Deep representation learning of electronic health records to unlock patient stratification at scale

An open-source framework for end-to-end analysis of electronic health record data

With progressive digitalization of healthcare systems worldwide, large-scale collection of electronic health records (EHRs) has become commonplace. However, an extensible framework for comprehensive exploratory analysis that accounts for data heterogeneity is missing. Here we introduce ehrapy, a modular open-source Python framework designed for exploratory analysis of heterogeneous epidemiology and EHR data. ehrapy incorporates a series of analytical steps, from data extraction and quality control to the generation of low-dimensional representations. Complemented by rich statistical modules, ehrapy facilitates associating patients with disease states, differential comparison between patient clusters, survival analysis, trajectory inference, causal inference and more. Leveraging ontologies, ehrapy further enables data sharing and training EHR deep learning models, paving the way for foundational models in biomedical research. We demonstrate ehrapy's features in six distinct examples. We applied ehrapy to stratify patients affected by unspecified pneumonia into finer-grained phenotypes. Furthermore, we reveal biomarkers for significant differences in survival among these groups. Additionally, we quantify medication-class effects of pneumonia medications on length of stay. We further leveraged ehrapy to analyze cardiovascular risks across different data modalities. We reconstructed disease state trajectories in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) based on imaging data. Finally, we conducted a case study to demonstrate how ehrapy can detect and mitigate biases in EHR data. ehrapy, thus, provides a framework that we envision will standardize analysis pipelines on EHR data and serve as a cornerstone for the community.

MyThisYourThat for interpretable identification of systematic bias in federated learning for biomedical images

Distributed collaborative learning is a promising approach for building predictive models for privacy-sensitive biomedical images. Here, several data owners (clients) train a joint model without sharing their original data. However, concealed systematic biases can compromise model performance and fairness. This study presents MyThisYourThat (MyTH) approach, which adapts an interpretable prototypical part learning network to a distributed setting, enabling each client to visualize feature differences learned by others on their own image: comparing one client's 'This' with others' 'That'. Our setting demonstrates four clients collaboratively training two diagnostic classifiers on a benchmark X-ray dataset. Without data bias, the global model reaches 74.14% balanced accuracy for cardiomegaly and 74.08% for pleural effusion. We show that with systematic visual bias in one client, the performance of global models drops to near-random. We demonstrate how differences between local and global prototypes reveal biases and allow their visualization on each client's data without compromising privacy.

Integrating Digital Health Solutions with Immunization Strategies: Improving Immunization Coverage and Monitoring in the Post-COVID-19 Era

The COVID-19 pandemic underscored the critical importance of vaccination to global health security and highlighted the potential of digital health solutions to improve immunization strategies. This article explores integrating digital health technologies with immunization programs to improve coverage, monitoring, and public health outcomes. It examines the current landscape of digital tools used in immunization initiatives, such as mobile health apps, electronic health records, and data analytics platforms. Case studies from different regions demonstrate the effectiveness of these technologies in addressing challenges such as vaccine hesitancy, logistics, and real-time monitoring of vaccine distribution and adverse events. The paper also examines ethical considerations, data privacy issues, and the need for a robust digital infrastructure to support these innovations. By analyzing the successes and limitations of digital health interventions in immunization campaigns during and after the COVID-19 pandemic, we provide recommendations for future integration strategies to ensure resilient and responsive immunization systems. This research aims to guide policymakers, health professionals, and technologists in leveraging digital health to strengthen immunization efforts and prepare for future public health emergencies.

Autoencoder to Identify Sex-Specific Sub-phenotypes in Alzheimer's Disease Progression Using Longitudinal Electronic Health Records

Alzheimer's Disease (AD) is a complex neurodegenerative disorder significantly influenced by sex differences, with approximately two-thirds of AD patients being women. Characterizing the sex-specific AD progression and identifying its progression trajectory is a crucial step to developing effective risk stratification and prevention strategies. In this study, we developed an autoencoder to uncover sex-specific sub-phenotypes in AD progression leveraging longitudinal electronic health record (EHR) data from OneFlorida+ Clinical Research Consortium. Specifically, we first constructed temporal patient representation using longitudinal EHRs from a sex-stratified AD cohort. We used a long short-term memory (LSTM)-based autoencoder to extract and generate latent representation embeddings from sequential clinical records of patients. We then applied hierarchical agglomerative clustering to the learned representations, grouping patients based on their progression sub-phenotypes. The experimental results show we successfully identified five primary sex-based AD sub-phenotypes with corresponding progression pathways with high confidence. These sex-specific sub-phenotypes not only illustrated distinct AD progression patterns but also revealed differences in clinical characteristics and comorbidities between females and males in AD development. These findings could provide valuable insights for advancing personalized AD intervention and treatment strategies.

Harnessing EHR data for health research

With the increasing availability of rich, longitudinal, real-world clinical data recorded in electronic health records (EHRs) for millions of patients, there is a growing interest in leveraging these records to improve the understanding of human health and disease and translate these insights into clinical applications. However, there is also a need to consider the limitations of these data due to various biases and to understand the impact of missing information. Recognizing and addressing these limitations can inform the design and interpretation of EHR-based informatics studies that avoid confusing or incorrect conclusions, particularly when applied to population or precision medicine. Here we discuss key considerations in the design, implementation and interpretation of EHR-based informatics studies, drawing from examples in the literature across hypothesis generation, hypothesis testing and machine learning applications. We outline the growing opportunities for EHR-based informatics studies, including association studies and predictive modeling, enabled by evolving AI capabilities-while addressing limitations and potential pitfalls to avoid.

Drop the shortcuts: image augmentation improves fairness and decreases AI detection of race and other demographics from medical images

BACKGROUND

It has been shown that AI models can learn race on medical images, leading to algorithmic bias. Our aim in this study was to enhance the fairness of medical image models by eliminating bias related to race, age, and sex. We hypothesise models may be learning demographics via shortcut learning and combat this using image augmentation.

METHODS

This study included 44,953 patients who identified as Asian, Black, or White (mean age, 60.68 years ±18.21; 23,499 women) for a total of 194,359 chest X-rays (CXRs) from MIMIC-CXR database. The included CheXpert images comprised 45,095 patients (mean age 63.10 years ±18.14; 20,437 women) for a total of 134,300 CXRs were used for external validation. We also collected 1195 3D brain magnetic resonance imaging (MRI) data from the ADNI database, which included 273 participants with an average age of 76.97 years ±14.22, and 142 females. DL models were trained on either non-augmented or augmented images and assessed using disparity metrics. The features learned by the models were analysed using task transfer experiments and model visualisation techniques.

