Efficacy of decentralised home-based antihypertensive treatment in older adults with multimorbidity and polypharmacy (ATEMPT): an open-label randomised controlled pilot trial.

BACKGROUND: Older patients with multimorbidity and polypharmacy have been under-represented in clinical trials. We aimed to assess the effect of different intensities of antihypertensive treatment on changes in blood pressure, major safety outcomes, and patient-reported outcomes in this population. METHODS: ATEMPT was a decentralised, two-armed, parallel-group, open-label randomised controlled pilot trial conducted in the Thames Valley area, South East England. Individuals aged 65 years or older with multimorbidity (three or more chronic conditions) or polypharmacy (five or more types of medications) and a systolic blood pressure of 115-165 mm Hg were eligible for inclusion. Participants were identified through a search of national hospital discharge databases, identification of patients registered with an online pharmacy, and via targeted advertising on social media platforms. Participants were randomly assigned to receive up to two more classes versus up to two fewer classes of antihypertensive medications. Apart from routine home visits for conducting the baseline assessment, all communication, monitoring, and management of participants by the trial team was conducted remotely. The primary outcome was change in home-measured blood pressure. FINDINGS: Between Dec 15, 2020, and Aug 31, 2022, 230 participants were randomly assigned (n=126 to more vs n=104 to fewer antihypertensive medications). The frequency of serious adverse events was similar across both groups; no cardiovascular events occurred in the more antihypertensive drugs group, compared with six in the fewer antihypertensive drugs group, of which two were fatal. Over a 13-month follow-up period, the mean systolic blood pressure in the group allocated to receive more antihypertensive medications decreased from 134·5 mm Hg (SD 10·7) at baseline to 122·1 mm Hg (10·5). By contrast, in the group allocated to receive fewer antihypertensive medications, it remained relatively unchanged, moving from 134·8 mm Hg (SD 11·2) at baseline to 132·9 mm Hg (15·3); this corresponded to a mean difference of -10·7 mm Hg (95% CI -17·5 to -4·0). INTERPRETATION: Remotely delivered antihypertensive treatment substantially reduced systolic blood pressure in older adults who are often less represented in trials, with no increase in the risk of serious adverse events. The results of this trial will inform a larger clinical trial focusing on assessing major cardiovascular events, safety, physical functioning, and cognitive function that is currently in the planning stages. These results also underscore the efficiency of decentralised trial designs, which might be of broader interest in other settings. FUNDING: National Institute for Health Research Oxford Biomedical Research Centre and the Oxford Martin School.

Gout and incidence of 12 cardiovascular diseases: a case-control study including 152 663 individuals with gout and 709 981 matched controls.

BACKGROUND: Gout, a common crystal arthropathy, is associated with increased risk of cardiovascular disease. We aimed to identify how this risk varies by individual cardiovascular disease across a broad spectrum of conditions. METHODS: In this matched case-control study, we used linked primary and secondary electronic health records from the UK Clinical Practice Research Datalink to assemble a cohort of individuals with a first-time diagnosis of gout between Jan 1, 2000 and Dec 31, 2017, who were aged 80 years or younger at diagnosis, and free of cardiovascular diseases up to 12 months after diagnosis. The control cohort comprised up to five control individuals per patient with gout, matched on age, sex, socioeconomic status, geographical region, and calendar time, randomly selected among individuals free of gout at any time before and during the study period. The cohorts were followed up until June 30, 2019. We investigated the incidence of 12 cardiovascular diseases and used Cox proportional hazards models to examine differences in people with and without gout, overall and by subgroups of sex, age, socioeconomic status, and year of study inclusion. We further adjusted models for known cardiovascular risk factors (blood pressure, BMI, smoking status, cholesterol, type 2 diabetes, chronic kidney disease, and history of hypertension). FINDINGS: We identified 152 663 individuals with gout (mean age 56·2 years [SD 13·3]; 120 324 [78·8%] men and 32 339 [21·2%] women) and 709 981 matched controls (mean age 56·5 years [13·2]; 561 002 [79·0%] men and 148 979 [21·0%] women). Of these individuals, 31 479 (20·6%) with gout and 106 520 (15·0%) without gout developed cardiovascular disease during a median follow-up of 6·5 years (IQR 3·1-10·5). Patients with gout had higher risk of cardiovascular diseases than matched controls (hazard ratio [HR] 1·58 [95% CI 1·52-1·63]). Excess risk of cardiovascular disease in gout was greater in women than men (women: HR 1·88 [1·75-2·02]; men: HR 1·49 [1·43-1·56]), and, among all age groups, was highest in younger individuals (HR in people aged <45 years: 2·22 [1·92-2·57]). Excess risk was observed across all 12 cardiovascular diseases investigated. Patients with gout had higher BMI than matched controls (mean difference 2·90 kg/m2 [95% CI 2·87-2·93]) and higher prevalence of chronic kidney disease, dyslipidaemia, history of hypertension, obesity, and type 2 diabetes. Adjusting for known cardiovascular risk factors attenuated but did not eliminate the excess risk of cardiovascular disease related to gout (adjusted HR 1·31 [1·27-1·36]). INTERPRETATION: Patients with gout had an excess risk of developing a broad range of cardiovascular diseases that extend beyond atherosclerotic diseases and include heart failure, arrhythmias, valve disease, and thromboembolic diseases. Excess risk was highest in women and younger individuals. These findings suggest that strategies to reduce cardiovascular risk in patients with gout need to evolve and be implemented in clinical practice. FUNDING: Research Foundation Flanders.

