Foresight-a generative pretrained transformer for modelling of patient timelines using electronic health records: a retrospective modelling study.

BACKGROUND: An electronic health record (EHR) holds detailed longitudinal information about a patient's health status and general clinical history, a large portion of which is stored as unstructured, free text. Existing approaches to model a patient's trajectory focus mostly on structured data and a subset of single-domain outcomes. This study aims to evaluate the effectiveness of Foresight, a generative transformer in temporal modelling of patient data, integrating both free text and structured formats, to predict a diverse array of future medical outcomes, such as disorders, substances (eg, to do with medicines, allergies, or poisonings), procedures, and findings (eg, relating to observations, judgements, or assessments). METHODS: Foresight is a novel transformer-based pipeline that uses named entity recognition and linking tools to convert EHR document text into structured, coded concepts, followed by providing probabilistic forecasts for future medical events, such as disorders, substances, procedures, and findings. The Foresight pipeline has four main components: (1) CogStack (data retrieval and preprocessing); (2) the Medical Concept Annotation Toolkit (structuring of the free-text information from EHRs); (3) Foresight Core (deep-learning model for biomedical concept modelling); and (4) the Foresight web application. We processed the entire free-text portion from three different hospital datasets (King's College Hospital [KCH], South London and Maudsley [SLaM], and the US Medical Information Mart for Intensive Care III [MIMIC-III]), resulting in information from 811 336 patients and covering both physical and mental health institutions. We measured the performance of models using custom metrics derived from precision and recall. FINDINGS: Foresight achieved a precision@10 (ie, of 10 forecasted candidates, at least one is correct) of 0·68 (SD 0·0027) for the KCH dataset, 0·76 (0·0032) for the SLaM dataset, and 0·88 (0·0018) for the MIMIC-III dataset, for forecasting the next new disorder in a patient timeline. Foresight also achieved a precision@10 value of 0·80 (0·0013) for the KCH dataset, 0·81 (0·0026) for the SLaM dataset, and 0·91 (0·0011) for the MIMIC-III dataset, for forecasting the next new biomedical concept. In addition, Foresight was validated on 34 synthetic patient timelines by five clinicians and achieved a relevancy of 33 (97% [95% CI 91-100]) of 34 for the top forecasted candidate disorder. As a generative model, Foresight can forecast follow-on biomedical concepts for as many steps as required. INTERPRETATION: Foresight is a general-purpose model for biomedical concept modelling that can be used for real-world risk forecasting, virtual trials, and clinical research to study the progression of disorders, to simulate interventions and counterfactuals, and for educational purposes. FUNDING: National Health Service Artificial Intelligence Laboratory, National Institute for Health and Care Research Biomedical Research Centre, and Health Data Research UK.

Artificial intelligence methods for improved detection of undiagnosed heart failure with preserved ejection fraction.

AIM: Heart failure with preserved ejection fraction (HFpEF) remains under-diagnosed in clinical practice despite accounting for nearly half of all heart failure (HF) cases. Accurate and timely diagnosis of HFpEF is crucial for proper patient management and treatment. In this study, we explored the potential of natural language processing (NLP) to improve the detection and diagnosis of HFpEF according to the European Society of Cardiology (ESC) diagnostic criteria. METHODS AND RESULTS: In a retrospective cohort study, we used an NLP pipeline applied to the electronic health record (EHR) to identify patients with a clinical diagnosis of HF between 2010 and 2022. We collected demographic, clinical, echocardiographic and outcome data from the EHR. Patients were categorized according to the left ventricular ejection fraction (LVEF). Those with LVEF ≥50% were further categorized based on whether they had a clinician-assigned diagnosis of HFpEF and if not, whether they met the ESC diagnostic criteria. Results were validated in a second, independent centre. We identified 8606 patients with HF. Of 3727 consecutive patients with HF and LVEF ≥50% on echocardiogram, only 8.3% had a clinician-assigned diagnosis of HFpEF, while 75.4% met ESC criteria but did not have a formal diagnosis of HFpEF. Patients with confirmed HFpEF were hospitalized more frequently; however the ESC criteria group had a higher 5-year mortality, despite being less comorbid and experiencing fewer acute cardiovascular events. CONCLUSIONS: This study demonstrates that patients with undiagnosed HFpEF are an at-risk group with high mortality. It is possible to use NLP methods to identify likely HFpEF patients from EHR data who would likely then benefit from expert clinical review and complement the use of diagnostic algorithms.

