If something so simple can transform intensive care, what else can it do?
Use of Deep Learning for Continuous Prediction of Mortality for All Admissions in Intensive Care Units
<p>The mortality rate in the intensive care unit (ICU) is a key metric of hospital clinical quality. To enhance hospital performance, many methods have been proposed for the stratification of patients’ different risk categories, such as severity scoring systems and machine learning models. However, these methods make capturing time sequence information difficult, posing challenges to the continuous assessment of a patient’s severity during their hospital stay. Therefore, we built a predictive model that can make predictions throughout the patient’s stay and obtain the patient’s risk of death in real time. Our proposed model performed much better than other machine learning methods, including logistic regression, random forest, and XGBoost, in a full set of performance evaluation processes. Thus, the proposed model can support physicians’ decisions by allowing them to pay more attention to high-risk patients and anticipate potential complications to reduce ICU mortality.</p>
A startup is pitching a mind-uploading service that is “100 percent fatal”
Nectome will preserve your brain, but you have to be euthanized first.
ChatGPT Health and what AI can do for a broken system
Healthcare isn’t working for patients or doctors, but AI tools can help.
‘Unbelievably dangerous’: experts sound alarm after ChatGPT Health fails to recognise medical emergencies
Study finds ChatGPT Health did not recommend a hospital visit when medically necessary in more than half of cases
ChatGPT Health performance in a structured test of triage recommendations
ChatGPT Health was launched in January 2026 as OpenAI’s consumer health tool and has reached millions of users. Here we conducted a structured stress test of triage recommendations using 60 clinician-authored vignettes across 21 clinical domains under 16 factorial conditions, yielding 960 total responses. Performance followed an inverted U-shaped pattern, with the most dangerous failures concentrated at clinical extremes—nonurgent presentations (35%) and emergency conditions (48%). Among gold-standard emergencies, the system undertriaged 52% of cases, directing patients with diabetic ketoacidosis or impending respiratory failure to 24–48 h evaluation rather than the emergency department, while correctly triaging classical emergencies such as stroke and anaphylaxis. When family or friends minimized symptoms, indicating anchoring bias, triage recommendations shifted significantly in edge cases (odds ratio = 11.7, 95% confidence interval = 3.7–36.6), with the majority of shifts toward less urgent care. Crisis-intervention messages activated unpredictably across suicidal ideation presentations, occurring more frequently when patients described no specific method than when they did. Patient race, sex and barriers to care did not show significant effects, although confidence intervals did not exclude clinically meaningful differences. These findings reveal missed high-risk emergencies and inconsistent activation of crisis safeguards, raising safety concerns that warrant prospective validation before consumer-scale deployment of artificial intelligence triage systems.
ChatGPT Health performance in a structured test of triage recommendations
ChatGPT Health was launched in January 2026 as OpenAI’s consumer health tool and has reached millions of users. Here we conducted a structured stress test of triage recommendations using 60 clinician-authored vignettes across 21 clinical domains under 16 factorial conditions, yielding 960 total responses. Performance followed an inverted U-shaped pattern, with the most dangerous failures concentrated at clinical extremes—nonurgent presentations (35%) and emergency conditions (48%). Among gold-standard emergencies, the system undertriaged 52% of cases, directing patients with diabetic ketoacidosis or impending respiratory failure to 24–48 h evaluation rather than the emergency department, while correctly triaging classical emergencies such as stroke and anaphylaxis. When family or friends minimized symptoms, indicating anchoring bias, triage recommendations shifted significantly in edge cases (odds ratio = 11.7, 95% confidence interval = 3.7–36.6), with the majority of shifts toward less urgent care. Crisis-intervention messages activated unpredictably across suicidal ideation presentations, occurring more frequently when patients described no specific method than when they did. Patient race, sex and barriers to care did not show significant effects, although confidence intervals did not exclude clinically meaningful differences. These findings reveal missed high-risk emergencies and inconsistent activation of crisis safeguards, raising safety concerns that warrant prospective validation before consumer-scale deployment of artificial intelligence triage systems.
'Critical' ramping leaves Bendigo and Ballarat without ambulance coverage
Patients in a life and death situation won't survive, union says.
About Us - 9/11 Health Watch
Why do we need a 911 Health Watch? Even with the passage of the James Zadroga 9/11 Health and Compensation Act in 2010, and the law’s reauthorization in 2015, 2019, and the 2026 full funding of the World Trade Center Health Program, there is still much that needs to be done to ensure the Federal […]
Preparing Physicians for the Clinical Algorithm Era
The U.S. government recently took steps to ensure that clinical decision support algorithms are safe for clinical use. The next and larger step will be teaching physicians how to use the algorithms ...
Preparing Physicians for the Clinical Algorithm Era
The U.S. government recently took steps to ensure that clinical decision support algorithms are safe for clinical use. The next and larger step will be teaching physicians how to use the algorithms ...
Implementation of Digital Monitoring Services During the COVID-19 Pandemic for Patients With Chronic Diseases: Design Science Approach
Background: The COVID-19 pandemic is straining health systems and disrupting the delivery of health care services, in particular, for older adults and people with chronic conditions, who are particularly vulnerable to COVID-19 infection. Objective: The aim of this project was to support primary health care provision with a digital health platform that will allow primary care physicians and nurses to remotely manage the care of patients with chronic diseases or COVID-19 infections. Methods: For the rapid design and implementation of a digital platform to support primary health care services, we followed the Design Science implementation framework: (1) problem identification and motivation, (2) definition of the objectives aligned with goal-oriented care, (3) artefact design and development based on Scrum, (4) solution demonstration, (5) evaluation, and (6) communication. Results: The digital platform was developed for the specific objectives of the project and successfully piloted in 3 primary health care centers in the Lisbon Health Region. Health professionals (n=53) were able to remotely manage their first patients safely and thoroughly, with high degrees of satisfaction. Conclusions: Although still in the first steps of implementation, its positive uptake, by both health care providers and patients, is a promising result. There were several limitations including the low number of participating health care units. Further research is planned to deploy the platform to many more primary health care centers and evaluate the impact on patient’s health related outcomes.
How we can stop antibiotic resistance
It’s been dubbed “the end of modern medicine”. BBC Future asked experts to explain how we might avoid the worst effects of antibiotic resistance – a grand challenge of our age.
Fact Sheet 3: ME/CFS: Information for Medical Professionals
Fact Sheet 3: ME/CFS: Information for Medical Professionals Published December 2025 Link to pdf: ME/CFS: Information for Medical Professionals.pdf Discussion thread: Fact sheet #3: Information...
Avoidable mortality, risk factors and policies for tackling noncommunicable diseases – leveraging data for impact: monitoring commitments in the WHO European Region ahead of the Fourth United Nations High-Level Meeting
The report notes that, in order to tackle NCDs and address the commercial, social, environmental and digital determinants of health, there is a need for accelerated multisectoral whole-of-government and whole-of-society action, integrated policies, enhanced preventive measures and strengthened health systems.