A comparative study of model-centric and data-centric approaches in the development of cardiovascular disease risk prediction models in the UK Biobank.

Aims: A diverse set of factors influence cardiovascular diseases (CVDs), but a systematic investigation of the interplay between these determinants and the contribution of each to CVD incidence prediction is largely missing from the literature. In this study, we leverage one of the most comprehensive biobanks worldwide, the UK Biobank, to investigate the contribution of different risk factor categories to more accurate incidence predictions in the overall population, by sex, different age groups, and ethnicity. Methods and results: The investigated categories include the history of medical events, behavioural factors, socioeconomic factors, environmental factors, and measurements. We included data from a cohort of 405 257 participants aged 37-73 years and trained various machine learning and deep learning models on different subsets of risk factors to predict CVD incidence. Each of the models was trained on the complete set of predictors and subsets where each category was excluded. The results were benchmarked against QRISK3. The findings highlight that (i) leveraging a more comprehensive medical history substantially improves model performance. Relative to QRISK3, the best performing models improved the discrimination by 3.78% and improved precision by 1.80%. (ii) Both model- and data-centric approaches are necessary to improve predictive performance. The benefits of using a comprehensive history of diseases were far more pronounced when a neural sequence model, BEHRT, was used. This highlights the importance of the temporality of medical events that existing clinical risk models fail to capture. (iii) Besides the history of diseases, socioeconomic factors and measurements had small but significant independent contributions to the predictive performance. Conclusion: These findings emphasize the need for considering broad determinants and novel modelling approaches to enhance CVD incidence prediction.

Sodium-based paracetamol: impact on blood pressure, cardiovascular events, and all-cause mortality.

BACKGROUND AND AIMS: Effervescent formulations of paracetamol containing sodium bicarbonate have been reported to associate with increased blood pressure and a higher risk of cardiovascular diseases and all-cause mortality. Given the major implications of these findings, the reported associations were re-examined. METHODS: Using linked electronic health records data, a cohort of 475 442 UK individuals with at least one prescription of paracetamol, aged between 60 and 90 years, was identified. Outcomes in patients taking sodium-based paracetamol were compared with those taking non-sodium-based formulations of the same. Using a deep learning approach, associations with systolic blood pressure (SBP), major cardiovascular events (myocardial infarction, heart failure, and stroke), and all-cause mortality within 1 year after baseline were investigated. RESULTS: A total of 460 980 and 14 462 patients were identified for the non-sodium-based and sodium-based paracetamol exposure groups, respectively (mean age: 74 years; 64% women). Analysis revealed no difference in SBP [mean difference -0.04 mmHg (95% confidence interval -0.51, 0.43)] and no association with major cardiovascular events [relative risk (RR) 1.03 (0.91, 1.16)]. Sodium-based paracetamol showed a positive association with all-cause mortality [RR 1.46 (1.40, 1.52)]. However, after further accounting of other sources of residual confounding, the observed association attenuated towards the null [RR 1.08 (1.01, 1.16)]. Exploratory analyses revealed dysphagia and related conditions as major sources of uncontrolled confounding by indication for this association. CONCLUSIONS: This study does not support previous suggestions of increased SBP and an elevated risk of cardiovascular events from short-term use of sodium bicarbonate paracetamol in routine clinical practice.