Problematic Smartphone Usage in Singaporean University Students: An Analysis of Self-Reported Versus Objectively Measured Smartphone Usage Patterns.

INTRODUCTION: Problematic smartphone usage is the excessive usage of the smartphone, leading to addiction symptoms that impair one's functional status. Self-administered surveys developed to describe the symptoms and measure the risk of problematic smartphone usage have been associated with depressive symptoms, symptoms of anxiety disorder, and perceived stress. However, self-reported smartphone usage can be unreliable, and previous studies have identified a better association between objectively measured smartphone usage and problematic smartphone usage. METHODOLOGY: A self-administered survey was used to investigate the relationships between the risk of problematic smartphone usage (SAS-SV) with depressive symptoms (PHQ-9), anxiety disorder symptoms (GAD-7), and perceived stress (PSS) in Singaporean full-time university students. Self-reported screentime and objectively measured screentime were collected to determine if there is any difference between perceived smartphone usage and objective smartphone usage. RESULTS: There was no statistical difference between self-reported and app-measured screentime in the study population. However, there were significant positive correlations between SAS-SV with PHQ-9, GAD-7, and PSS. In the logistic regression model, PHQ-9 was found to be the sole predictor for variances in SAS-SV score in the study population. CONCLUSION: This study suggests that problematic smartphone usage may potentially related to depressive symptoms, symptoms of anxiety disorder, and greater perceived stress in university students.

Artificial intelligence (AI) for neurologists: do digital neurones dream of electric sheep?

Artificial intelligence (AI) is routinely mentioned in journals and newspapers, and non-technical outsiders may have difficulty in distinguishing hyperbole from reality. We present a practical guide to help non-technical neurologists to understand healthcare AI. AI is being used to support clinical decisions in treating neurological disorders. We introduce basic concepts of AI, such as machine learning and natural language processing, and explain how AI is being used in healthcare, giving examples its benefits and challenges. We also cover how AI performance is measured, and its regulatory aspects in healthcare. An important theme is that AI is a general-purpose technology like medical statistics, with broad utility applicable in various scenarios, such that niche approaches are outpaced by approaches that are broadly applicable in many disease areas and specialties. By understanding AI basics and its potential applications, neurologists can make informed decisions when evaluating AI used in their clinical practice. This article was written by four humans, with generative AI helping with formatting and image generation.

Quantifying digital health inequality across a national healthcare system.

OBJECTIVES: Digital health inequality, observed as differential utilisation of digital tools between population groups, has not previously been quantified in the National Health Service (NHS). Deployment of universal digital health interventions, including a national smartphone app and online primary care services, allows measurement of digital inequality across a nation. We aimed to measure population factors associated with digital utilisation across 6356 primary care providers serving the population of England. METHODS: We used multivariable regression to test association of population and provider characteristics (including patient demographics, socioeconomic deprivation, disease burden, prescribing burden, geography and healthcare provider resource) with activation of two independent digital services during 2021/2022. RESULTS: We find a significant adjusted association between increased population deprivation and reduced digital utilisation across both interventions. Multivariable regression coefficients for most deprived quintiles correspond to 4.27 million patients across England where deprivation is associated with non-activation of the NHS App. CONCLUSION: Results are concerning for technologically driven widening of healthcare inequalities. Targeted incentive to digital is necessary to prevent digital disparity from becoming health outcomes disparity.

Federated learning for medical imaging radiology.

Federated learning (FL) is gaining wide acceptance across the medical AI domains. FL promises to provide a fairly acceptable clinical-grade accuracy, privacy, and generalisability of machine learning models across multiple institutions. However, the research on FL for medical imaging AI is still in its early stages. This paper presents a review of recent research to outline the difference between state-of-the-art [SOTA] (published literature) and state-of-the-practice [SOTP] (applied research in realistic clinical environments). Furthermore, the review outlines the future research directions considering various factors such as data, learning models, system design, governance, and human-in-loop to translate the SOTA into SOTP and effectively collaborate across multiple institutions.

Latent Transformer Models for out-of-distribution detection.

