Eye tracking insights into physician behaviour with safe and unsafe explainable AI recommendations.

We studied clinical AI-supported decision-making as an example of a high-stakes setting in which explainable AI (XAI) has been proposed as useful (by theoretically providing physicians with context for the AI suggestion and thereby helping them to reject unsafe AI recommendations). Here, we used objective neurobehavioural measures (eye-tracking) to see how physicians respond to XAI with N = 19 ICU physicians in a hospital's clinical simulation suite. Prescription decisions were made both pre- and post-reveal of either a safe or unsafe AI recommendation and four different types of simultaneously presented XAI. We used overt visual attention as a marker for where physician mental attention was directed during the simulations. Unsafe AI recommendations attracted significantly greater attention than safe AI recommendations. However, there was no appreciably higher level of attention placed onto any of the four types of explanation during unsafe AI scenarios (i.e. XAI did not appear to 'rescue' decision-makers). Furthermore, self-reported usefulness of explanations by physicians did not correlate with the level of attention they devoted to the explanations reinforcing the notion that using self-reports alone to evaluate XAI tools misses key aspects of the interaction behaviour between human and machine.

Machine Learning Tools for Acute Respiratory Distress Syndrome Detection and Prediction.

Machine learning (ML) tools for acute respiratory distress syndrome (ARDS) detection and prediction are increasingly used. Therefore, understanding risks and benefits of such algorithms is relevant at the bedside. ARDS is a complex and severe lung condition that can be challenging to define precisely due to its multifactorial nature. It often arises as a response to various underlying medical conditions, such as pneumonia, sepsis, or trauma, leading to widespread inflammation in the lungs. ML has shown promising potential in supporting the recognition of ARDS in ICU patients. By analyzing a variety of clinical data, including vital signs, laboratory results, and imaging findings, ML models can identify patterns and risk factors associated with the development of ARDS. This detection and prediction could be crucial for timely interventions, diagnosis and treatment. In summary, leveraging ML for the early prediction and detection of ARDS in ICU patients holds great potential to enhance patient care, improve outcomes, and contribute to the evolving landscape of precision medicine in critical care settings. This article is a concise definitive review on artificial intelligence and ML tools for the prediction and detection of ARDS in critically ill patients.

Infections in long-duration space missions.

As government space agencies and private companies announce plans for deep space exploration and colonisation, prioritisation of medical preparedness is becoming crucial. Among all medical conditions, infections pose one of the biggest threats to astronaut health and mission success. To gain a comprehensive understanding of these risks, we review the measured and estimated incidence of infections in space, effect of space environment on the human immune system and microbial behaviour, current preventive and management strategies for infections, and future perspectives for diagnosis and treatment. This information will enable space agencies to enhance their comprehension of the risk of infection in space, highlight gaps in knowledge, aid better crew preparation, and potentially contribute to sepsis management in terrestrial settings, including not only isolated or austere environments but also conventional clinical settings.

A scoping review of machine learning for sepsis prediction- feature engineering strategies and model performance: a step towards explainability.

BACKGROUND: Sepsis, an acute and potentially fatal systemic response to infection, significantly impacts global health by affecting millions annually. Prompt identification of sepsis is vital, as treatment delays lead to increased fatalities through progressive organ dysfunction. While recent studies have delved into leveraging Machine Learning (ML) for predicting sepsis, focusing on aspects such as prognosis, diagnosis, and clinical application, there remains a notable deficiency in the discourse regarding feature engineering. Specifically, the role of feature selection and extraction in enhancing model accuracy has been underexplored. OBJECTIVES: This scoping review aims to fulfill two primary objectives: To identify pivotal features for predicting sepsis across a variety of ML models, providing valuable insights for future model development, and To assess model efficacy through performance metrics including AUROC, sensitivity, and specificity. RESULTS: The analysis included 29 studies across diverse clinical settings such as Intensive Care Units (ICU), Emergency Departments, and others, encompassing 1,147,202 patients. The review highlighted the diversity in prediction strategies and timeframes. It was found that feature extraction techniques notably outperformed others in terms of sensitivity and AUROC values, thus indicating their critical role in improving sepsis prediction models. CONCLUSION: Key dynamic indicators, including vital signs and critical laboratory values, are instrumental in the early detection of sepsis. Applying feature selection methods significantly boosts model precision, with models like Random Forest and XG Boost showing promising results. Furthermore, Deep Learning models (DL) reveal unique insights, spotlighting the pivotal role of feature engineering in sepsis prediction, which could greatly benefit clinical practice.