Stratification of diabetes in the context of comorbidities, using representation learning and topological data analysis.

Diabetes is a heterogenous, multimorbid disorder with a large variation in manifestations, trajectories, and outcomes. The aim of this study is to validate a novel machine learning method for the phenotyping of diabetes in the context of comorbidities. Data from 9967 multimorbid patients with a new diagnosis of diabetes were extracted from Clinical Practice Research Datalink. First, using BEHRT (a transformer-based deep learning architecture), the embeddings corresponding to diabetes were learned. Next, topological data analysis (TDA) was carried out to test how different areas in high-dimensional manifold correspond to different risk profiles. The following endpoints were considered when profiling risk trajectories: major adverse cardiovascular events (MACE), coronary artery disease (CAD), stroke (CVA), heart failure (HF), renal failure (RF), diabetic neuropathy, peripheral arterial disease, reduced visual acuity and all-cause mortality. Kaplan Meier curves were plotted for each derived phenotype. Finally, we tested the performance of an established risk prediction model (QRISK) by adding TDA-derived features. We identified four subgroups of patients with diabetes and divergent comorbidity patterns differing in their risk of future cardiovascular, renal, and other microvascular outcomes. Phenotype 1 (young with chronic inflammatory conditions) and phenotype 2 (young with CAD) included relatively younger patients with diabetes compared to phenotypes 3 (older with hypertension and renal disease) and 4 (older with previous CVA), and those subgroups had a higher frequency of pre-existing cardio-renal diseases. Within ten years of follow-up, 2592 patients (26%) experienced MACE, 2515 patients (25%) died, and 2020 patients (20%) suffered RF. QRISK3 model's AUC was augmented from 67.26% (CI 67.25-67.28%) to 67.67% (CI 67.66-67.69%) by adding specific TDA-derived phenotype and the distances to both extremities of the TDA graph improving its performance in the prediction of CV outcomes. We confirmed the importance of accounting for multimorbidity when risk stratifying heterogenous cohort of patients with new diagnosis of diabetes. Our unsupervised machine learning method improved the prediction of clinical outcomes.

Long-Term Exposure to Road Traffic Noise and Incident Heart Failure: Evidence From UK Biobank.

BACKGROUND: Evidence on road traffic noise and heart failure (HF) is limited, and little is known on the potential mediation roles of acute myocardial infarction (AMI), hypertension, or diabetes. OBJECTIVES: The purpose of this study was to evaluate the impacts of long-term road traffic noise exposure on the risk of incident HF considering air pollution, and explore the mediations of the previously mentioned diseases. METHODS: This prospective study included 424,767 participants without HF at baseline in UK Biobank. The residential-level noise and air pollution exposure was estimated, and the incident HF was identified through linkages with medical records. Cox proportional hazard models were used to estimate HRs. Furthermore, time-dependent mediation was performed. RESULTS: During a median 12.5 years of follow-up, 12,817 incident HF were ascertained. The HRs were 1.08 (95% CI: 1.00-1.16) per 10 dB[A] increase in weighted average 24-hour road traffic noise level (Lden), and 1.15 (95% CI: 1.02-1.31) for exposure to Lden >65 dB[A] compared with the reference category (Lden ≤55 dB[A]), respectively. Furthermore, the strongest combined effects were found in those with both high exposures to road traffic noise and air pollution including fine particles and nitrogen dioxide. Prior AMI before HF within 2 years' time interval mediated 12.5% of the association of road traffic noise with HF. CONCLUSIONS: More attention should be paid and a preventive strategy should be considered to alleviate the disease burden of HF related to road traffic noise exposure, especially in participants who survived AMI and developed HF within 2 years.