Any clinically-deployed image-processing pipeline must be robust to the full range of inputs it may be presented with. One popular approach to this challenge is to develop predictive models that can provide a measure of their uncertainty. Another approach is to use generative modelling to quantify the likelihood of inputs. Inputs with a low enough likelihood are deemed to be out-of-distribution and are not presented to the downstream predictive model. In this work, we evaluate several approaches to segmentation with uncertainty for the task of segmenting bleeds in 3D CT of the head. We show that these models can fail catastrophically when operating in the far out-of-distribution domain, often providing predictions that are both highly confident and wrong. We propose to instead perform out-of-distribution detection using the Latent Transformer Model: a VQ-GAN is used to provide a highly compressed latent representation of the input volume, and a transformer is then used to estimate the likelihood of this compressed representation of the input. We demonstrate this approach can identify images that are both far- and near- out-of-distribution, as well as provide spatial maps that highlight the regions considered to be out-of-distribution. Furthermore, we find a strong relationship between an image's likelihood and the quality of a model's segmentation on it, demonstrating that this approach is viable for filtering out unsuitable images.

Mapping and evaluating national data flows: transparency, privacy, and guiding infrastructural transformation.

The importance of big health data is recognised worldwide. Most UK National Health Service (NHS) care interactions are recorded in electronic health records, resulting in an unmatched potential for population-level datasets. However, policy reviews have highlighted challenges from a complex data-sharing landscape relating to transparency, privacy, and analysis capabilities. In response, we used public information sources to map all electronic patient data flows across England, from providers to more than 460 subsequent academic, commercial, and public data consumers. Although NHS data support a global research ecosystem, we found that multistage data flow chains limit transparency and risk public trust, most data interactions do not fulfil recommended best practices for safe data access, and existing infrastructure produces aggregation of duplicate data assets, thus limiting diversity of data and added value to end users. We provide recommendations to support data infrastructure transformation and have produced a website (https://DataInsights.uk) to promote transparency and showcase NHS data assets.

Streams, rivers and data lakes: an introduction to understanding modern electronic healthcare records.

As foundation doctors, we have often found ourselves informing patients that a certain aspect of their medical information cannot be immediately found, either because it is on an electronic system we cannot access, or it is in a hospital that is unlinked to our own. Unsurprisingly, this frequently leaves patients flabbergasted and confused. We started to wonder: if patients' data are entered onto an electronic system: where do those data go? If medical data are searched for, where do those data come from? Why are there so many hidden sources of information that clinicians cannot access? In an ever-increasing digital sphere, electronic data will be the future of holistic health and social care planning, impacting every clinician's day-to-day role. From electronic healthcare records to the use of artificial intelligence solutions, this article will serve as an introduction to how data flows in modern healthcare systems.

AI chatbots not yet ready for clinical use.

As large language models (LLMs) expand and become more advanced, so do the natural language processing capabilities of conversational AI, or "chatbots". OpenAI's recent release, ChatGPT, uses a transformer-based model to enable human-like text generation and question-answering on general domain knowledge, while a healthcare-specific Large Language Model (LLM) such as GatorTron has focused on the real-world healthcare domain knowledge. As LLMs advance to achieve near human-level performances on medical question and answering benchmarks, it is probable that Conversational AI will soon be developed for use in healthcare. In this article we discuss the potential and compare the performance of two different approaches to generative pretrained transformers-ChatGPT, the most widely used general conversational LLM, and Foresight, a GPT (generative pretrained transformer) based model focused on modelling patients and disorders. The comparison is conducted on the task of forecasting relevant diagnoses based on clinical vignettes. We also discuss important considerations and limitations of transformer-based chatbots for clinical use.

Hospital-wide natural language processing summarising the health data of 1 million patients.

Electronic health records (EHRs) represent a major repository of real world clinical trajectories, interventions and outcomes. While modern enterprise EHR's try to capture data in structured standardised formats, a significant bulk of the available information captured in the EHR is still recorded only in unstructured text format and can only be transformed into structured codes by manual processes. Recently, Natural Language Processing (NLP) algorithms have reached a level of performance suitable for large scale and accurate information extraction from clinical text. Here we describe the application of open-source named-entity-recognition and linkage (NER+L) methods (CogStack, MedCAT) to the entire text content of a large UK hospital trust (King's College Hospital, London). The resulting dataset contains 157M SNOMED concepts generated from 9.5M documents for 1.07M patients over a period of 9 years. We present a summary of prevalence and disease onset as well as a patient embedding that captures major comorbidity patterns at scale. NLP has the potential to transform the health data lifecycle, through large-scale automation of a traditionally manual task.

Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models.

Brief Hospital Course (BHC) summaries are succinct summaries of an entire hospital encounter, embedded within discharge summaries, written by senior clinicians responsible for the overall care of a patient. Methods to automatically produce summaries from inpatient documentation would be invaluable in reducing clinician manual burden of summarising documents under high time-pressure to admit and discharge patients. Automatically producing these summaries from the inpatient course, is a complex, multi-document summarisation task, as source notes are written from various perspectives (e.g. nursing, doctor, radiology), during the course of the hospitalisation. We demonstrate a range of methods for BHC summarisation demonstrating the performance of deep learning summarisation models across extractive and abstractive summarisation scenarios. We also test a novel ensemble extractive and abstractive summarisation model that incorporates a medical concept ontology (SNOMED) as a clinical guidance signal and shows superior performance in 2 real-world clinical data sets.

Diagnostic signature for heart failure with preserved ejection fraction (HFpEF): a machine learning approach using multi-modality electronic health record data.

BACKGROUND: Heart failure with preserved ejection fraction (HFpEF) is thought to be highly prevalent yet remains underdiagnosed. Evidence-based treatments are available that increase quality of life and decrease hospitalization. We sought to develop a data-driven diagnostic model to predict from electronic health records (EHR) the likelihood of HFpEF among patients with unexplained dyspnea and preserved left ventricular EF. METHODS AND RESULTS: The derivation cohort comprised patients with dyspnea and echocardiography results. Structured and unstructured data were extracted using an automated informatics pipeline. Patients were retrospectively diagnosed as HFpEF (cases), non-HF (control cohort I), or HF with reduced EF (HFrEF; control cohort II). The ability of clinical parameters and investigations to discriminate cases from controls was evaluated by extreme gradient boosting. A likelihood scoring system was developed and validated in a separate test cohort. The derivation cohort included 1585 consecutive patients: 133 cases of HFpEF (9%), 194 non-HF cases (Control cohort I) and 1258 HFrEF cases (Control cohort II). Two HFpEF diagnostic signatures were derived, comprising symptoms, diagnoses and investigation results. A final prediction model was generated based on the averaged likelihood scores from these two models. In a validation cohort consisting of 269 consecutive patients [with 66 HFpEF cases (24.5%)], the diagnostic power of detecting HFpEF had an AUROC of 90% (P < 0.001) and average precision of 74%. CONCLUSION: This diagnostic signature enables discrimination of HFpEF from non-cardiac dyspnea or HFrEF from EHR and can assist in the diagnostic evaluation in patients with unexplained dyspnea. This approach will enable identification of HFpEF patients who may then benefit from new evidence-based therapies.

Moving towards vertically integrated artificial intelligence development.

Substantial interest and investment in clinical artificial intelligence (AI) research has not resulted in widespread translation to deployed AI solutions. Current attention has focused on bias and explainability in AI algorithm development, external validity and model generalisability, and lack of equity and representation in existing data. While of great importance, these considerations also reflect a model-centric approach seen in published clinical AI research, which focuses on optimising architecture and performance of an AI model on best available datasets. However, even robustly built models using state-of-the-art algorithms may fail once tested in realistic environments due to unpredictability of real-world conditions, out-of-dataset scenarios, characteristics of deployment infrastructure, and lack of added value to clinical workflows relative to cost and potential clinical risks. In this perspective, we define a vertically integrated approach to AI development that incorporates early, cross-disciplinary, consideration of impact evaluation, data lifecycles, and AI production, and explore its implementation in two contrasting AI development pipelines: a scalable "AI factory" (Mayo Clinic, Rochester, United States), and an end-to-end cervical cancer screening platform for resource poor settings (Paps AI, Mbarara, Uganda). We provide practical recommendations for implementers, and discuss future challenges and novel approaches (including a decentralised federated architecture being developed in the NHS (AI4VBH, London, UK)). Growth in global clinical AI research continues unabated, and introduction of vertically integrated teams and development practices can increase the translational potential of future clinical AI projects.

Anticoagulation for atrial fibrillation in people with serious mental illness in the general hospital setting.