Can Machine Learning Personalize Cardiovascular Therapy in Sepsis?

Large randomized trials in sepsis have generally failed to find effective novel treatments. This is increasingly attributed to patient heterogeneity, including heterogeneous cardiovascular changes in septic shock. We discuss the potential for machine learning systems to personalize cardiovascular resuscitation in sepsis. While the literature is replete with proofs of concept, the technological readiness of current systems is low, with a paucity of clinical trials and proven patient benefit. Systems may be vulnerable to confounding and poor generalization to new patient populations or contemporary patterns of care. Typical electronic health records do not capture rich enough data, at sufficient temporal resolution, to produce systems that make actionable treatment suggestions. To resolve these issues, we recommend a simultaneous focus on technical challenges and removing barriers to translation. This will involve improving data quality, adopting causally grounded models, prioritizing safety assessment and integration into healthcare workflows, conducting randomized clinical trials and aligning with regulatory requirements.

A locally optimised machine learning approach to early prognostication of long-term neurological outcomes after out-of-hospital cardiac arrest.

Background: Out-of-hospital cardiac arrest (OHCA) represents a major burden for society and health care, with an average incidence in adults of 67 to 170 cases per 100,000 person-years in Europe and in-hospital survival rates of less than 10%. Patients and practitioners would benefit from a prognostication tool for long-term good neurological outcomes. Objective: We aim to develop a machine learning (ML) pipeline on a local database to classify patients according to their neurological outcomes and identify prognostic features. Methods: We collected clinical and biological data consecutively from 595 patients who presented OHCA and were routed to a single regional cardiac arrest centre in the south of France. We applied recursive feature elimination and ML analyses to identify the main features associated with a good neurological outcome, defined as a Cerebral Performance Category score less than or equal to 2 at six months post-OHCA. Results: We identified 12 variables 24 h after admission, capable of predicting a six-month good neurological outcome. The best model (extreme gradient boosting) achieved an AUC of 0.96 and an accuracy of 0.92 in the test cohort. Conclusion: We demonstrated that it is possible to build accurate, locally optimised prediction and prognostication scores using datasets of limited size and breadth. We proposed and shared a generic machine-learning pipeline which allows external teams to replicate the approach locally.

Clinical practice, decision-making, and use of clinical decision support systems in invasive mechanical ventilation: a narrative review.

Invasive mechanical ventilation is a key supportive therapy for patients on intensive care. There is increasing emphasis on personalised ventilation strategies. Clinical decision support systems (CDSS) have been developed to support this. We conducted a narrative review to assess evidence that could inform device implementation. A search was conducted in MEDLINE (Ovid) and EMBASE. Twenty-nine studies met the inclusion criteria. Role allocation is well described, with interprofessional collaboration dependent on culture, nurse:patient ratio, the use of protocols, and perception of responsibility. There were no descriptions of process measures, quality metrics, or clinical workflow. Nurse-led weaning is well-described, with factors grouped by patient, nurse, and system. Physician-led weaning is heterogenous, guided by subjective and objective information, and 'gestalt'. No studies explored decision-making with CDSS. Several explored facilitators and barriers to implementation, grouped by clinician (facilitators: confidence using CDSS, retaining decision-making ownership; barriers: undermining clinician's role, ambiguity moving off protocol), intervention (facilitators: user-friendly interface, ease of workflow integration, minimal training requirement; barriers: increased documentation time), and organisation (facilitators: system-level mandate; barriers: poor communication, inconsistent training, lack of technical support). One study described factors that support CDSS implementation. There are gaps in our understanding of ventilation practice. A coordinated approach grounded in implementation science is required to support CDSS implementation. Future research should describe factors that guide clinical decision-making throughout mechanical ventilation, with and without CDSS, map clinical workflow, and devise implementation toolkits. Novel research design analogous to a learning organisation, that considers the commercial aspects of device design, is required.

Quantifying the impact of AI recommendations with explanations on prescription decision making.