Mediation of metabolic syndrome in the association between long-term co-exposure to road traffic noise, air pollution and incident type 2 diabetes.

OBJECTIVES: Recent studies have linked exposure to road traffic noise or air pollution with incident type 2 diabetes (T2D), but investigation on their co-exposure was limited and underlying mechanisms remain unclear. We hypothesized that long-term co-exposure to road traffic noise and air pollution increases the risk of incident T2D via the development of metabolic syndrome (MetS). METHODS: This prospective study included 390,834 participants in UK Biobank. Cumulative risk index (CRI), the health-based weighted levels of multiple exposures, was applied to characterize the co-exposure to 24-hour road traffic noise (Lden), particulate matter with aerodynamic diameter ≤ 2.5 µm (PM2.5), and nitrogen dioxide (NO2). Lden was modeled by the Common Noise Assessment Methods in Europe and air pollutant levels were measured by the Land Use Regression model at participants' residential addresses. Incident T2D was ascertained through linkages to inpatient hospital records. MetS was defined by five (central obesity, triglycerides, HDL cholesterol, glucose, and blood pressure) or six factors (C-reactive protein additionally). Cox proportional hazard models were used to assess the association between environmental exposures and incident T2D, and mediation analyses were applied to investigate the role of MetS. RESULTS: After a median of 10.9 years of follow-up, 13,214 (3.4%) incident T2D cases were ascertained. The exposure to Lden, PM2.5, and NO2, as well as their co-exposure, were significantly associated with an elevated risk of incident T2D, with HRs of 1.03 (95%CI: 1.00, 1.05) per 3.5 dB(A) increase in Lden, 1.05 (95%CI: 1.01, 1.10) per 1.3 µg/m3 increase in PM2.5, 1.07 (95%CI: 1.02, 1.11) per 9.8 µg/m3 increase in NO2, and 1.06 (95%CI: 1.02, 1.09) per interquartile range increase in CRI. MetS significantly mediated 43.5%- 54.7% of the CRI-T2D relationship. CONCLUSIONS: Long-term co-exposure to road traffic noise and air pollution is associated with an elevated risk of incident T2D, which may partly be mediated by MetS.

Systolic blood pressure, chronic obstructive pulmonary disease and cardiovascular risk.

OBJECTIVE: In individuals with complex underlying health problems, the association between systolic blood pressure (SBP) and cardiovascular disease is less well recognised. The association between SBP and risk of cardiovascular events in patients with chronic obstructive pulmonary disease (COPD) was investigated. METHODS AND ANALYSIS: In this cohort study, 39 602 individuals with a diagnosis of COPD aged 55-90 years between 1990 and 2009 were identified from validated electronic health records (EHR) in the UK. The association between SBP and risk of cardiovascular end points (composite of ischaemic heart disease, heart failure, stroke and cardiovascular death) was analysed using a deep learning approach. RESULTS: In the selected cohort (46.5% women, median age 69 years), 10 987 cardiovascular events were observed over a median follow-up period of 3.9 years. The association between SBP and risk of cardiovascular end points was found to be monotonic; the lowest SBP exposure group of <120 mm Hg presented nadir of risk. With respect to reference SBP (between 120 and 129 mm Hg), adjusted risk ratios for the primary outcome were 0.99 (95% CI 0.93 to 1.05) for SBP of <120 mm Hg, 1.02 (0.97 to 1.07) for SBP between 130 and 139 mm Hg, 1.07 (1.01 to 1.12) for SBP between 140 and 149 mm Hg, 1.11 (1.05 to 1.17) for SBP between 150 and 159 mm Hg and 1.16 (1.10 to 1.22) for SBP ≥160 mm Hg. CONCLUSION: Using deep learning for modelling EHR, we identified a monotonic association between SBP and risk of cardiovascular events in patients with COPD.

Deep Perceptual Image Enhancement Network for Exposure Restoration.

