Off-Protocol Radiation Therapy in Phase 3 Metastatic Solid Tumor Trials.

PURPOSE: Increasing data suggest that radiation therapy, particularly ablative radiation therapy, alters the natural history of metastatic disease. For patients with metastatic disease enrolled in prospective trials testing systemic therapy, the use of off-protocol radiation therapy to improve clinical symptoms or extend the duration of study systemic therapy may influence study endpoints. We sought to evaluate how often off-protocol radiation therapy was permitted among systemic therapy phase 3 trials, how often off-protocol radiation therapy is used, and whether off-protocol radiation therapy correlated with study outcomes. METHODS AND MATERIALS: Two-arm, superiority-design, phase 3 randomized trials testing systemic therapy were screened from ClinicalTrials.gov. Protocol availability was required to assess the trial approach to off-protocol radiation therapy if not described in the manuscript. Adjusted odds ratios with 95% CI were calculated by logistic regression. RESULTS: A total of 112 trials enrolling 80,134 patients were included, with publication dates between 2010 and 2019. Of these, off-protocol radiation therapy was allowed, not discussed, or prohibited during study systemic therapy in 52% (N =58), 25% (N = 28), and 23% (N = 26) of trials, respectively. However, only 2% (2 of 112) of trials reported off-protocol radiation therapy utilization rates, although no data were reported on the use of ablative off-protocol radiation therapy. No trials evaluated or adjusted for the potential influence of off-protocol radiation therapy on study endpoints. Among the subset of open-label studies, trials permissive toward off-protocol radiation therapy were more likely to meet their primary endpoint (adjusted odds ratio, 4.50; 95% CI, 1.23-20.23; P = .04). CONCLUSIONS: Although most trials allowed off-protocol radiation therapy during the receipt of the study systemic therapy, the influence of off-protocol radiation therapy, especially ablative radiation therapy, on study outcomes is underevaluated among phase 3 systemic therapy trials.

Technical Development and In Silico Implementation of SyntheticMR in Head and Neck Adaptive Radiation Therapy: A Prospective R-IDEAL Stage 0/1 Technology Development Report.

OBJECTIVE: The purpose of this study was to investigate the technical feasibility of integrating the quantitative maps available from SyntheticMR into the head and neck adaptive radiation oncology workflow. While SyntheticMR has been investigated for diagnostic applications, no studies have investigated its feasibility and potential for MR-Simulation or MR-Linac workflow. Demonstrating the feasibility of using this technique will facilitate rapid quantitative biomarker extraction which can be leveraged to guide adaptive radiation therapy decision making. APPROACH: Two phantoms, two healthy volunteers, and one patient were scanned using SyntheticMR on the MR-Simulation and MR-Linac devices with scan times between four to six minutes. Images in phantoms and volunteers were conducted in a test/retest protocol. The correlation between measured and reference quantitative T1, T2, and PD values were determined across clinical ranges in the phantom. Distortion was also studied. Contours of head and neck organs-at-risk (OAR) were drawn and applied to extract T1, T2, and PD. These values were plotted against each other, clusters were computed, and their separability significance was determined to evaluate SyntheticMR for differentiating tumor and normal tissue. MAIN RESULTS: The Lin's Concordance Correlation Coefficient between the measured and phantom reference values was above 0.98 for both the MR-Sim and MR-Linac. No significant levels of distortion were measured. The mean bias between the measured and phantom reference values across repeated scans was below 4% for T1, 7% for T2, and 4% for PD for both the MR-Sim and MR-Linac. For T1 vs. T2 and T1 vs. PD, the GTV contour exhibited perfect purity against neighboring OARs while being 0.7 for T2 vs. PD. All cluster significance levels between the GTV and the nearest OAR, the tongue, using the SigClust method was p < 0.001. SIGNIFICANCE: The technical feasibility of SyntheticMR was confirmed. Application of this technique to the head and neck adaptive radiation therapy workflow can enrich the current quantitative biomarker landscape.

DITTO: A Visual Digital Twin for Interventions and Temporal Treatment Outcomes in Head and Neck Cancer.