FINDINGS

In the detection of radiological findings, training a model using augmented CXR images was shown to reduce disparities in error rate among racial groups (-5.45%), age groups (-13.94%), and sex (-22.22%). For AD detection, the model trained with augmented MRI images was shown 53.11% and 31.01% reduction of disparities in error rate among age and sex groups, respectively. Image augmentation led to a reduction in the model's ability to identify demographic attributes and resulted in the model trained for clinical purposes incorporating fewer demographic features.

INTERPRETATION

The model trained using the augmented images was less likely to be influenced by demographic information in detecting image labels. These results demonstrate that the proposed augmentation scheme could enhance the fairness of interpretations by DL models when dealing with data from patients with different demographic backgrounds.

FUNDING

National Science and Technology Council (Taiwan), National Institutes of Health.

Temporal self-attention for risk prediction from electronic health records using non-stationary kernel approximation

Effective modeling of patient representation from electronic health records (EHRs) is increasingly becoming a vital research topic. Yet, modeling the non-stationarity in EHR data has received less attention. Most existing studies follow a strong assumption of stationarity in patient representation from EHRs. However, in practice, a patient's visits are irregularly spaced over a relatively long period of time, and disease progression patterns exhibit non-stationarity. Furthermore, the time gaps between patient visits often encapsulate significant domain knowledge, potentially revealing undiscovered patterns that characterize specific medical conditions. To address these challenges, we introduce a new method which combines the self-attention mechanism with non-stationary kernel approximation to capture both contextual information and temporal relationships between patient visits in EHRs. To assess the effectiveness of our proposed approach, we use two real-world EHR datasets, comprising a total of 76,925 patients, for the task of predicting the next diagnosis code for a patient, given their EHR history. The first dataset is a general EHR cohort and consists of 11,451 patients with a total of 3,485 unique diagnosis codes. The second dataset is a disease-specific cohort that includes 65,474 pregnant patients and encompasses a total of 9,782 unique diagnosis codes. Our experimental evaluation involved nine prediction models, categorized into three distinct groups. Group 1 comprises the baselines: original self-attention with positional encoding model, RETAIN model, and LSTM model. Group 2 includes models employing self-attention with stationary kernel approximations, specifically incorporating three variations of Bochner's feature maps. Lastly, Group 3 consists of models utilizing self-attention with non-stationary kernel approximations, including quadratic, cubic, and bi-quadratic polynomials. The experimental results demonstrate that non-stationary kernels significantly outperformed baseline methods for NDCG@10 and Hit@10 metrics in both datasets. The performance boost was more substantial in dataset 1 for the NDCG@10 metric. On the other hand, stationary Kernels showed significant but smaller gains over baselines and were nearly as effective as Non-stationary Kernels for Hit@10 in dataset 2. These findings robustly validate the efficacy of employing non-stationary kernels for temporal modeling of EHR data, and emphasize the importance of modeling non-stationary temporal information in healthcare prediction tasks.

Digital Health and Machine Learning Technologies for Blood Glucose Monitoring and Management of Gestational Diabetes

Innovations in digital health and machine learning are changing the path of clinical health and care. People from different geographical locations and cultural backgrounds can benefit from the mobility of wearable devices and smartphones to monitor their health ubiquitously. This paper focuses on reviewing the digital health and machine learning technologies used in gestational diabetes - a subtype of diabetes that occurs during pregnancy. This paper reviews sensor technologies used in blood glucose monitoring devices, digital health innovations and machine learning models for gestational diabetes monitoring and management, in clinical and commercial settings, and discusses future directions. Despite one in six mothers having gestational diabetes, digital health applications were underdeveloped, especially the techniques that can be deployed in clinical practice. There is an urgent need to (1) develop clinically interpretable machine learning methods for patients with gestational diabetes, assisting health professionals with treatment, monitoring, and risk stratification before, during and after their pregnancies; (2) adapt and develop clinically-proven devices for patient self-management of health and well-being at home settings ("virtual ward" and virtual consultation), thereby improving clinical outcomes by facilitating timely intervention; and (3) ensure innovations are affordable and sustainable for all women with different socioeconomic backgrounds and clinical resources.

Split Learning for Distributed Collaborative Training of Deep Learning Models in Health Informatics

Deep learning continues to rapidly evolve and is now demonstrating remarkable potential for numerous medical prediction tasks. However, realizing deep learning models that generalize across healthcare organizations is challenging. This is due, in part, to the inherent siloed nature of these organizations and patient privacy requirements. To address this problem, we illustrate how split learning can enable collaborative training of deep learning models across disparate and privately maintained health datasets, while keeping the original records and model parameters private. We introduce a new privacy-preserving distributed learning framework that offers a higher level of privacy compared to conventional federated learning. We use several biomedical imaging and electronic health record (EHR) datasets to show that deep learning models trained via split learning can achieve highly similar performance to their centralized and federated counterparts while greatly improving computational efficiency and reducing privacy risks.

Identification of Outcome-Oriented Progression Subtypes from Mild Cognitive Impairment to Alzheimer's Disease Using Electronic Health Records

Alzheimer's disease (AD) is a complex heterogeneous neurodegenerative disease that requires an in-depth understanding of its progression pathways and contributing factors to develop effective risk stratification and prevention strategies. In this study, we proposed an outcome-oriented model to identify progression pathways from mild cognitive impairment (MCI) to AD using electronic health records (EHRs) from the OneFlorida+ Clinical Research Consortium. To achieve this, we employed the long short-term memory (LSTM) network to extract relevant information from the sequential records of each patient. The hierarchical agglomerative clustering was then applied to the learned representation to group patients based on their progression subtypes. Our approach identified multiple progression pathways, each of which represented distinct patterns of disease progression from MCI to AD. These pathways can serve as a valuable resource for researchers to understand the factors influencing AD progression and to develop personalized interventions to delay or prevent the onset of the disease.

Chinese experts' consensus on the application of intensive care big data

The development of intensive care medicine is inseparable from the diversified monitoring data. Intensive care medicine has been closely integrated with data since its birth. Critical care research requires an integrative approach that embraces the complexity of critical illness and the computational technology and algorithms that can make it possible. Considering the need of standardization of application of big data in intensive care, Intensive Care Medicine Branch of China Health Information and Health Care Big Data Society, Standard Committee has convened expert group, secretary group and the external audit expert group to formulate Chinese Experts' Consensus on the Application of Intensive Care Big Data (2022). This consensus makes 29 recommendations on the following five parts: Concept of intensive care big data, Important scientific issues, Standards and principles of database, Methodology in solving big data problems, Clinical application and safety consideration of intensive care big data. The consensus group believes this consensus is the starting step of application big data in the field of intensive care. More explorations and big data based retrospective research should be carried out in order to enhance safety and reliability of big data based models of critical care field.