Image restoration techniques process degraded images to highlight obscure details or enhance the scene with good contrast and vivid color for the best possible visibility. Poor illumination condition causes issues, such as high-level noise, unlikely color or texture distortions, nonuniform exposure, halo artifacts, and lack of sharpness in the images. This article presents a novel end-to-end trainable deep convolutional neural network called the deep perceptual image enhancement network (DPIENet) to address these challenges. The novel contributions of the proposed work are: 1) a framework to synthesize multiple exposures from a single image and utilizing the exposure variation to restore the image and 2) a loss function based on the approximation of the logarithmic response of the human eye. Extensive computer simulations on the benchmark MIT-Adobe FiveK and user studies performed using Google high dynamic range, DIV2K, and low light image datasets show that DPIENet has clear advantages over state-of-the-art techniques. It has the potential to be useful for many everyday applications such as modernizing traditional camera technologies that currently capture images/videos with under/overexposed regions due to their sensors limitations, to be used in consumer photography to help the users capture appealing images, or for a variety of intelligent systems, including automated driving and video surveillance applications.

Systolic Blood Pressure and Cardiovascular Risk in Patients With Diabetes: A Prospective Cohort Study.

BACKGROUND: Whether the association between systolic blood pressure (SBP) and risk of cardiovascular disease is monotonic or whether there is a nadir of optimal blood pressure remains controversial. We investigated the association between SBP and cardiovascular events in patients with diabetes across the full spectrum of SBP. METHODS: A cohort of 49 000 individuals with diabetes aged 50 to 90 years between 1990 and 2005 was identified from linked electronic health records in the United Kingdom. Associations between SBP and cardiovascular outcomes (ischemic heart disease, heart failure, stroke, and cardiovascular death) were analyzed using a deep learning approach. RESULTS: Over a median follow-up of 7.3 years, 16 378 cardiovascular events were observed. The relationship between SBP and cardiovascular events followed a monotonic pattern, with the group with the lowest baseline SBP of <120 mm Hg exhibiting the lowest risk of cardiovascular events. In comparison to the reference group with the lowest SBP (<120 mm Hg), the adjusted risk ratio for cardiovascular disease was 1.03 (95% CI, 0.97-1.10) for SBP between 120 and 129 mm Hg, 1.05 (0.99-1.11) for SBP between 130 and 139 mm Hg, 1.08 (1.01-1.15) for SBP between 140 and 149 mm Hg, 1.12 (1.03-1.20) for SBP between 150 and 159 mm Hg, and 1.19 (1.09-1.28) for SBP ≥160 mm Hg. CONCLUSIONS: Using deep learning modeling, we found a monotonic relationship between SBP and risk of cardiovascular outcomes in patients with diabetes, without evidence of a J-shaped relationship.

Hi-BEHRT: Hierarchical Transformer-Based Model for Accurate Prediction of Clinical Events Using Multimodal Longitudinal Electronic Health Records.

Electronic health records (EHR) represent a holistic overview of patients' trajectories. Their increasing availability has fueled new hopes to leverage them and develop accurate risk prediction models for a wide range of diseases. Given the complex interrelationships of medical records and patient outcomes, deep learning models have shown clear merits in achieving this goal. However, a key limitation of current study remains their capacity in processing long sequences, and long sequence modelling and its application in the context of healthcare and EHR remains unexplored. Capturing the whole history of medical encounters is expected to lead to more accurate predictions, but the inclusion of records collected for decades and from multiple resources can inevitably exceed the receptive field of the most existing deep learning architectures. This can result in missing crucial, long-term dependencies. To address this gap, we present Hi-BEHRT, a hierarchical Transformer-based model that can significantly expand the receptive field of Transformers and extract associations from much longer sequences. Using a multimodal large-scale linked longitudinal EHR, the Hi-BEHRT exceeds the state-of-the-art deep learning models 1% to 5% for area under the receiver operating characteristic (AUROC) curve and 1% to 8% for area under the precision recall (AUPRC) curve on average, and 2% to 8% (AUROC) and 2% to 11% (AUPRC) for patients with long medical history for 5-year heart failure, diabetes, chronic kidney disease, and stroke risk prediction. Additionally, because pretraining for hierarchical Transformer is not well-established, we provide an effective end-to-end contrastive pre-training strategy for Hi-BEHRT using EHR, improving its transferability on predicting clinical events with relatively small training dataset.