The influence of AI recommendations on physician behaviour remains poorly characterised. We assess how clinicians' decisions may be influenced by additional information more broadly, and how this influence can be modified by either the source of the information (human peers or AI) and the presence or absence of an AI explanation (XAI, here using simple feature importance). We used a modified between-subjects design where intensive care doctors (N = 86) were presented on a computer for each of 16 trials with a patient case and prompted to prescribe continuous values for two drugs. We used a multi-factorial experimental design with four arms, where each clinician experienced all four arms on different subsets of our 24 patients. The four arms were (i) baseline (control), (ii) peer human clinician scenario showing what doses had been prescribed by other doctors, (iii) AI suggestion and (iv) XAI suggestion. We found that additional information (peer, AI or XAI) had a strong influence on prescriptions (significantly for AI, not so for peers) but simple XAI did not have higher influence than AI alone. There was no correlation between attitudes to AI or clinical experience on the AI-supported decisions and nor was there correlation between what doctors self-reported about how useful they found the XAI and whether the XAI actually influenced their prescriptions. Our findings suggest that the marginal impact of simple XAI was low in this setting and we also cast doubt on the utility of self-reports as a valid metric for assessing XAI in clinical experts.

Evaluation of free-floating tracheal intubation in weightlessness via ice-pick position with a direct laryngoscopy and classic approach with indirect videolaryngoscopy.

Long duration spaceflights to the Moon or Mars are at risk for emergency medical events. Managing a hypoxemic distress and performing an advanced airway procedure such as oro-tracheal intubation may be complicated under weightlessness due to ergonomic constraints. An emergency free-floating intubation would be dangerous because of high failure rates due to stabilization issues that prohibits its implementation in a space environment. Nevertheless, we hypothesized that two configurations could lead to a high first-pass success score for intubation performed by a free-floating operator. In a non-randomized, controlled, cross-over simulation study during a parabolic flight campaign, we evaluated and compared the intubation performance of free-floating trained operators, using either a conventional direct laryngoscope in an ice-pick position or an indirect laryngoscopy with a video-laryngoscope in a classic position at the head of a high-fidelity simulation manikin, in weightlessness and in normogravity. Neither of the two tested conditions reached the minimal terrestrial ILCOR recommendations (95% first-pass success) and therefore could not be recommended for general implementation under weightlessness conditions. Free-floating video laryngoscopy at the head of the manikin had a significant better success score than conventional direct laryngoscopy in an ice-pick position. Our results, combined with the preexisting literature, emphasis the difficulties of performing oro-tracheal intubation, even for experts using modern airway devices, under postural instability in weightlessness. ClinicalTrials registration number NCT05303948.

Multi-site, Multi-domain Airway Tree Modeling.

Open international challenges are becoming the de facto standard for assessing computer vision and image analysis algorithms. In recent years, new methods have extended the reach of pulmonary airway segmentation that is closer to the limit of image resolution. Since EXACT'09 pulmonary airway segmentation, limited effort has been directed to the quantitative comparison of newly emerged algorithms driven by the maturity of deep learning based approaches and extensive clinical efforts for resolving finer details of distal airways for early intervention of pulmonary diseases. Thus far, public annotated datasets are extremely limited, hindering the development of data-driven methods and detailed performance evaluation of new algorithms. To provide a benchmark for the medical imaging community, we organized the Multi-site, Multi-domain Airway Tree Modeling (ATM'22), which was held as an official challenge event during the MICCAI 2022 conference. ATM'22 provides large-scale CT scans with detailed pulmonary airway annotation, including 500 CT scans (300 for training, 50 for validation, and 150 for testing). The dataset was collected from different sites and it further included a portion of noisy COVID-19 CTs with ground-glass opacity and consolidation. Twenty-three teams participated in the entire phase of the challenge and the algorithms for the top ten teams are reviewed in this paper. Both quantitative and qualitative results revealed that deep learning models embedded with the topological continuity enhancement achieved superior performance in general. ATM'22 challenge holds as an open-call design, the training data and the gold standard evaluation are available upon successful registration via its homepage (https://atm22.grand-challenge.org/).

Artificial Intelligence Applications in Space Medicine.

