There is a growing need to understand how digital systems can support clinical decision-making, particularly as artificial intelligence (AI) models become increasingly complex and less human-interpretable. This complexity raises concerns about trustworthiness, impacting safe and effective adoption of such technologies. Improved understanding of decision-making processes and requirements for explanations coming from decision support tools is a vital component in providing effective explainable solutions. This is particularly relevant in the data-intensive, fast-paced environments of intensive care units (ICUs). To explore these issues, group interviews were conducted with seven ICU clinicians, representing various roles and experience levels. Thematic analysis revealed three core themes: (T1) ICU decision-making relies on a wide range of factors, (T2) the complexity of patient state is challenging for shared decision-making, and (T3) requirements and capabilities of AI decision support systems. We include design recommendations from clinical input, providing insights to inform future AI systems for intensive care.

Reinforcement Learning in Healthcare is typically concerned with narrow self-contained tasks such as sepsis prediction or anesthesia control. However, previous research has demonstrated the potential of generalist models (the prime example being Large Language Models) to outperform task-specific approaches due to their capability for implicit transfer learning. To enable training of foundation models for Healthcare as well as leverage the capabilities of state of the art Transformer architectures, we propose the paradigm of Healthcare as Sequence Modeling, in which interaction between the patient and the healthcare provider is represented as an event stream and tasks like diagnosis and treatment selection are modeled as prediction of future events in the stream. To explore this paradigm experimentally we develop MIMIC-SEQ, a sequence modeling benchmark derived by translating heterogenous clinical records from MIMIC-IV dataset into a uniform event stream format, train a baseline model and explore its capabilities.

Foundation models hold promise for transforming AI in healthcare by providing modular components that are easily adaptable to downstream healthcare tasks, making AI development more scalable and cost-effective. Structured EHR foundation models, trained on coded medical records from millions of patients, demonstrated benefits including increased performance with fewer training labels, and improved robustness to distribution shifts. However, questions remain on the feasibility of sharing these models across different hospitals and their performance for local task adaptation. This multi-center study examined the adaptability of a recently released structured EHR foundation model ($FM_{SM}$), trained on longitudinal medical record data from 2.57M Stanford Medicine patients. Experiments were conducted using EHR data at The Hospital for Sick Children and MIMIC-IV. We assessed both adaptability via continued pretraining on local data, and task adaptability compared to baselines of training models from scratch at each site, including a local foundation model. We evaluated the performance of these models on 8 clinical prediction tasks. In both datasets, adapting the off-the-shelf $FM_{SM}$ matched the performance of GBM models locally trained on all data while providing a 13% improvement in settings with few task-specific training labels. With continued pretraining on local data, label efficiency substantially improved, such that $FM_{SM}$ required fewer than 1% of training examples to match the fully trained GBM's performance. Continued pretraining was also 60 to 90% more sample-efficient than training local foundation models from scratch. Our findings show that adapting shared EHR foundation models across hospitals provides improved prediction performance at less cost, underscoring the utility of base foundation models as modular components to streamline the development of healthcare AI.

Covid-19 can cause lasting cognitive and mental health issues, including brain fog, fatigue and even post-traumatic stress disorder. To better understand the scale of the problem, researchers at the University of Cambridge analysed 46 people who were hospitalised due to the infection between March and July 2020. The participants underwent cognitive tests on average six months after their initial illness. These results were compared against those of more than 66,000 people from the general population. Those hospitalised with covid-19 scored worse on verbal analogical reasoning tests, which assess an individual's ability to recognise relationships between ideas and think methodically. They also recorded slower processing speeds. Previous studies suggest glucose is less efficiently used by the part of the brain responsible for attention, complex problem-solving and working memory after covid-19. Scores and reaction speeds improved over time, however, any recovery was gradual at best, according to the researchers. This cognitive impairment probably has multiple causes, including inadequate blood supply to the brain, blood vessel blockage and microscopic bleeds caused by SARS-CoV-2 virus, as well as damage triggered by an overactive immune system, they added. "Around 40,000 people have been through intensive care with covid-19 in England alone and many more will have been very sick, but not admitted to hospital," Adam Hampshire at Imperial College London said in a statement. "This means there is a large number of people out there still experiencing problems with cognition many months later." The biological mechanism behind a rare and severe covid-19 response seen in some children may have been uncovered by researchers at the Murdoch Children's Research Institute in Melbourne, Australia. Doctors have so far been unable to identify why some children develop multisystem inflammatory syndrome (MIS) in response to covid-19, which can cause symptoms such as fever, abdominal pain and heart disease.