Targeted-BEHRT: Deep Learning for Observational Causal Inference on Longitudinal Electronic Health Records.

Observational causal inference is useful for decision-making in medicine when randomized clinical trials (RCTs) are infeasible or nongeneralizable. However, traditional approaches do not always deliver unconfounded causal conclusions in practice. The rise of "doubly robust" nonparametric tools coupled with the growth of deep learning for capturing rich representations of multimodal data offers a unique opportunity to develop and test such models for causal inference on comprehensive electronic health records (EHRs). In this article, we investigate causal modeling of an RCT-established causal association: the effect of classes of antihypertensive on incident cancer risk. We develop a transformer-based model, targeted bidirectional EHR transformer (T-BEHRT) coupled with doubly robust estimation to estimate average risk ratio (RR). We compare our model to benchmark statistical and deep learning models for causal inference in multiple experiments on semi-synthetic derivations of our dataset with various types and intensities of confounding. In order to further test the reliability of our approach, we test our model on situations of limited data. We find that our model provides more accurate estimates of relative risk least sum absolute error (SAE) from ground truth compared with benchmark estimations. Finally, our model provides an estimate of class-wise antihypertensive effect on cancer risk that is consistent with results derived from RCTs.

An Explainable Transformer-Based Deep Learning Model for the Prediction of Incident Heart Failure.

Predicting the incidence of complex chronic conditions such as heart failure is challenging. Deep learning models applied to rich electronic health records may improve prediction but remain unexplainable hampering their wider use in medical practice. We aimed to develop a deep-learning framework for accurate and yet explainable prediction of 6-month incident heart failure (HF). Using 100,071 patients from longitudinal linked electronic health records across the U.K., we applied a novel Transformer-based risk model using all community and hospital diagnoses and medications contextualized within the age and calendar year for each patient's clinical encounter. Feature importance was investigated with an ablation analysis to compare model performance when alternatively removing features and by comparing the variability of temporal representations. A post-hoc perturbation technique was conducted to propagate the changes in the input to the outcome for feature contribution analyses. Our model achieved 0.93 area under the receiver operator curve and 0.69 area under the precision-recall curve on internal 5-fold cross validation and outperformed existing deep learning models. Ablation analysis indicated medication is important for predicting HF risk, calendar year is more important than chronological age, which was further reinforced by temporal variability analysis. Contribution analyses identified risk factors that are closely related to HF. Many of them were consistent with existing knowledge from clinical and epidemiological research but several new associations were revealed which had not been considered in expert-driven risk prediction models. In conclusion, the results highlight that our deep learning model, in addition high predictive performance, can inform data-driven risk factor identification.

Tufts Dental Database: A Multimodal Panoramic X-Ray Dataset for Benchmarking Diagnostic Systems.

The application of Artificial Intelligence in dental healthcare has a very promising role due to the abundance of imagery and non-imagery-based clinical data. Expert analysis of dental radiographs can provide crucial information for clinical diagnosis and treatment. In recent years, Convolutional Neural Networks have achieved the highest accuracy in various benchmarks, including analyzing dental X-ray images to improve clinical care quality. The Tufts Dental Database, a new X-ray panoramic radiography image dataset, has been presented in this paper. This dataset consists of 1000 panoramic dental radiography images with expert labeling of abnormalities and teeth. The classification of radiography images was performed based on five different levels: anatomical location, peripheral characteristics, radiodensity, effects on the surrounding structure, and the abnormality category. This first-of-its-kind multimodal dataset also includes the radiologist's expertise captured in the form of eye-tracking and think-aloud protocol. The contributions of this work are 1) publicly available dataset that can help researchers to incorporate human expertise into AI and achieve more robust and accurate abnormality detection; 2) a benchmark performance analysis for various state-of-the-art systems for dental radiograph image enhancement and image segmentation using deep learning; 3) an in-depth review of various panoramic dental image datasets, along with segmentation and detection systems. The release of this dataset aims to propel the development of AI-powered automated abnormality detection and classification in dental panoramic radiographs, enhance tooth segmentation algorithms, and the ability to distill the radiologist's expertise into AI.