Longitudinal dynamic clinical phenotypes of in-hospital COVID-19 patients across three dominant virus variants in New York

BACKGROUND

COVID-19 is a challenging disease to characterize given its wide-ranging heterogeneous symptomatology. Several studies have attempted to extract clinical phenotypes but often relied on data from small patient cohorts, usually limited to only one viral variant and utilizing a static snapshot of patient data.

OBJECTIVE

This study aimed to identify clinical phenotypes of hospitalized COVID-19 patients and investigate their longitudinal dynamics throughout the pandemic, with the goal to relate these phenotypes to clinical outcomes and treatment strategies.

METHODS

We utilized routinely collected demographic and clinical data throughout the hospitalization of 38,077 patients admitted between 3/2020 to 5/2022, in 12 New York hospitals. Uniform Manifold Approximation and Projection and agglomerative hierarchical clustering were used to derive the clusters, followed by exploratory data analysis to compare the prevalence of comorbidities and treatments per cluster.

RESULTS

4 distinct clinical phenotypes remained robust in multi-site validation and were associated with different mortality rates. The temporal progression of these phenotypes throughout the COVID-19 pandemic demonstrated increased variability across the waves of the three dominant viral variants (alpha, delta, omicron). Longitudinal analysis evaluating changes in clinical phenotypes of each patient throughout the course of a 4-week hospital stay exemplified the dynamic nature of the disease progression. Factors such as sex, race/ethnicity and specific treatment modalities revealed significant and clinically relevant differences between the observed phenotypes.

CONCLUSIONS

Our proposed methodology has the potential of enabling clinicians and policy makers to draw evidence-based conclusions for guiding treatment modalities in a dynamic fashion.

Large-scale deep learning analysis to identify adult patients at risk for combined and common variable immunodeficiencies

BACKGROUND

Primary immunodeficiency (PI) is a group of heterogeneous disorders resulting from immune system defects. Over 70% of PI is undiagnosed, leading to increased mortality, co-morbidity and healthcare costs. Among PI disorders, combined immunodeficiencies (CID) are characterized by complex immune defects. Common variable immunodeficiency (CVID) is among the most common types of PI. In light of available treatments, it is critical to identify adult patients at risk for CID and CVID, before the development of serious morbidity and mortality.

METHODS

We developed a deep learning-based method (named "TabMLPNet") to analyze clinical history from nationally representative medical claims from electronic health records (Optum® data, covering all US), evaluated in the setting of identifying CID/CVID in adults. Further, we revealed the most important CID/CVID-associated antecedent phenotype combinations. Four large cohorts were generated: a total of 47,660 PI cases and (1:1 matched) controls.

RESULTS

The sensitivity/specificity of TabMLPNet modeling ranges from 0.82-0.88/0.82-0.85 across cohorts. Distinctive combinations of antecedent phenotypes associated with CID/CVID are identified, consisting of respiratory infections/conditions, genetic anomalies, cardiac defects, autoimmune diseases, blood disorders and malignancies, which can possibly be useful to systematize the identification of CID and CVID.

CONCLUSIONS

We demonstrated an accurate method in terms of CID and CVID detection evaluated on large-scale medical claims data. Our predictive scheme can potentially lead to the development of new clinical insights and expanded guidelines for identification of adult patients at risk for CID and CVID as well as be used to improve patient outcomes on population level.

Cardiometabolic and renal phenotypes and transitions in the United States population

Cardiovascular and renal conditions have both shared and distinct determinants. In this study, we applied unsupervised clustering to multiple rounds of the National Health and Nutrition Examination Survey from 1988 to 2018, and identified 10 cardiometabolic and renal phenotypes. These included a 'low risk' phenotype; two groups with average risk factor levels but different heights; one group with low body-mass index and high levels of high-density lipoprotein cholesterol; five phenotypes with high levels of one or two related risk factors ('high heart rate', 'high cholesterol', 'high blood pressure', 'severe obesity' and 'severe hyperglycemia'); and one phenotype with low diastolic blood pressure (DBP) and low estimated glomerular filtration rate (eGFR). Prevalence of the 'high blood pressure' and 'high cholesterol' phenotypes decreased over time, contrasted by a rise in the 'severe obesity' and 'low DBP, low eGFR' phenotypes. The cardiometabolic and renal traits of the US population have shifted from phenotypes with high blood pressure and cholesterol toward poor kidney function, hyperglycemia and severe obesity.

AI in Rehabilitation Medicine: Opportunities and Challenges

Artificial intelligence (AI) tools are increasingly able to learn from larger and more complex data, thus allowing clinicians and scientists to gain new insights from the information they collect about their patients every day. In rehabilitation medicine, AI can be used to find patterns in huge amounts of healthcare data. These patterns can then be leveraged at the individual level, to design personalized care strategies and interventions to optimize each patient's outcomes. However, building effective AI tools requires many careful considerations about how we collect and handle data, how we train the models, and how we interpret results. In this perspective, we discuss some of the current opportunities and challenges for AI in rehabilitation. We first review recent trends in AI for the screening, diagnosis, treatment, and continuous monitoring of disease or injury, with a special focus on the different types of healthcare data used for these applications. We then examine potential barriers to designing and integrating AI into the clinical workflow, and we propose an end-to-end framework to address these barriers and guide the development of effective AI for rehabilitation. Finally, we present ideas for future work to pave the way for AI implementation in real-world rehabilitation practices.

Fair patient model: Mitigating bias in the patient representation learned from the electronic health records

OBJECTIVE

To pre-train fair and unbiased patient representations from Electronic Health Records (EHRs) using a novel weighted loss function that reduces bias and improves fairness in deep representation learning models.

METHODS

We defined a new loss function, called weighted loss function, in the deep representation learning model to balance the importance of different groups of patients and features. We applied the proposed model, called Fair Patient Model (FPM), to a sample of 34,739 patients from the MIMIC-III dataset and learned patient representations for four clinical outcome prediction tasks.

RESULTS

FPM outperformed the baseline models in terms of three fairness metrics: demographic parity, equality of opportunity difference, and equalized odds ratio. FPM also achieved comparable predictive performance with the baselines, with an average accuracy of 0.7912. Feature analysis revealed that FPM captured more information from clinical features than the baselines.