Digital twin models are of high interest to Head and Neck Cancer (HNC) oncologists, who have to navigate a series of complex treatment decisions that weigh the efficacy of tumor control against toxicity and mortality risks. Evaluating individual risk profiles necessitates a deeper understanding of the interplay between different factors such as patient health, spatial tumor location and spread, and risk of subsequent toxicities that can not be adequately captured through simple heuristics. To support clinicians in better understanding tradeoffs when deciding on treatment courses, we developed DITTO, a digital-twin and visual computing system that allows clinicians to analyze detailed risk profiles for each patient, and decide on a treatment plan. DITTO relies on a sequential Deep Reinforcement Learning digital twin (DT) to deliver personalized risk of both long-term and short-term disease outcome and toxicity risk for HNC patients. Based on a participatory collaborative design alongside oncologists, we also implement several visual explainability methods to promote clinical trust and encourage healthy skepticism when using our system. We evaluate the efficacy of DITTO through quantitative evaluation of performance and case studies with qualitative feedback. Finally, we discuss design lessons for developing clinical visual XAI applications for clinical end users.

Generalizability assessment of AI models across hospitals in a low-middle and high income country.

The integration of artificial intelligence (AI) into healthcare systems within low-middle income countries (LMICs) has emerged as a central focus for various initiatives aiming to improve healthcare access and delivery quality. In contrast to high-income countries (HICs), which often possess the resources and infrastructure to adopt innovative healthcare technologies, LMICs confront resource limitations such as insufficient funding, outdated infrastructure, limited digital data, and a shortage of technical expertise. Consequently, many algorithms initially trained on data from non-LMIC settings are now being employed in LMIC contexts. However, the effectiveness of these systems in LMICs can be compromised when the unique local contexts and requirements are not adequately considered. In this study, we evaluate the feasibility of utilizing models developed in the United Kingdom (a HIC) within hospitals in Vietnam (a LMIC). Consequently, we present and discuss practical methodologies aimed at improving model performance, emphasizing the critical importance of tailoring solutions to the distinct healthcare systems found in LMICs. Our findings emphasize the necessity for collaborative initiatives and solutions that are sensitive to the local context in order to effectively tackle the healthcare challenges that are unique to these regions.

Atrial fibrillation after cardiac surgery: identifying candidate predictors through a Delphi process.

OBJECTIVES: This study was undertaken to identify potential predictors of atrial fibrillation after cardiac surgery (AFACS) through a modified Delphi process and expert consensus. These will supplement predictors identified through a systematic review and cohort study to inform the development of two AFACS prediction models as part of the PARADISE project (NCT05255224). Atrial fibrillation is a common complication after cardiac surgery. It is associated with worse postoperative outcomes. Reliable prediction of AFACS would enable risk stratification and targeted prevention. Systematic identification of candidate predictors is important to improve validity of AFACS prediction tools. DESIGN: This study is a Delphi consensus exercise. SETTING: This study was undertaken through remote participation. PARTICIPANTS: The participants are an international multidisciplinary panel of experts selected through national research networks. INTERVENTIONS: This is a two-stage consensus exercise consisting of generating a long list of variables, followed by refinement by voting and retaining variables selected by at least 40% of panel members. RESULTS: The panel comprised 15 experts who participated in both stages, comprising cardiac intensive care physicians (n=3), cardiac anaesthetists (n=2), cardiac surgeons (n=1), cardiologists (n=4), cardiac pharmacists (n=1), critical care nurses (n=1), cardiac nurses (n=1) and patient representatives (n=2). Our Delphi process highlighted candidate AFACS predictors, including both patient factors and those related to the surgical intervention. We generated a final list of 72 candidate predictors. The final list comprised 3 demographic, 29 comorbidity, 4 vital sign, 13 intraoperative, 10 postoperative investigation and 13 postoperative intervention predictors. CONCLUSIONS: A Delphi consensus exercise has the potential to highlight predictors beyond the scope of existing literature. This method proved effective in identifying a range of candidate AFACS predictors. Our findings will inform the development of future AFACS prediction tools as part of the larger PARADISE project. TRIAL REGISTRATION NUMBER: NCT05255224.

Mine Your Own Anatomy: Revisiting Medical Image Segmentation With Extremely Limited Labels.