Descriptive analysis of Acute Ischemic stroke in COVID-19 patients through the course of the COVID-19 pandemic.

Coronavirus disease 2019 (COVID-19) has been associated with Acute Ischemic Stroke (AIS). Here, we characterize our institutional experience with management of COVID-19 and AIS. Baseline demographics, clinical, imaging, and outcomes data were determined in patients with COVID-19 and AIS presenting within March 2020 to October 2020, and November 2020 to August 2021, based on institutional COVID-19 hospitalization volume. Of 2512 COVID-19 patients, 35 (1.39%, mean age 63.3 years, 54% women) had AIS. AIS recognition was frequently delayed after COVID-19 symptoms (median 19.5 days). Four patients (11%) were on therapeutic anticoagulation at AIS recognition. AIS mechanism was undetermined or due to multiple etiologies in most cases (n = 20, 57%). Three patients underwent IV TPA, and three underwent mechanical thrombectomy, of which two suffered re-occlusion. Three patients had incomplete mRNA vaccination course. Fourteen (40%) died, with 26 (74%) having poor outcomes. Critical COVID-19 severity was associated with worsened mortality (p = 0.02). More patients (12/16; 75%) had either worsened or similar 3-month functional outcomes, than those with improvement, indicating the devastating impact of co-existing AIS and COVID-19. Comparative analysis showed that patients in the later cohort had earlier AIS presentation, fewer stroke risk factors, more comprehensive workup, more defined stroke mechanisms, less instance of critical COVID-19 severity, more utilization of IV TPA, and a trend towards worse outcomes for the sub-group of mild-to-moderate COVID-19 severity. AIS incidence, NIHSS, and overall outcomes were similar. Further studies should investigate outcomes beyond 3 months and their predictive factors, impact of completed vaccination course, and access to neurologic care.

Validation of risk prediction models applied to longitudinal electronic health record data for the prediction of major cardiovascular events in the presence of data shifts.

Aims: Deep learning has dominated predictive modelling across different fields, but in medicine it has been met with mixed reception. In clinical practice, simple, statistical models and risk scores continue to inform cardiovascular disease risk predictions. This is due in part to the knowledge gap about how deep learning models perform in practice when they are subject to dynamic data shifts; a key criterion that common internal validation procedures do not address. We evaluated the performance of a novel deep learning model, BEHRT, under data shifts and compared it with several ML-based and established risk models. Methods and results: Using linked electronic health records of 1.1 million patients across England aged at least 35 years between 1985 and 2015, we replicated three established statistical models for predicting 5-year risk of incident heart failure, stroke, and coronary heart disease. The results were compared with a widely accepted machine learning model (random forests), and a novel deep learning model (BEHRT). In addition to internal validation, we investigated how data shifts affect model discrimination and calibration. To this end, we tested the models on cohorts from (i) distinct geographical regions; (ii) different periods. Using internal validation, the deep learning models substantially outperformed the best statistical models by 6%, 8%, and 11% in heart failure, stroke, and coronary heart disease, respectively, in terms of the area under the receiver operating characteristic curve. Conclusion: The performance of all models declined as a result of data shifts; despite this, the deep learning models maintained the best performance in all risk prediction tasks. Updating the model with the latest information can improve discrimination but if the prior distribution changes, the model may remain miscalibrated.

Deep Bayesian Gaussian processes for uncertainty estimation in electronic health records.

One major impediment to the wider use of deep learning for clinical decision making is the difficulty of assigning a level of confidence to model predictions. Currently, deep Bayesian neural networks and sparse Gaussian processes are the main two scalable uncertainty estimation methods. However, deep Bayesian neural networks suffer from lack of expressiveness, and more expressive models such as deep kernel learning, which is an extension of sparse Gaussian process, captures only the uncertainty from the higher-level latent space. Therefore, the deep learning model under it lacks interpretability and ignores uncertainty from the raw data. In this paper, we merge features of the deep Bayesian learning framework with deep kernel learning to leverage the strengths of both methods for a more comprehensive uncertainty estimation. Through a series of experiments on predicting the first incidence of heart failure, diabetes and depression applied to large-scale electronic medical records, we demonstrate that our method is better at capturing uncertainty than both Gaussian processes and deep Bayesian neural networks in terms of indicating data insufficiency and identifying misclassifications, with a comparable generalization performance. Furthermore, by assessing the accuracy and area under the receiver operating characteristic curve over the predictive probability, we show that our method is less susceptible to making overconfident predictions, especially for the minority class in imbalanced datasets. Finally, we demonstrate how uncertainty information derived by the model can inform risk factor analysis towards model interpretability.