CONCLUSION

FPM is a novel method to pre-train fair and unbiased patient representations from the EHR data using a weighted loss function. The learned representations can be used for various downstream tasks in healthcare and can be extended to other domains where fairness is important.

Patient Information Summarization in Clinical Settings: Scoping Review

BACKGROUND

Information overflow, a common problem in the present clinical environment, can be mitigated by summarizing clinical data. Although there are several solutions for clinical summarization, there is a lack of a complete overview of the research relevant to this field.

OBJECTIVE

This study aims to identify state-of-the-art solutions for clinical summarization, to analyze their capabilities, and to identify their properties.

METHODS

A scoping review of articles published between 2005 and 2022 was conducted. With a clinical focus, PubMed and Web of Science were queried to find an initial set of reports, later extended by articles found through a chain of citations. The included reports were analyzed to answer the questions of where, what, and how medical information is summarized; whether summarization conserves temporality, uncertainty, and medical pertinence; and how the propositions are evaluated and deployed. To answer how information is summarized, methods were compared through a new framework "collect-synthesize-communicate" referring to information gathering from data, its synthesis, and communication to the end user.

RESULTS

Overall, 128 articles were included, representing various medical fields. Exclusively structured data were used as input in 46.1% (59/128) of papers, text in 41.4% (53/128) of articles, and both in 10.2% (13/128) of papers. Using the proposed framework, 42.2% (54/128) of the records contributed to information collection, 27.3% (35/128) contributed to information synthesis, and 46.1% (59/128) presented solutions for summary communication. Numerous summarization approaches have been presented, including extractive (n=13) and abstractive summarization (n=19); topic modeling (n=5); summary specification (n=11); concept and relation extraction (n=30); visual design considerations (n=59); and complete pipelines (n=7) using information extraction, synthesis, and communication. Graphical displays (n=53), short texts (n=41), static reports (n=7), and problem-oriented views (n=7) were the most common types in terms of summary communication. Although temporality and uncertainty information were usually not conserved in most studies (74/128, 57.8% and 113/128, 88.3%, respectively), some studies presented solutions to treat this information. Overall, 115 (89.8%) articles showed results of an evaluation, and methods included evaluations with human participants (median 15, IQR 24 participants): measurements in experiments with human participants (n=31), real situations (n=8), and usability studies (n=28). Methods without human involvement included intrinsic evaluation (n=24), performance on a proxy (n=10), or domain-specific tasks (n=11). Overall, 11 (8.6%) reports described a system deployed in clinical settings.

CONCLUSIONS

The scientific literature contains many propositions for summarizing patient information but reports very few comparisons of these proposals. This work proposes to compare these algorithms through how they conserve essential aspects of clinical information and through the "collect-synthesize-communicate" framework. We found that current propositions usually address these 3 steps only partially. Moreover, they conserve and use temporality, uncertainty, and pertinent medical aspects to varying extents, and solutions are often preliminary.

Precision subclassification of type 2 diabetes: a systematic review

BACKGROUND

Heterogeneity in type 2 diabetes presentation and progression suggests that precision medicine interventions could improve clinical outcomes. We undertook a systematic review to determine whether strategies to subclassify type 2 diabetes were associated with high quality evidence, reproducible results and improved outcomes for patients.

METHODS

We searched PubMed and Embase for publications that used 'simple subclassification' approaches using simple categorisation of clinical characteristics, or 'complex subclassification' approaches which used machine learning or 'omics approaches in people with established type 2 diabetes. We excluded other diabetes subtypes and those predicting incident type 2 diabetes. We assessed quality, reproducibility and clinical relevance of extracted full-text articles and qualitatively synthesised a summary of subclassification approaches.

RESULTS

Here we show data from 51 studies that demonstrate many simple stratification approaches, but none have been replicated and many are not associated with meaningful clinical outcomes. Complex stratification was reviewed in 62 studies and produced reproducible subtypes of type 2 diabetes that are associated with outcomes. Both approaches require a higher grade of evidence but support the premise that type 2 diabetes can be subclassified into clinically meaningful subtypes.

CONCLUSION

Critical next steps toward clinical implementation are to test whether subtypes exist in more diverse ancestries and whether tailoring interventions to subtypes will improve outcomes.

Unsupervised pre-training of graph transformers on patient population graphs

Pre-training has shown success in different areas of machine learning, such as Computer Vision, Natural Language Processing (NLP), and medical imaging. However, it has not been fully explored for clinical data analysis. An immense amount of clinical records are recorded, but still, data and labels can be scarce for data collected in small hospitals or dealing with rare diseases. In such scenarios, pre-training on a larger set of unlabeled clinical data could improve performance. In this paper, we propose novel unsupervised pre-training techniques designed for heterogeneous, multi-modal clinical data for patient outcome prediction inspired by masked language modeling (MLM), by leveraging graph deep learning over population graphs. To this end, we further propose a graph-transformer-based network, designed to handle heterogeneous clinical data. By combining masking-based pre-training with a transformer-based network, we translate the success of masking-based pre-training in other domains to heterogeneous clinical data. We show the benefit of our pre-training method in a self-supervised and a transfer learning setting, utilizing three medical datasets TADPOLE, MIMIC-III, and a Sepsis Prediction Dataset. We find that our proposed pre-training methods help in modeling the data at a patient and population level and improve performance in different fine-tuning tasks on all datasets.

A scoping review of the clinical application of machine learning in data-driven population segmentation analysis

OBJECTIVE

Data-driven population segmentation is commonly used in clinical settings to separate the heterogeneous population into multiple relatively homogenous groups with similar healthcare features. In recent years, machine learning (ML) based segmentation algorithms have garnered interest for their potential to speed up and improve algorithm development across many phenotypes and healthcare situations. This study evaluates ML-based segmentation with respect to (1) the populations applied, (2) the segmentation details, and (3) the outcome evaluations.

MATERIALS AND METHODS

MEDLINE, Embase, Web of Science, and Scopus were used following the PRISMA-ScR criteria. Peer-reviewed studies in the English language that used data-driven population segmentation analysis on structured data from January 2000 to October 2022 were included.

RESULTS

We identified 6077 articles and included 79 for the final analysis. Data-driven population segmentation analysis was employed in various clinical settings. K-means clustering is the most prevalent unsupervised ML paradigm. The most common settings were healthcare institutions. The most common targeted population was the general population.

DISCUSSION

Although all the studies did internal validation, only 11 papers (13.9%) did external validation, and 23 papers (29.1%) conducted methods comparison. The existing papers discussed little validating the robustness of ML modeling.