A single blood sample from a critically ill COVID-19 patient can be analyzed by a machine learning model which uses blood plasma proteins to predict survival, weeks before the outcome, according to a new study published this week in the open-access journal PLOS Digital Health by Florian Kurth and Markus Ralser of the Charité – Universitätsmedizin Berlin, Germany, and colleagues. Healthcare systems around the world are struggling to accommodate high numbers of severely ill COVID-19 patients who need special medical attention, especially if they are identified as being at high risk. Clinically established risk assessments in intensive care medicine, such as the SOFA or APACHE II, show only limited reliability in predicting future disease outcomes for COVID-19. In the new study, researchers studied the levels of 321 proteins in blood samples taken at 349 timepoints from 50 critically ill COVID-19 patients being treated in two independent health care centers in Germany and Austria. A machine learning approach was used to find associations between the measured proteins and patient survival.

Levels of 14 proteins in the blood of critically ill COVID-19 patients are associated with survival. A single blood sample from a critically ill COVID-19 patient can be analyzed by a machine learning model which uses blood plasma proteins to predict survival, weeks before the outcome, according to a new study published this week in the open-access journal PLOS Digital Health by Florian Kurth and Markus Ralser of the Charité – Universitätsmedizin Berlin, Germany, and colleagues. Healthcare systems around the world are struggling to accommodate high numbers of severely ill COVID-19 patients who need special medical attention, especially if they are identified as being at high risk. Clinically established risk assessments in intensive care medicine, such as the SOFA or APACHE II, show only limited reliability in predicting future disease outcomes for COVID-19. In the new study, researchers studied the levels of 321 proteins in blood samples taken at 349 timepoints from 50 critically ill COVID-19 patients being treated in two independent health care centers in Germany and Austria.

Researchers may have developed a new tool that uses machine learning to better predict health outcomes for hospitalized Covid patients, and help physicians make more informed treatment decisions. A German research team from Charity-University Medicine in Berlin – one of the country's largest university hospitals – developed an Artificial Intelligence tool that can estimate how well an infected person will fare based off of a blood sample. The levels of fourteen proteins found in a person's blood can indicate whether a person who suffers a severe enough hospitalization will survive or die from the virus, and the tool developed by researchers can accurately asses their risk. In times of crisis, where resources are especially scarce, the tool can help determine what patients require the most intensive care to survive, and who is more fit to fight off the virus themselves. Using blood samples from Covid patients, a German research team has found that levels of 14 proteins can help determine whether a person survives the virus.

Author summary Healthcare systems around the world are struggling to accommodate high numbers of the most severely ill patients with COVID-19. Moreover, the pandemic creates a pressing need to accelerate clinical trials investigating potential new therapeutics. While various biomarkers can discriminate and predict the future course of disease for patients of different disease severity, prognosis remains difficult for patient groups with similar disease severity, e.g. patients requiring intensive care. Established risk assessments in intensive care medicine such as the SOFA or APACHE II show only limited reliability in predicting future disease outcomes for COVID-19. In this study we hypothesized that the plasma proteome, which reflects the complete set of proteins that are expressed by an organism and are present in the blood, and which is known to comprehensively capture the host response to COVID-19, can be leveraged to allow for prediction of survival in the most critically ill patients with COVID-19. Here, we found 14 proteins, which over time changed in opposite directions for patients who survive compared to patients who do not survive on intensive care. Using a machine learning model which combines the measurements of multiple proteins, we were able to accurately predict survival in critically ill patients with COVID-19 from single blood samples, weeks before the outcome, substantially outperforming established risk predictors.

Despite the recent progress in the field of causal inference, to date there is no agreed upon methodology to glean treatment effect estimation from observational data. The consequence on clinical practice is that, when lacking results from a randomized trial, medical personnel is left without guidance on what seems to be effective in a real-world scenario. This article showcases a pragmatic methodology to obtain preliminary estimation of treatment effect from observational studies. Our approach was tested on the estimation of treatment effect of the proning maneuver on oxygenation levels, on a cohort of COVID-19 Intensive Care patients. We modeled our study design on a recent RCT for proning (the PROSEVA trial). Linear regression, propensity score models such as blocking and DR-IPW, BART and two versions of Counterfactual Regression were employed to provide estimates on observational data comprising first wave COVID-19 ICU patient data from 25 Dutch hospitals. 6371 data points, from 745 mechanically ventilated patients, were included in the study. Estimates for the early effect of proning -- P/F ratio from 2 to 8 hours after proning -- ranged between 14.54 and 20.11 mm Hg depending on the model. Estimates for the late effect of proning -- oxygenation from 12 to 24 hours after proning -- ranged between 13.53 and 15.26 mm Hg. All confidence interval being strictly above zero indicated that the effect of proning on oxygenation for COVID-19 patient was positive and comparable in magnitude to the effect on non COVID-19 patients. These results provide further evidence on the effectiveness of proning on the treatment of COVID-19 patients. This study, along with the accompanying open-source code, provides a blueprint for treatment effect estimation in scenarios where RCT data is lacking. Funding: SIDN fund, CovidPredict consortium, Pacmed.