Association between cardiometabolic disease multimorbidity and all-cause mortality in 2 million women and men registered in UK general practices.

BACKGROUND: Myocardial infarction (MI), stroke and diabetes share underlying risk factors and commonalities in clinical management. We examined if their combined impact on mortality is proportional, amplified or less than the expected risk separately of each disease and whether the excess risk is explained by their associated comorbidities. METHODS: Using large-scale electronic health records, we identified 2,007,731 eligible patients (51% women) and registered with general practices in the UK and extracted clinical information including diagnosis of myocardial infarction (MI), stroke, diabetes and 53 other long-term conditions before 2005 (study baseline). We used Cox regression to determine the risk of all-cause mortality with age as the underlying time variable and tested for excess risk due to interaction between cardiometabolic conditions. RESULTS: At baseline, the mean age was 51 years, and 7% (N = 145,910) have had a cardiometabolic condition. After a 7-year mean follow-up, 146,994 died. The sex-adjusted hazard ratios (HR) (95% confidence interval [CI]) of all-cause mortality by baseline disease status, compared to those without cardiometabolic disease, were MI = 1.51 (1.49-1.52), diabetes = 1.52 (1.51-1.53), stroke = 1.84 (1.82-1.86), MI and diabetes = 2.14 (2.11-2.17), MI and stroke = 2.35 (2.30-2.39), diabetes and stroke = 2.53 (2.50-2.57) and all three = 3.22 (3.15-3.30). Adjusting for other concurrent comorbidities attenuated these estimates, including the risk associated with having all three conditions (HR = 1.81 [95% CI 1.74-1.89]). Excess risks due to interaction between cardiometabolic conditions, particularly when all three conditions were present, were not significantly greater than expected from the individual disease effects. CONCLUSION: Myocardial infarction, stroke and diabetes were associated with excess mortality, without evidence of any amplification of risk in people with all three diseases. The presence of other comorbidities substantially contributed to the excess mortality risks associated with cardiometabolic disease multimorbidity.

Perinatal health predictors using artificial intelligence: A review.

Advances in public health and medical care have enabled better pregnancy and birth outcomes. The rates of perinatal health indicators such as maternal mortality and morbidity; fetal, neonatal, and infant mortality; low birthweight; and preterm birth have reduced over time. However, they are still a public health concern, and considerable disparities exist within and between countries. For perinatal researchers who are engaged in unraveling the tangled web of causation for maternal and child health outcomes and for clinicians involved in the care of pregnant women and infants, artificial intelligence offers novel approaches to prediction modeling, diagnosis, early detection, and monitoring in perinatal health. Machine learning, a commonly used artificial intelligence method, has been used to predict preterm birth, birthweight, preeclampsia, mortality, hypertensive disorders, and postpartum depression. Real-time electronic health recording and predictive modeling using artificial intelligence have found early success in fetal monitoring and monitoring of women with gestational diabetes especially in low-resource settings. Artificial intelligence-based methodologies have the potential to improve prenatal diagnosis of birth defects and outcomes in assisted reproductive technology too. In this scenario, we envision artificial intelligence for perinatal research to be based on three goals: (1) availability of population-representative, routine clinical data (rich multimodal data of large sample size) for perinatal research; (2) modification and application of current state-of-the-art artificial intelligence for prediction and classification in health care research to the field of perinatal health; and (3) development of methods for explaining the decision-making processes of artificial intelligence models for perinatal health indicators. Achieving these three goals via a multidisciplinary approach to the development of artificial intelligence tools will enable trust in these tools and advance research, clinical practice, and policies to ensure optimal perinatal health.