CONCLUSION

Existing ML applications on population segmentation need more evaluations regarding giving tailored, efficient integrated healthcare solutions compared to traditional segmentation analysis. Future ML applications in the field should emphasize methods' comparisons and external validation and investigate approaches to evaluate individual consistency using different methods.

Identification of Outcome-Oriented Progression Subtypes from Mild Cognitive Impairment to Alzheimer's Disease Using Electronic Health Records

Alzheimer's disease (AD) is a complex heterogeneous neurodegenerative disease that requires an in-depth understanding of its progression pathways and contributing factors to develop effective risk stratification and prevention strategies. In this study, we proposed an outcome-oriented model to identify progression pathways from mild cognitive impairment (MCI) to AD using electronic health records (EHRs) from the OneFlorida+ Clinical Research Consortium. To achieve this, we employed the long short-term memory (LSTM) network to extract relevant information from the sequential records of each patient. The hierarchical agglomerative clustering was then applied to the learned representation to group patients based on their progression subtypes. Our approach identified multiple progression pathways, each of which represented distinct patterns of disease progression from MCI to AD. These pathways can serve as a valuable resource for researchers to understand the factors influencing AD progression and to develop personalized interventions to delay or prevent the onset of the disease.

An Alzheimer's disease category progression sub-grouping analysis using manifold learning on ADNI

Many current statistical and machine learning methods have been used to explore Alzheimer's disease (AD) and its associated patterns that contribute to the disease. However, there has been limited success in understanding the relationship between cognitive tests, biomarker data, and patient AD category progressions. In this work, we perform exploratory data analysis of AD health record data by analyzing various learned lower dimensional manifolds to separate early-stage AD categories further. Specifically, we used Spectral embedding, Multidimensional scaling, Isomap, t-Distributed Stochastic Neighbour Embedding, Uniform Manifold Approximation and Projection, and sparse denoising autoencoder based manifolds on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. We then determine the clustering potential of the learned embeddings and then determine if category sub-groupings or sub-categories can be found. We then used a Kruskal-sWallis H test to determine the statistical significance of the discovered AD subcategories. Our results show that the existing AD categories do exhibit sub-groupings, especially in mild cognitive impairment transitions in many of the tested manifolds, showing there may be a need for further subcategories to describe AD progression.

Representation Learning and Spectral Clustering for the Development and External Validation of Dynamic Sepsis Phenotypes: Observational Cohort Study

BACKGROUND

Recent attempts at clinical phenotyping for sepsis have shown promise in identifying groups of patients with distinct treatment responses. Nonetheless, the replicability and actionability of these phenotypes remain an issue because the patient trajectory is a function of both the patient's physiological state and the interventions they receive.

OBJECTIVE

We aimed to develop a novel approach for deriving clinical phenotypes using unsupervised learning and transition modeling.

METHODS

Forty commonly used clinical variables from the electronic health record were used as inputs to a feed-forward neural network trained to predict the onset of sepsis. Using spectral clustering on the representations from this network, we derived and validated consistent phenotypes across a diverse cohort of patients with sepsis. We modeled phenotype dynamics as a Markov decision process with transitions as a function of the patient's current state and the interventions they received.

RESULTS

Four consistent and distinct phenotypes were derived from over 11,500 adult patients who were admitted from the University of California, San Diego emergency department (ED) with sepsis between January 1, 2016, and January 31, 2020. Over 2000 adult patients admitted from the University of California, Irvine ED with sepsis between November 4, 2017, and August 4, 2022, were involved in the external validation. We demonstrate that sepsis phenotypes are not static and evolve in response to physiological factors and based on interventions. We show that roughly 45% of patients change phenotype membership within the first 6 hours of ED arrival. We observed consistent trends in patient dynamics as a function of interventions including early administration of antibiotics.

CONCLUSIONS

We derived and describe 4 sepsis phenotypes present within 6 hours of triage in the ED. We observe that the administration of a 30 mL/kg fluid bolus may be associated with worse outcomes in certain phenotypes, whereas prompt antimicrobial therapy is associated with improved outcomes.

Bottom-up and top-down paradigms of artificial intelligence research approaches to healthcare data science using growing real-world big data

OBJECTIVES

As the real-world electronic health record (EHR) data continue to grow exponentially, novel methodologies involving artificial intelligence (AI) are becoming increasingly applied to enable efficient data-driven learning and, ultimately, to advance healthcare. Our objective is to provide readers with an understanding of evolving computational methods and help in deciding on methods to pursue.

TARGET AUDIENCE

The sheer diversity of existing methods presents a challenge for health scientists who are beginning to apply computational methods to their research. Therefore, this tutorial is aimed at scientists working with EHR data who are early entrants into the field of applying AI methodologies.

SCOPE

This manuscript describes the diverse and growing AI research approaches in healthcare data science and categorizes them into 2 distinct paradigms, the bottom-up and top-down paradigms to provide health scientists venturing into artificial intelligent research with an understanding of the evolving computational methods and help in deciding on methods to pursue through the lens of real-world healthcare data.

Hypergraph Transformers for EHR-based Clinical Predictions

Electronic health records (EHR) data contain rich information about patients' health conditions including diagnosis, procedures, medications and etc., which have been widely used to facilitate digital medicine. Despite its importance, it is often non-trivial to learn useful representations for patients' visits that support downstream clinical predictions, as each visit contains massive and diverse medical codes. As a result, the complex interactions among medical codes are often not captured, which leads to substandard predictions. To better model these complex relations, we leverage hypergraphs, which go beyond pairwise relations to jointly learn the representations for visits and medical codes. We also propose to use the self-attention mechanism to automatically identify the most relevant medical codes for each visit based on the downstream clinical predictions with better generalization power. Experiments on two EHR datasets show that our proposed method not only yields superior performance, but also provides reasonable insights towards the target tasks.

Leveraging electronic health record data for endometriosis research

Endometriosis is a chronic, complex disease for which there are vast disparities in diagnosis and treatment between sociodemographic groups. Clinical presentation of endometriosis can vary from asymptomatic disease-often identified during (in)fertility consultations-to dysmenorrhea and debilitating pelvic pain. Because of this complexity, delayed diagnosis (mean time to diagnosis is 1.7-3.6 years) and misdiagnosis is common. Early and accurate diagnosis of endometriosis remains a research priority for patient advocates and healthcare providers. Electronic health records (EHRs) have been widely adopted as a data source in biomedical research. However, they remain a largely untapped source of data for endometriosis research. EHRs capture diverse, real-world patient populations and care trajectories and can be used to learn patterns of underlying risk factors for endometriosis which, in turn, can be used to inform screening guidelines to help clinicians efficiently and effectively recognize and diagnose the disease in all patient populations reducing inequities in care. Here, we provide an overview of the advantages and limitations of using EHR data to study endometriosis. We describe the prevalence of endometriosis observed in diverse populations from multiple healthcare institutions, examples of variables that can be extracted from EHRs to enhance the accuracy of endometriosis prediction, and opportunities to leverage longitudinal EHR data to improve our understanding of long-term health consequences for all patients.