Recent studies on contrastive learning have achieved remarkable performance solely by leveraging few labels in the context of medical image segmentation. Existing methods mainly focus on instance discrimination and invariant mapping (i.e., pulling positive samples closer and negative samples apart in the feature space). However, they face three common pitfalls: (1) tailness: medical image data usually follows an implicit long-tail class distribution. Blindly leveraging all pixels in training hence can lead to the data imbalance issues, and cause deteriorated performance; (2) consistency: it remains unclear whether a segmentation model has learned meaningful and yet consistent anatomical features due to the intra-class variations between different anatomical features; and (3) diversity: the intra-slice correlations within the entire dataset have received significantly less attention. This motivates us to seek a principled approach for strategically making use of the dataset itself to discover similar yet distinct samples from different anatomical views. In this paper, we introduce a novel semi-supervised 2D medical image segmentation framework termed Mine yOur owNAnatomy (MONA), and make three contributions. First, prior work argues that every pixel equally matters to the model training; we observe empirically that this alone is unlikely to define meaningful anatomical features, mainly due to lacking the supervision signal. We show two simple solutions towards learning invariances-through the use of stronger data augmentations and nearest neighbors. Second, we construct a set of objectives that encourage the model to be capable of decomposing medical images into a collection of anatomical features in an unsupervised manner. Lastly, we both empirically and theoretically, demonstrate the efficacy of our MONA on three benchmark datasets with different labeled settings, achieving new state-of-the-art under different labeled semi-supervised settings. MONA makes minimal assumptions on domain expertise, and hence constitutes a practical and versatile solution in medical image analysis. We provide the PyTorch-like pseudo-code in supplementary.

Improving the clinical meaning of surrogate endpoints: An empirical assessment of clinical progression in phase III oncology trials.

Disease progression in clinical trials is commonly defined by radiologic measures. However, clinical progression may be more meaningful to patients, may occur even when radiologic criteria for progression are not met, and often requires a change in therapy in clinical practice. The objective of this study was to determine the utilization of clinical progression criteria within progression-based trial endpoints among phase III trials testing systemic therapies for metastatic solid tumors. The primary manuscripts and protocols of phase III trials were reviewed for whether clinical events, such as refractory pain, tumor bleeding, or neurologic compromise, could constitute a progression event. Univariable logistic regression computed odds ratios (OR) and 95% CI for associations between trial-level covariates and clinical progression. A total of 216 trials enrolling 148,190 patients were included, with publication dates from 2006 through 2020. A major change in clinical status was included in the progression criteria of 13% of trials (n = 27), most commonly as a secondary endpoint (n = 22). Only 59% of trials (n = 16) reported distinct clinical progression outcomes that constituted the composite surrogate endpoint. Compared with other disease sites, genitourinary trials were more likely to include clinical progression definitions (16/33 [48%] vs. 11/183 [6%]; OR, 14.72; 95% CI, 5.99 to 37.84; p < .0001). While major tumor-related clinical events were seldom considered as disease progression events, increased attention to clinical progression may improve the meaningfulness and clinical applicability of surrogate endpoints for patients with metastatic solid tumors.

CliqueFluxNet: Unveiling EHR Insights with Stochastic Edge Fluxing and Maximal Clique Utilisation Using Graph Neural Networks.

Electronic Health Records (EHRs) play a crucial role in shaping predictive are models, yet they encounter challenges such as significant data gaps and class imbalances. Traditional Graph Neural Network (GNN) approaches have limitations in fully leveraging neighbourhood data or demanding intensive computational requirements for regularisation. To address this challenge, we introduce CliqueFluxNet, a novel framework that innovatively constructs a patient similarity graph to maximise cliques, thereby highlighting strong inter-patient connections. At the heart of CliqueFluxNet lies its stochastic edge fluxing strategy - a dynamic process involving random edge addition and removal during training. This strategy aims to enhance the model's generalisability and mitigate overfitting. Our empirical analysis, conducted on MIMIC-III and eICU datasets, focuses on the tasks of mortality and readmission prediction. It demonstrates significant progress in representation learning, particularly in scenarios with limited data availability. Qualitative assessments further underscore CliqueFluxNet's effectiveness in extracting meaningful EHR representations, solidifying its potential for advancing GNN applications in healthcare analytics.