INTRODUCTION:During future interplanetary space missions, a number of health conditions may arise, owing to the hostile environment of space and the myriad of stressors experienced by the crew. When managing these conditions, crews will be required to make accurate, timely clinical decisions at a high level of autonomy, as telecommunication delays and increasing distances restrict real-time support from the ground. On Earth, artificial intelligence (AI) has proven successful in healthcare, augmenting expert clinical decision-making or enhancing medical knowledge where it is lacking. Similarly, deploying AI tools in the context of a space mission could improve crew self-reliance and healthcare delivery.METHODS: We conducted a narrative review to discuss existing AI applications that could improve the prevention, recognition, evaluation, and management of the most mission-critical conditions, including psychological and mental health, acute radiation sickness, surgical emergencies, spaceflight-associated neuro-ocular syndrome, infections, and cardiovascular deconditioning.RESULTS: Some examples of the applications we identified include AI chatbots designed to prevent and mitigate psychological and mental health conditions, automated medical imaging analysis, and closed-loop systems for hemodynamic optimization. We also discuss at length gaps in current technologies, as well as the key challenges and limitations of developing and deploying AI for space medicine to inform future research and innovation. Indeed, shifts in patient cohorts, space-induced physiological changes, limited size and breadth of space biomedical datasets, and changes in disease characteristics may render the models invalid when transferred from ground settings into space.Cheung HC, De Louche C, Komorowski M. Artificial intelligence applications in space medicine. Aerosp Med Hum Perform. 2023; 94(8):610-622.

Sepsis biomarkers and diagnostic tools with a focus on machine learning.

Over the last years, there have been advances in the use of data-driven techniques to improve the definition, early recognition, subtypes characterisation, prognostication and treatment personalisation of sepsis. Some of those involve the discovery or evaluation of biomarkers or digital signatures of sepsis or sepsis sub-phenotypes. It is hoped that their identification may improve timeliness and accuracy of diagnosis, suggest physiological pathways and therapeutic targets, inform targeted recruitment into clinical trials, and optimise clinical management. Given the complexities of the sepsis response, panels of biomarkers or models combining biomarkers and clinical data are necessary, as well as specific data analysis methods, which broadly fall under the scope of machine learning. This narrative review gives a brief overview of the main machine learning techniques (mainly in the realms of supervised and unsupervised methods) and published applications that have been used to create sepsis diagnostic tools and identify biomarkers.

Assuring the safety of AI-based clinical decision support systems: a case study of the AI Clinician for sepsis treatment.

OBJECTIVES: Establishing confidence in the safety of Artificial Intelligence (AI)-based clinical decision support systems is important prior to clinical deployment and regulatory approval for systems with increasing autonomy. Here, we undertook safety assurance of the AI Clinician, a previously published reinforcement learning-based treatment recommendation system for sepsis. METHODS: As part of the safety assurance, we defined four clinical hazards in sepsis resuscitation based on clinical expert opinion and the existing literature. We then identified a set of unsafe scenarios, intended to limit the action space of the AI agent with the goal of reducing the likelihood of hazardous decisions. RESULTS: Using a subset of the Medical Information Mart for Intensive Care (MIMIC-III) database, we demonstrated that our previously published 'AI clinician' recommended fewer hazardous decisions than human clinicians in three out of our four predefined clinical scenarios, while the difference was not statistically significant in the fourth scenario. Then, we modified the reward function to satisfy our safety constraints and trained a new AI Clinician agent. The retrained model shows enhanced safety, without negatively impacting model performance. DISCUSSION: While some contextual patient information absent from the data may have pushed human clinicians to take hazardous actions, the data were curated to limit the impact of this confounder. CONCLUSION: These advances provide a use case for the systematic safety assurance of AI-based clinical systems towards the generation of explicit safety evidence, which could be replicated for other AI applications or other clinical contexts, and inform medical device regulatory bodies.

Early Intensive Physical Rehabilitation Combined with a Protocolized Decannulation Process in Tracheostomized Survivors from Severe COVID-19 Pneumonia with Chronic Critical Illness.

(1) Background: Intensive care unit (ICU) survivors from severe COVID-19 acute respiratory distress syndrome (CARDS) with chronic critical illness (CCI) may be considered vast resource consumers with a poor prognosis. We hypothesized that a holistic approach combining an early intensive rehabilitation with a protocol of difficult weaning would improve patient outcomes (2) Methods: A single-center retrospective study in a five-bed post-ICU weaning and intensive rehabilitation center with a dedicated fitness room specifically equipped to safely deliver physical activity sessions in frail patients with CCI. (3) Results: Among 502 CARDS patients admitted to the ICU from March 2020 to March 2022, 50 consecutive tracheostomized patients were included in the program. After a median of 39 ICU days, 25 days of rehabilitation were needed to restore patients' autonomy (ADL, from 0 to 6; p < 0.001), to significantly improve their aerobic capacity (6-min walking test distance, from 0 to 253 m; p < 0.001) and to reduce patients' vulnerability (frailty score, from 7 to 3; p < 0.001) and hospital anxiety and depression scale (HADS, from 18 to 10; p < 0.001). Forty-eight decannulated patients (96%) were discharged home. (4) Conclusions: A protocolized weaning strategy combined with early intensive rehabilitation in a dedicated specialized center boosted the physical and mental recovery.