OBJECTIVE: People with serious mental illnesses (SMI) have an increased risk of stroke compared to the general population. This study aims to evaluate oral anticoagulation prescription trends in atrial fibrillation (AF) patients with and without a comorbid SMI. METHODS: An open-source retrieval system for clinical data (CogStack) was used to identify a cohort of AF patients with SMI who ever had an inpatient admission to King's College Hospital from 2011 to 2020. A Natural Language Processing pipeline was used to calculate CHA2DS2-VASc and HASBLED risk scores from Electronic Health Records free text. Antithrombotic prescriptions of warfarin and Direct acting oral anti-coagulants (DOACs) (apixaban, rivaroxaban, dabigatran, edoxaban) were extracted from discharge summaries. RESULTS: Among patients included in the study (n = 16 916), 2.7% had a recorded co-morbid SMI diagnosis. Compared to non-SMI patients, those with SMI had significantly higher CHA2DS2-VASc (mean (SD): 5.3 (1.96) vs 4.7 (2.08), p < 0.001) and HASBLED scores (mean (SD): 3.2 (1.27) vs 2.5 (1.29), p < 0.001). Among AF patients having a CHA2DS2-VASc ≥2, those with co-morbid SMI were less likely than non-SMI patients to be prescribed an OAC (44% vs 54%, p < 0.001). However, there was no evidence of a significant difference between the two groups since 2019. CONCLUSION: Over recent years, DOAC prescription rates have increased among AF patients with SMI in acute hospitals. More research is needed to confirm whether the introduction of DOACs has reduced OAC treatment gaps in people with serious mental illness and to assess whether the use of DOACs has improved health outcomes in this population.

Interaction Between Race, Ethnicity, Severe Mental Illness, and Cardiovascular Disease.

Severe mental illnesses, such as schizophrenia or bipolar disorder, affect ≈1% of the population who, as a group, experience significant disadvantage in terms of physical health and reduced life expectancy. In this review, we explore the interaction between race, ethnicity, severe mental illness, and cardiovascular disease, with a focus on cardiovascular care pathways. Finally, we discuss strategies to investigate and address disparities in cardiovascular care for patients with severe mental illness.

Improving stroke pathways using an adhesive ambulatory ECG patch: reducing time for patients to ECGs and subsequent results.

Three south-London hospital trusts undertook a feasibility study, comparing data from 93 patients who received the 14-day adhesive ambulatory electrocardiography (ECG) patch Zio XT with retrospective data from 125 patients referred for 24-hour Holter for cryptogenic stroke and transient ischaemic attack following negative 12-lead ECG. As the ECG patch was fitted the same day as the clinical decision for ambulatory ECG monitoring was made, median time to the patient having the monitor fitted was significantly reduced in all three hospital trusts compared with 24-hour Holter being ordered and fitted. Hospital visits reduced by a median of two for patients receiving Zio XT. This project supports that it is feasible to use a patch as part of routine clinical care with a positive impact on care pathways.

A Critical Investigation of Cerebellar Associative Learning in Isolated Dystonia.

BACKGROUND: Impaired eyeblink conditioning is often cited as evidence for cerebellar dysfunction in isolated dystonia yet the results from individual studies are conflicting and underpowered. OBJECTIVE: To systematically examine the influence of dystonia, dystonia subtype, and clinical features over eyeblink conditioning within a statistical model which controlled for the covariates age and sex. METHODS: Original neurophysiological data from all published studies (until 2019) were shared and compared to an age- and sex-matched control group. Two raters blinded to participant identity rescored all recordings (6732 trials). After higher inter-rater agreement was confirmed, mean conditioning per block across raters was entered into a mixed repetitive measures model. RESULTS: Isolated dystonia (P = 0.517) and the subtypes of isolated dystonia (cervical dystonia, DYT-TOR1A, DYT-THAP1, and focal hand dystonia) had similar levels of eyeblink conditioning relative to controls. The presence of tremor did not significantly influence levels of eyeblink conditioning. A large range of eyeblink conditioning behavior was seen in both health and dystonia and sample size estimates are provided for future studies. CONCLUSIONS: The similarity of eyeblink conditioning behavior in dystonia and controls is against a global cerebellar learning deficit in isolated dystonia. Precise mechanisms for how the cerebellum interplays mechanistically with other key neuroanatomical nodes within the dystonic network remains an open research question. © 2022 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson Movement Disorder Society.