Mitigating machine learning bias between high income and low-middle income countries for enhanced model fairness and generalizability.

Collaborative efforts in artificial intelligence (AI) are increasingly common between high-income countries (HICs) and low- to middle-income countries (LMICs). Given the resource limitations often encountered by LMICs, collaboration becomes crucial for pooling resources, expertise, and knowledge. Despite the apparent advantages, ensuring the fairness and equity of these collaborative models is essential, especially considering the distinct differences between LMIC and HIC hospitals. In this study, we show that collaborative AI approaches can lead to divergent performance outcomes across HIC and LMIC settings, particularly in the presence of data imbalances. Through a real-world COVID-19 screening case study, we demonstrate that implementing algorithmic-level bias mitigation methods significantly improves outcome fairness between HIC and LMIC sites while maintaining high diagnostic sensitivity. We compare our results against previous benchmarks, utilizing datasets from four independent United Kingdom Hospitals and one Vietnamese hospital, representing HIC and LMIC settings, respectively.

Addressing label noise for electronic health records: insights from computer vision for tabular data.

The analysis of extensive electronic health records (EHR) datasets often calls for automated solutions, with machine learning (ML) techniques, including deep learning (DL), taking a lead role. One common task involves categorizing EHR data into predefined groups. However, the vulnerability of EHRs to noise and errors stemming from data collection processes, as well as potential human labeling errors, poses a significant risk. This risk is particularly prominent during the training of DL models, where the possibility of overfitting to noisy labels can have serious repercussions in healthcare. Despite the well-documented existence of label noise in EHR data, few studies have tackled this challenge within the EHR domain. Our work addresses this gap by adapting computer vision (CV) algorithms to mitigate the impact of label noise in DL models trained on EHR data. Notably, it remains uncertain whether CV methods, when applied to the EHR domain, will prove effective, given the substantial divergence between the two domains. We present empirical evidence demonstrating that these methods, whether used individually or in combination, can substantially enhance model performance when applied to EHR data, especially in the presence of noisy/incorrect labels. We validate our methods and underscore their practical utility in real-world EHR data, specifically in the context of COVID-19 diagnosis. Our study highlights the effectiveness of CV methods in the EHR domain, making a valuable contribution to the advancement of healthcare analytics and research.

Lost in the plot: missing visual elements in Kaplan-Meier plots of phase III oncology trials.

Missing visual elements (MVE) in Kaplan-Meier (KM) curves can misrepresent data, preclude curve reconstruction, and hamper transparency. This study evaluated KM plots of phase III oncology trials. MVE were defined as an incomplete y-axis range or missing number at risk table in a KM curve. Surrogate endpoint KM curves were additionally evaluated for complete interpretability, defined by (1) reporting the number of censored patients and (2) correspondence of the disease assessment interval with the number at risk interval. Among 641 trials enrolling 518 235 patients, 116 trials (18%) had MVE in KM curves. Industry sponsorship, larger trials, and more recently published trials were correlated with lower odds of MVE. Only 3% of trials (15 of 574) published surrogate endpoint KM plots with complete interpretability. Improvements in the quality of KM curves of phase III oncology trials, particularly for surrogate endpoints, are needed for greater interpretability, reproducibility, and transparency in oncology research.

Self-Supervised Machine Learning to Characterize Step Counts from Wrist-Worn Accelerometers in the UK Biobank.