Early detection of Parkinson's disease through enriching the electronic health record using a biomedical knowledge graph

Introduction

Early diagnosis of Parkinson's disease (PD) is important to identify treatments to slow neurodegeneration. People who develop PD often have symptoms before the disease manifests and may be coded as diagnoses in the electronic health record (EHR).

Methods

To predict PD diagnosis, we embedded EHR data of patients onto a biomedical knowledge graph called Scalable Precision medicine Open Knowledge Engine (SPOKE) and created patient embedding vectors. We trained and validated a classifier using these vectors from 3,004 PD patients, restricting records to 1, 3, and 5 years before diagnosis, and 457,197 non-PD group.

Results

The classifier predicted PD diagnosis with moderate accuracy (AUC = 0.77 ± 0.06, 0.74 ± 0.05, 0.72 ± 0.05 at 1, 3, and 5 years) and performed better than other benchmark methods. Nodes in the SPOKE graph, among cases, revealed novel associations, while SPOKE patient vectors revealed the basis for individual risk classification.

Discussion

The proposed method was able to explain the clinical predictions using the knowledge graph, thereby making the predictions clinically interpretable. Through enriching EHR data with biomedical associations, SPOKE may be a cost-efficient and personalized way to predict PD diagnosis years before its occurrence.

Fusion of imaging and non-imaging data for disease trajectory prediction for coronavirus disease 2019 patients

Purpose

Our study investigates whether graph-based fusion of imaging data with non-imaging electronic health records (EHR) data can improve the prediction of the disease trajectories for patients with coronavirus disease 2019 (COVID-19) beyond the prediction performance of only imaging or non-imaging EHR data.

Approach

We present a fusion framework for fine-grained clinical outcome prediction [discharge, intensive care unit (ICU) admission, or death] that fuses imaging and non-imaging information using a similarity-based graph structure. Node features are represented by image embedding, and edges are encoded with clinical or demographic similarity.

Results

Experiments on data collected from the Emory Healthcare Network indicate that our fusion modeling scheme performs consistently better than predictive models developed using only imaging or non-imaging features, with area under the receiver operating characteristics curve of 0.76, 0.90, and 0.75 for discharge from hospital, mortality, and ICU admission, respectively. External validation was performed on data collected from the Mayo Clinic. Our scheme highlights known biases in the model prediction, such as bias against patients with alcohol abuse history and bias based on insurance status.

Conclusions

Our study signifies the importance of the fusion of multiple data modalities for the accurate prediction of clinical trajectories. The proposed graph structure can model relationships between patients based on non-imaging EHR data, and graph convolutional networks can fuse this relationship information with imaging data to effectively predict future disease trajectory more effectively than models employing only imaging or non-imaging data. Our graph-based fusion modeling frameworks can be easily extended to other prediction tasks to efficiently combine imaging data with non-imaging clinical data.

Systematic review of precision subclassification of type 2 diabetes

Heterogeneity in type 2 diabetes presentation, progression and treatment has the potential for precision medicine interventions that can enhance care and outcomes for affected individuals. We undertook a systematic review to ascertain whether strategies to subclassify type 2 diabetes are associated with improved clinical outcomes, show reproducibility and have high quality evidence. We reviewed publications that deployed 'simple subclassification' using clinical features, biomarkers, imaging or other routinely available parameters or 'complex subclassification' approaches that used machine learning and/or genomic data. We found that simple stratification approaches, for example, stratification based on age, body mass index or lipid profiles, had been widely used, but no strategy had been replicated and many lacked association with meaningful outcomes. Complex stratification using clustering of simple clinical data with and without genetic data did show reproducible subtypes of diabetes that had been associated with outcomes such as cardiovascular disease and/or mortality. Both approaches require a higher grade of evidence but support the premise that type 2 diabetes can be subclassified into meaningful groups. More studies are needed to test these subclassifications in more diverse ancestries and prove that they are amenable to interventions.

A scoping review of neurodegenerative manifestations in explainable digital phenotyping

Neurologists nowadays no longer view neurodegenerative diseases, like Parkinson's and Alzheimer's disease, as single entities, but rather as a spectrum of multifaceted symptoms with heterogeneous progression courses and treatment responses. The definition of the naturalistic behavioral repertoire of early neurodegenerative manifestations is still elusive, impeding early diagnosis and intervention. Central to this view is the role of artificial intelligence (AI) in reinforcing the depth of phenotypic information, thereby supporting the paradigm shift to precision medicine and personalized healthcare. This suggestion advocates the definition of disease subtypes in a new biomarker-supported nosology framework, yet without empirical consensus on standardization, reliability and interpretability. Although the well-defined neurodegenerative processes, linked to a triad of motor and non-motor preclinical symptoms, are detected by clinical intuition, we undertake an unbiased data-driven approach to identify different patterns of neuropathology distribution based on the naturalistic behavior data inherent to populations in-the-wild. We appraise the role of remote technologies in the definition of digital phenotyping specific to brain-, body- and social-level neurodegenerative subtle symptoms, emphasizing inter- and intra-patient variability powered by deep learning. As such, the present review endeavors to exploit digital technologies and AI to create disease-specific phenotypic explanations, facilitating the understanding of neurodegenerative diseases as "bio-psycho-social" conditions. Not only does this translational effort within explainable digital phenotyping foster the understanding of disease-induced traits, but it also enhances diagnostic and, eventually, treatment personalization.

Building Practical Risk Prediction Models for Nasopharyngeal Carcinoma Screening with Patient Graph Analysis and Machine Learning

BACKGROUND

To expand nasopharyngeal carcinoma (NPC) screening to larger populations, more practical NPC risk prediction models independent of Epstein-Barr virus (EBV) and other lab tests are necessary.

METHODS

Patient data before diagnosis of NPC were collected from hospital electronic medical records (EMR) and used to develop machine learning (ML) models for NPC risk prediction using XGBoost. NPC risk factor distributions were generated through connection delta ratio (CDR) analysis of patient graphs. By combining EMR-wide ML with patient graph analysis, the number of variables in these risk models was reduced, allowing for more practical NPC risk prediction ML models.