Randomized Comparison of Two New Methods for Chest Compressions during CPR in Microgravity-A Manikin Study.

BACKGROUND: Although there have been no reported cardiac arrests in space to date, the risk of severe medical events occurring during long-duration spaceflights is a major concern. These critical events can endanger both the crew as well as the mission and include cardiac arrest, which would require cardiopulmonary resuscitation (CPR). Thus far, five methods to perform CPR in microgravity have been proposed. However, each method seems insufficient to some extent and not applicable at all locations in a spacecraft. The aim of the present study is to describe and gather data for two new CPR methods in microgravity. MATERIALS AND METHODS: A randomized, controlled trial (RCT) compared two new methods for CPR in a free-floating underwater setting. Paramedics performed chest compressions on a manikin (Ambu Man, Ambu, Germany) using two new methods for a free-floating position in a parallel-group design. The first method (Schmitz-Hinkelbein method) is similar to conventional CPR on earth, with the patient in a supine position lying on the operator's knees for stabilization. The second method (Cologne method) is similar to the first, but chest compressions are conducted with one elbow while the other hand stabilizes the head. The main outcome parameters included the total number of chest compressions (n) during 1 min of CPR (compression rate), the rate of correct chest compressions (%), and no-flow time (s). The study was registered on clinicaltrials.gov (NCT04354883). RESULTS: Fifteen volunteers (age 31.0 ± 8.8 years, height 180.3 ± 7.5 cm, and weight 84.1 ± 13.2 kg) participated in this study. Compared to the Cologne method, the Schmitz-Hinkelbein method showed superiority in compression rates (100.5 ± 14.4 compressions/min), correct compression depth (65 ± 23%), and overall high rates of correct thoracic release after compression (66% high, 20% moderate, and 13% low). The Cologne method showed correct depth rates (28 ± 27%) but was associated with a lower mean compression rate (73.9 ± 25.5/min) and with lower rates of correct thoracic release (20% high, 7% moderate, and 73% low). CONCLUSIONS: Both methods are feasible without any equipment and could enable immediate CPR during cardiac arrest in microgravity, even in a single-helper scenario. The Schmitz-Hinkelbein method appears superior and could allow the delivery of high-quality CPR immediately after cardiac arrest with sufficient quality.

Is artificial intelligence ready to solve mechanical ventilation? Computer says blow.

Artificial intelligence (AI) has the potential to identify treatable phenotypes, optimise ventilation strategies, and provide clinical decision support for patients who require mechanical ventilation. Gallifant and colleagues performed a systematic review to identify studies using AI to solve a diverse range of clinical problems in the ventilated patient. They identify 95 studies, the majority of which were reported in the last 5 yr. Their findings indicate that the majority of studies have significant methodological bias and are a long way from deployment.

The global mental health burden of COVID-19 on critical care staff.

BACKGROUND: Although the mental health burden in healthcare workers caused by COVID-19 has gained increasing attention both within the profession and through public opinion, there has been a lack of data describing their experience; specifically, the mental wellbeing of healthcare workers in the intensive care unit (ICU), including those redeployed. AIMS: The authors aimed to compare the mental health status of ICU healthcare workers (physicians, nurses and allied health professionals) affected by various factors during the COVID-19 pandemic; and highlight to policymakers areas of staff vulnerabilities in order to improve wellbeing strategies within healthcare systems. METHODS: An online survey using three validated scales was conducted in France, the UK, Italy, Mainland China, Taiwan, Egypt and Belgium. FINDINGS: The proportion of respondents who screened positive on the three scales across the countries was 16-49% for depression, 60-86% for insomnia and 17-35% for post-traumatic stress disorder. The authors also identified an increase in the scores with longer time spent in personal protective equipment, female gender, advancing age and redeployed status. CONCLUSION: The high prevalence of mental disorders among ICU staff during the COVID-19 crisis should inform local and national wellbeing policies.