PURPOSE: Step count is an intuitive measure of physical activity frequently quantified in health-related studies; however, accurate step counting is difficult in the free-living environment, with error routinely above 20% in wrist-worn devices against camera-annotated ground truth. This study aimed to describe the development and validation of step count derived from a wrist-worn accelerometer and assess its association with cardiovascular and all-cause mortality in a large prospective cohort. METHODS: We developed and externally validated a self-supervised machine learning step detection model, trained on an open-source and step-annotated free-living dataset. Thirty-nine individuals will free-living ground-truth annotated step counts were used for model development. An open-source dataset with 30 individuals was used for external validation. Epidemiological analysis was performed using 75,263 UK Biobank participants without prevalent cardiovascular disease (CVD) or cancer. Cox regression was used to test the association of daily step count with fatal CVD and all-cause mortality after adjustment for potential confounders. RESULTS: The algorithm substantially outperformed reference models (free-living mean absolute percent error of 12.5% vs 65%-231%). Our data indicate an inverse dose-response association, where taking 6430-8277 daily steps was associated with 37% (25%-48%) and 28% (20%-35%) lower risk of fatal CVD and all-cause mortality up to 7 yr later, compared with those taking fewer steps each day. CONCLUSIONS: We have developed an open and transparent method that markedly improves the measurement of steps in large-scale wrist-worn accelerometer datasets. The application of this method demonstrated expected associations with CVD and all-cause mortality, indicating excellent face validity. This reinforces public health messaging for increasing physical activity and can help lay the groundwork for the inclusion of target step counts in future public health guidelines.

Continuous patient state attention model for addressing irregularity in electronic health records.

BACKGROUND: Irregular time series (ITS) are common in healthcare as patient data is recorded in an electronic health record (EHR) system as per clinical guidelines/requirements but not for research and depends on a patient's health status. Due to irregularity, it is challenging to develop machine learning techniques to uncover vast intelligence hidden in EHR big data, without losing performance on downstream patient outcome prediction tasks. METHODS: In this paper, we propose Perceiver, a cross-attention-based transformer variant that is computationally efficient and can handle long sequences of time series in healthcare. We further develop continuous patient state attention models, using Perceiver and transformer to deal with ITS in EHR. The continuous patient state models utilise neural ordinary differential equations to learn patient health dynamics, i.e., patient health trajectory from observed irregular time steps, which enables them to sample patient state at any time. RESULTS: The proposed models' performance on in-hospital mortality prediction task on PhysioNet-2012 challenge and MIMIC-III datasets is examined. Perceiver model either outperforms or performs at par with baselines, and reduces computations by about nine times when compared to the transformer model, with no significant loss of performance. Experiments to examine irregularity in healthcare reveal that continuous patient state models outperform baselines. Moreover, the predictive uncertainty of the model is used to refer extremely uncertain cases to clinicians, which enhances the model's performance. Code is publicly available and verified at https://codeocean.com/capsule/4587224 . CONCLUSIONS: Perceiver presents a computationally efficient potential alternative for processing long sequences of time series in healthcare, and the continuous patient state attention models outperform the traditional and advanced techniques to handle irregularity in the time series. Moreover, the predictive uncertainty of the model helps in the development of transparent and trustworthy systems, which can be utilised as per the availability of clinicians.

Deep reinforcement learning for multi-class imbalanced training: applications in healthcare.

With the rapid growth of memory and computing power, datasets are becoming increasingly complex and imbalanced. This is especially severe in the context of clinical data, where there may be one rare event for many cases in the majority class. We introduce an imbalanced classification framework, based on reinforcement learning, for training extremely imbalanced data sets, and extend it for use in multi-class settings. We combine dueling and double deep Q-learning architectures, and formulate a custom reward function and episode-training procedure, specifically with the capability of handling multi-class imbalanced training. Using real-world clinical case studies, we demonstrate that our proposed framework outperforms current state-of-the-art imbalanced learning methods, achieving more fair and balanced classification, while also significantly improving the prediction of minority classes. Supplementary Information: The online version contains supplementary material available at 10.1007/s10994-023-06481-z.

Decoding 2.3 million ECGs: interpretable deep learning for advancing cardiovascular diagnosis and mortality risk stratification.