RESULTS

Using data collected from 1,357 patients with NPC and 1,448 patients with control, an optimal set of 100 variables (ov100) was determined for building NPC risk prediction ML models that had, the following performance metrics: 0.93-0.96 recall, 0.80-0.92 precision, and 0.83-0.94 AUC. Aided by the analysis of top CDR-ranked risk factors, the models were further refined to contain only 20 practical variables (pv20), excluding EBV. The pv20 NPC risk XGBoost model achieved 0.79 recall, 0.94 precision, 0.96 specificity, and 0.87 AUC.

CONCLUSIONS

This study demonstrated the feasibility of developing practical NPC risk prediction models using EMR-wide ML and patient graph CDR analysis, without requiring EBV data. These models could enable broader implementation of NPC risk evaluation and screening recommendations for larger populations in urban community health centers and rural clinics.

IMPACT

These more practical NPC risk models could help increase NPC screening rate and identify more patients with early-stage NPC.

Evaluating and mitigating bias in machine learning models for cardiovascular disease prediction

OBJECTIVE

The study aims to investigate whether machine learning-based predictive models for cardiovascular disease (CVD) risk assessment show equivalent performance across demographic groups (such as race and gender) and if bias mitigation methods can reduce any bias present in the models. This is important as systematic bias may be introduced when collecting and preprocessing health data, which could affect the performance of the models on certain demographic sub-cohorts. The study is to investigate this using electronic health records data and various machine learning models.

METHODS

The study used large de-identified Electronic Health Records data from Vanderbilt University Medical Center. Machine learning (ML) algorithms including logistic regression, random forest, gradient-boosting trees, and long short-term memory were applied to build multiple predictive models. Model bias and fairness were evaluated using equal opportunity difference (EOD, 0 indicates fairness) and disparate impact (DI, 1 indicates fairness). In our study, we also evaluated the fairness of a non-ML baseline model, the American Heart Association (AHA) Pooled Cohort Risk Equations (PCEs). Moreover, we compared the performance of three different de-biasing methods: removing protected attributes (e.g., race and gender), resampling the imbalanced training dataset by sample size, and resampling by the proportion of people with CVD outcomes.

RESULTS

The study cohort included 109,490 individuals (mean [SD] age 47.4 [14.7] years; 64.5% female; 86.3% White; 13.7% Black). The experimental results suggested that most ML models had smaller EOD and DI than PCEs. For ML models, the mean EOD ranged from -0.001 to 0.018 and the mean DI ranged from 1.037 to 1.094 across race groups. There was a larger EOD and DI across gender groups, with EOD ranging from 0.131 to 0.136 and DI ranging from 1.535 to 1.587. For debiasing methods, removing protected attributes didn't significantly reduced the bias for most ML models. Resampling by sample size also didn't consistently decrease bias. Resampling by case proportion reduced the EOD and DI for gender groups but slightly reduced accuracy in many cases.

CONCLUSIONS

Among the VUMC cohort, both PCEs and ML models were biased against women, suggesting the need to investigate and correct gender disparities in CVD risk prediction. Resampling by proportion reduced the bias for gender groups but not for race groups.

Explaining predictive factors in patient pathways using autoencoders

This paper introduces an end-to-end methodology to predict a pathway-related outcome and identifying predictive factors using autoencoders. A formal description of autoencoders for explainable binary predictions is presented, along with two objective functions that allows for filtering and inverting negative examples during training. A methodology to model and transform complex medical event logs is also proposed, which keeps the pathway information in terms of events and time, as well as the hierarchy information carried in medical codes. A case study is presented, in which the short-term mortality after the implementation of an Implantable Cardioverter-Defibrillator is predicted. Proposed methodologies have been tested and compared to other predictive methods, both explainable and not explainable. Results show the competitiveness of the method in terms of performances, particularly the use of a Variational Auto Encoder with an inverse objective function. Finally, the explainability of the method has been demonstrated, allowing for the identification of interesting predictive factors validated using relative risks.

Representational ethical model calibration

Equity is widely held to be fundamental to the ethics of healthcare. In the context of clinical decision-making, it rests on the comparative fidelity of the intelligence - evidence-based or intuitive - guiding the management of each individual patient. Though brought to recent attention by the individuating power of contemporary machine learning, such epistemic equity arises in the context of any decision guidance, whether traditional or innovative. Yet no general framework for its quantification, let alone assurance, currently exists. Here we formulate epistemic equity in terms of model fidelity evaluated over learnt multidimensional representations of identity crafted to maximise the captured diversity of the population, introducing a comprehensive framework for Representational Ethical Model Calibration. We demonstrate the use of the framework on large-scale multimodal data from UK Biobank to derive diverse representations of the population, quantify model performance, and institute responsive remediation. We offer our approach as a principled solution to quantifying and assuring epistemic equity in healthcare, with applications across the research, clinical, and regulatory domains.

Simulation of a machine learning enabled learning health system for risk prediction using synthetic patient data

When enabled by machine learning (ML), Learning Health Systems (LHS) hold promise for improving the effectiveness of healthcare delivery to patients. One major barrier to LHS research and development is the lack of access to EHR patient data. To overcome this challenge, this study demonstrated the feasibility of developing a simulated ML-enabled LHS using synthetic patient data. The ML-enabled LHS was initialized using a dataset of 30,000 synthetic Synthea patients and a risk prediction XGBoost base model for lung cancer. 4 additional datasets of 30,000 patients were generated and added to the previous updated dataset sequentially to simulate addition of new patients, resulting in datasets of 60,000, 90,000, 120,000 and 150,000 patients. New XGBoost models were built in each instance, and performance improved with data size increase, attaining 0.936 recall and 0.962 AUC (area under curve) in the 150,000 patients dataset. The effectiveness of the new ML-enabled LHS process was verified by implementing XGBoost models for stroke risk prediction on the same Synthea patient populations. By making the ML code and synthetic patient data publicly available for testing and training, this first synthetic LHS process paves the way for more researchers to start developing LHS with real patient data.

Modeling electronic health record data using an end-to-end knowledge-graph-informed topic model

The rapid growth of electronic health record (EHR) datasets opens up promising opportunities to understand human diseases in a systematic way. However, effective extraction of clinical knowledge from EHR data has been hindered by the sparse and noisy information. We present Graph ATtention-Embedded Topic Model (GAT-ETM), an end-to-end taxonomy-knowledge-graph-based multimodal embedded topic model. GAT-ETM distills latent disease topics from EHR data by learning the embedding from a constructed medical knowledge graph. We applied GAT-ETM to a large-scale EHR dataset consisting of over 1 million patients. We evaluated its performance based on topic quality, drug imputation, and disease diagnosis prediction. GAT-ETM demonstrated superior performance over the alternative methods on all tasks. Moreover, GAT-ETM learned clinically meaningful graph-informed embedding of the EHR codes and discovered interpretable and accurate patient representations for patient stratification and drug recommendations. GAT-ETM code is available at https://github.com/li-lab-mcgill/GAT-ETM .

