendly vision-language model is shown to achieve state-of-the-art accuracy in 16 out of 25 biomedical tasks, with promising performance in a series of potential clinical applications Hongfang Liu, PhD Lichao Sun James Zou nature.com/articles/s4159…

This is a cool study from PathAI in Nature Medicine AI-based automation of enrollment criteria and endpoint assessment in clinical trials in liver diseases nature.com/articles/s4159… The combination of digital pathology (I.e. remote accessibility from anywhere) & #AI assisted

An open source and computing-friendly vision-language model is shown to achieve state-of-the-art accuracy in 16 out of 25 biomedical tasks, with promising performance in a series of potential clinical applications Lichao Sun James Zou Hongfang Liu, PhD nature.com/articles/s4159…

Our paper in npj Digital Medicine uncovers major geo-economic and gender disparities in AI-enabled clinical studies. We need diverse research teams and global collaboration! 🌏💡 #AI #DigitalMedicine #HealthEquity #GlobalHealth #DukeNUS nature.com/articles/s4174…

A randomised controlled #trial evaluating the selection of a single #blastocyst for transfer by deep learning did not demonstrate non-inferiority in clinical pregnancy rates when compared to the trained embryologists using standard morphology criteria. nature.com/articles/s4159…

The 5 different patterns of brain aging from ~50,000 individuals, with their relationship to lifestyle, genetics and outcomes, derived from MRI and #AI. nature.com/articles/s4159… Nature Medicine Christos Davatzikos and colleagues

An implementation trial in Rwanda found that a digital school-based intervention did not affect the primary outcomes of modern contraceptive use, childbearing and HIV testing but was associated with higher contraceptive use in sexually active adolescents nature.com/articles/s4159…

A modelling study quantifies the health-economic burden of Lassa virus infection across West Africa and projects impacts of a series of reactive and preventive #vaccination campaigns against the disease. University of Oxford David Smith Health Economics Research Centre (HERC) nature.com/articles/s4159…

Presentation and publications #latebreaking The strategy was effective to 📈GDMT  #ESCCongress European Society of Cardiology @Rafavidalperez Nico Bruining EiC EHJ Digital Health HFAPresident @nature Wilfried Mullens Jozine ter Maaten Henry Han Gregg Fonarow MD Steve Greene Harriette Van Spall, MD MPH Victoria Delgado Charles Fauvel Novi Tham

Happy that our ehrapy toolbox for the exploratory analysis of electronic health records is out today Nature Medicine! It enables early QC&imputation at scale, visualization and downstream clustering & patient trajectory learning. nature.com/articles/s4159… github.com/theislab/ehrapy

In this pragmatic, randomized clinical trial from 6 hospitals in Nigeria, AI-based ECG screening proved accurate in detecting peripartum cardiomyopathies and suggests that it could improve detection of these conditions Demi Adedinsewo, MD Nature Medicine nature.com/articles/s4159…

A modelling study quantifies the health-economic burden of Lassa virus infection across West Africa and projects impacts of vaccination campaigns, presenting significant impacts in terms of averted DALYs and healthcare cost reductions DeirdreHollingsworth nature.com/articles/s4159…

New Science Magazine How to reduce belief in conspiracy theories? Generative #AI !! —Original research showing ~20% decrease, lasting at least 2 months, even among people whose beliefs are entrenched science.org/doi/10.1126/sc… —Perspective science.org/doi/10.1126/sc… —Editorial

Going from data extraction and quality control to the generation of low-dimensional representations and to longitudinal analyses, an open source software is proposed to standardize current EHR data processing and analysis pipelines Fabian Theis nature.com/articles/s4159…