Aims: Electrocardiogram (ECG) is widely considered the primary test for evaluating cardiovascular diseases. However, the use of artificial intelligence (AI) to advance these medical practices and learn new clinical insights from ECGs remains largely unexplored. We hypothesize that AI models with a specific design can provide fine-grained interpretation of ECGs to advance cardiovascular diagnosis, stratify mortality risks, and identify new clinically useful information. Methods and results: Utilizing a data set of 2 322 513 ECGs collected from 1 558 772 patients with 7 years follow-up, we developed a deep-learning model with state-of-the-art granularity for the interpretable diagnosis of cardiac abnormalities, gender identification, and hypertension screening solely from ECGs, which are then used to stratify the risk of mortality. The model achieved the area under the receiver operating characteristic curve (AUC) scores of 0.998 (95% confidence interval (CI), 0.995-0.999), 0.964 (95% CI, 0.963-0.965), and 0.839 (95% CI, 0.837-0.841) for the three diagnostic tasks separately. Using ECG-predicted results, we find high risks of mortality for subjects with sinus tachycardia (adjusted hazard ratio (HR) of 2.24, 1.96-2.57), and atrial fibrillation (adjusted HR of 2.22, 1.99-2.48). We further use salient morphologies produced by the deep-learning model to identify key ECG leads that achieved similar performance for the three diagnoses, and we find that the V1 ECG lead is important for hypertension screening and mortality risk stratification of hypertensive cohorts, with an AUC of 0.816 (0.814-0.818) and a univariate HR of 1.70 (1.61-1.79) for the two tasks separately. Conclusion: Using ECGs alone, our developed model showed cardiologist-level accuracy in interpretable cardiac diagnosis and the advancement in mortality risk stratification. In addition, it demonstrated the potential to facilitate clinical knowledge discovery for gender and hypertension detection which are not readily available.

Prevention and Management of Osteoradionecrosis in Patients With Head and Neck Cancer Treated With Radiation Therapy: ISOO-MASCC-ASCO Guideline.

PURPOSE: To provide evidence-based recommendations for prevention and management of osteoradionecrosis (ORN) of the jaw secondary to head and neck radiation therapy in patients with cancer. METHODS: The International Society of Oral Oncology-Multinational Association for Supportive Care in Cancer (ISOO-MASCC) and ASCO convened a multidisciplinary Expert Panel to evaluate the evidence and formulate recommendations. PubMed, EMBASE, and Cochrane Library databases were searched for randomized controlled trials and observational studies, published between January 1, 2009, and December 1, 2023. The guideline also incorporated systematic reviews conducted by ISOO-MASCC, which included studies published from January 1, 1990, through December 31, 2008. RESULTS: A total of 1,539 publications were initially identified. There were 487 duplicate publications, resulting in 1,052 studies screened by abstract, 104 screened by full text, and 80 included for systematic review evaluation. RECOMMENDATIONS: Due to limitations of available evidence, the guideline relied on informal consensus for some recommendations. Recommendations that were deemed evidence-based with strong evidence by the Expert Panel were those pertaining to best practices in prevention of ORN and surgical management. No recommendation was possible for the utilization of leukocyte- and platelet-rich fibrin or photobiomodulation for prevention of ORN. The use of hyperbaric oxygen in prevention and management of ORN remains largely unjustified, with limited evidence to support its practice.Additional information is available at www.asco.org/head-neck-cancer-guidelines.

Development of an enhanced scoring system to predict ICU readmission or in-hospital death within 24 hours using routine patient data from two NHS Foundation Trusts.

RATIONALE: Intensive care units (ICUs) admit the most severely ill patients. Once these patients are discharged from the ICU to a step-down ward, they continue to have their vital signs monitored by nursing staff, with Early Warning Score (EWS) systems being used to identify those at risk of deterioration. OBJECTIVES: We report the development and validation of an enhanced continuous scoring system for predicting adverse events, which combines vital signs measured routinely on acute care wards (as used by most EWS systems) with a risk score of a future adverse event calculated on discharge from the ICU. DESIGN: A modified Delphi process identified candidate variables commonly available in electronic records as the basis for a 'static' score of the patient's condition immediately after discharge from the ICU. L1-regularised logistic regression was used to estimate the in-hospital risk of future adverse event. We then constructed a model of physiological normality using vital sign data from the day of hospital discharge. This is combined with the static score and used continuously to quantify and update the patient's risk of deterioration throughout their hospital stay. SETTING: Data from two National Health Service Foundation Trusts (UK) were used to develop and (externally) validate the model. PARTICIPANTS: A total of 12 394 vital sign measurements were acquired from 273 patients after ICU discharge for the development set, and 4831 from 136 patients in the validation cohort. RESULTS: Outcome validation of our model yielded an area under the receiver operating characteristic curve of 0.724 for predicting ICU readmission or in-hospital death within 24 hours. It showed an improved performance with respect to other competitive risk scoring systems, including the National EWS (0.653). CONCLUSIONS: We showed that a scoring system incorporating data from a patient's stay in the ICU has better performance than commonly used EWS systems based on vital signs alone. TRIAL REGISTRATION NUMBER: ISRCTN32008295.

Self-supervised learning for human activity recognition using 700,000 person-days of wearable data.

Accurate physical activity monitoring is essential to understand the impact of physical activity on one's physical health and overall well-being. However, advances in human activity recognition algorithms have been constrained by the limited availability of large labelled datasets. This study aims to leverage recent advances in self-supervised learning to exploit the large-scale UK Biobank accelerometer dataset-a 700,000 person-days unlabelled dataset-in order to build models with vastly improved generalisability and accuracy. Our resulting models consistently outperform strong baselines across eight benchmark datasets, with an F1 relative improvement of 2.5-130.9% (median 24.4%). More importantly, in contrast to previous reports, our results generalise across external datasets, cohorts, living environments, and sensor devices. Our open-sourced pre-trained models will be valuable in domains with limited labelled data or where good sampling coverage (across devices, populations, and activities) is hard to achieve.

AutoNet-Generated Deep Layer-Wise Convex Networks for ECG Classification.

The design of neural networks typically involves trial-and-error, a time-consuming process for obtaining an optimal architecture, even for experienced researchers. Additionally, it is widely accepted that loss functions of deep neural networks are generally non-convex with respect to the parameters to be optimised. We propose the Layer-wise Convex Theorem to ensure that the loss is convex with respect to the parameters of a given layer, achieved by constraining each layer to be an overdetermined system of non-linear equations. Based on this theorem, we developed an end-to-end algorithm (the AutoNet) to automatically generate layer-wise convex networks (LCNs) for any given training set. We then demonstrate the performance of the AutoNet-generated LCNs (AutoNet-LCNs) compared to state-of-the-art models on three electrocardiogram (ECG) classification benchmark datasets, with further validation on two non-ECG benchmark datasets for more general tasks. The AutoNet-LCN was able to find networks customised for each dataset without manual fine-tuning under 2 GPU-hours, and the resulting networks outperformed the state-of-the-art models with fewer than 5% parameters on all the above five benchmark datasets. The efficiency and robustness of the AutoNet-LCN markedly reduce model discovery costs and enable efficient training of deep learning models in resource-constrained settings.

AZD1222 effectiveness against severe COVID-19 in individuals with comorbidity or frailty: The RAVEN cohort study.

OBJECTIVES: Despite being prioritized during initial COVID-19 vaccine rollout, vulnerable individuals at high risk of severe COVID-19 (hospitalization, intensive care unit admission, or death) remain underrepresented in vaccine effectiveness (VE) studies. The RAVEN cohort study (NCT05047822) assessed AZD1222 (ChAdOx1 nCov-19) two-dose primary series VE in vulnerable populations. METHODS: Using the Oxford-Royal College of General Practitioners Clinical Informatics Digital Hub, linked to secondary care, death registration, and COVID-19 datasets in England, COVID-19 outcomes in 2021 were compared in vaccinated and unvaccinated individuals matched on age, sex, region, and multimorbidity. RESULTS: Over 4.5 million AZD1222 recipients were matched (mean follow-up ∼5 months); 68% were ≥50 years, 57% had high multimorbidity. Overall, high VE against severe COVID-19 was demonstrated, with lower VE observed in vulnerable populations. VE against hospitalization was higher in the lowest multimorbidity quartile (91.1%; 95% CI: 90.1, 92.0) than the highest quartile (80.4%; 79.7, 81.1), and among individuals ≥65 years, higher in the 'fit' (86.2%; 84.5, 87.6) than the frailest (71.8%; 69.3, 74.2). VE against hospitalization was lowest in immunosuppressed individuals (64.6%; 60.7, 68.1). CONCLUSIONS: Based on integrated and comprehensive UK health data, overall population-level VE with AZD1222 was high. VEs were notably lower in vulnerable groups, particularly the immunosuppressed.