Artificial Intelligence in Breast Cancer Screening and Diagnosis

Cancer is a disease that continues to plague our modern society. Among all types of cancer, breast cancer is now the most common type of cancer occurring in women worldwide. Various factors, including genetics, lifestyle, and the environment, have contributed to the rise in the prevalence of breast cancer among women of all socioeconomic strata. Therefore, proper screening for early diagnosis and treatment becomes a major factor when fighting the disease. Artificial intelligence (AI) continues to revolutionize various spheres of our lives with its numerous applications. Using AI in the existing screening process makes obtaining results even easier and more convenient. Faster, more accurate results are some of the benefits of AI methods in breast cancer screening. Nonetheless, there are many challenges in the process of the integration of AI that needs to be addressed systematically. The following is a review of the application of AI in breast cancer screening.

Personalized survival probabilities for SARS-CoV-2 positive patients by explainable machine learning

Interpretable risk assessment of SARS-CoV-2 positive patients can aid clinicians to implement precision medicine. Here we trained a machine learning model to predict mortality within 12 weeks of a first positive SARS-CoV-2 test. By leveraging data on 33,938 confirmed SARS-CoV-2 cases in eastern Denmark, we considered 2723 variables extracted from electronic health records (EHR) including demographics, diagnoses, medications, laboratory test results and vital parameters. A discrete-time framework for survival modelling enabled us to predict personalized survival curves and explain individual risk factors. Performance on the test set was measured with a weighted concordance index of 0.95 and an area under the curve for precision-recall of 0.71. Age, sex, number of medications, previous hospitalizations and lymphocyte counts were identified as top mortality risk factors. Our explainable survival model developed on EHR data also revealed temporal dynamics of the 22 selected risk factors. Upon further validation, this model may allow direct reporting of personalized survival probabilities in routine care.

Preparing for the next pandemic via transfer learning from existing diseases with hierarchical multi-modal BERT: a study on COVID-19 outcome prediction

Developing prediction models for emerging infectious diseases from relatively small numbers of cases is a critical need for improving pandemic preparedness. Using COVID-19 as an exemplar, we propose a transfer learning methodology for developing predictive models from multi-modal electronic healthcare records by leveraging information from more prevalent diseases with shared clinical characteristics. Our novel hierarchical, multi-modal model (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\textsc {TransMED}}$$\end{document}TRANSMED) integrates baseline risk factors from the natural language processing of clinical notes at admission, time-series measurements of biomarkers obtained from laboratory tests, and discrete diagnostic, procedure and drug codes. We demonstrate the alignment of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\textsc {TransMED}}$$\end{document}TRANSMED’s predictions with well-established clinical knowledge about COVID-19 through univariate and multivariate risk factor driven sub-cohort analysis. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\textsc {TransMED}}$$\end{document}TRANSMED’s superior performance over state-of-the-art methods shows that leveraging patient data across modalities and transferring prior knowledge from similar disorders is critical for accurate prediction of patient outcomes, and this approach may serve as an important tool in the early response to future pandemics.

A framework for employing longitudinally collected multicenter electronic health records to stratify heterogeneous patient populations on disease history

OBJECTIVE

To facilitate patient disease subset and risk factor identification by constructing a pipeline which is generalizable, provides easily interpretable results, and allows replication by overcoming electronic health records (EHRs) batch effects.

MATERIAL AND METHODS

We used 1872 billing codes in EHRs of 102 880 patients from 12 healthcare systems. Using tools borrowed from single-cell omics, we mitigated center-specific batch effects and performed clustering to identify patients with highly similar medical history patterns across the various centers. Our visualization method (PheSpec) depicts the phenotypic profile of clusters, applies a novel filtering of noninformative codes (Ranked Scope Pervasion), and indicates the most distinguishing features.

RESULTS

We observed 114 clinically meaningful profiles, for example, linking prostate hyperplasia with cancer and diabetes with cardiovascular problems and grouping pediatric developmental disorders. Our framework identified disease subsets, exemplified by 6 "other headache" clusters, where phenotypic profiles suggested different underlying mechanisms: migraine, convulsion, injury, eye problems, joint pain, and pituitary gland disorders. Phenotypic patterns replicated well, with high correlations of ≥0.75 to an average of 6 (2-8) of the 12 different cohorts, demonstrating the consistency with which our method discovers disease history profiles.

DISCUSSION

Costly clinical research ventures should be based on solid hypotheses. We repurpose methods from single-cell omics to build these hypotheses from observational EHR data, distilling useful information from complex data.

CONCLUSION

We establish a generalizable pipeline for the identification and replication of clinically meaningful (sub)phenotypes from widely available high-dimensional billing codes. This approach overcomes datatype problems and produces comprehensive visualizations of validation-ready phenotypes.

Pretrained transformer framework on pediatric claims data for population specific tasks

The adoption of electronic health records (EHR) has become universal during the past decade, which has afforded in-depth data-based research. By learning from the large amount of healthcare data, various data-driven models have been built to predict future events for different medical tasks, such as auto diagnosis and heart-attack prediction. Although EHR is abundant, the population that satisfies specific criteria for learning population-specific tasks is scarce, making it challenging to train data-hungry deep learning models. This study presents the Claim Pre-Training (Claim-PT) framework, a generic pre-training model that first trains on the entire pediatric claims dataset, followed by a discriminative fine-tuning on each population-specific task. The semantic meaning of medical events can be captured in the pre-training stage, and the effective knowledge transfer is completed through the task-aware fine-tuning stage. The fine-tuning process requires minimal parameter modification without changing the model architecture, which mitigates the data scarcity issue and helps train the deep learning model adequately on small patient cohorts. We conducted experiments on a real-world pediatric dataset with more than one million patient records. Experimental results on two downstream tasks demonstrated the effectiveness of our method: our general task-agnostic pre-training framework outperformed tailored task-specific models, achieving more than 10% higher in model performance as compared to baselines. In addition, our framework showed a potential to transfer learned knowledge from one institution to another, which may pave the way for future healthcare model pre-training across institutions.

Predictive structured-unstructured interactions in EHR models: A case study of suicide prediction

Clinical risk prediction models powered by electronic health records (EHRs) are becoming increasingly widespread in clinical practice. With suicide-related mortality rates rising in recent years, it is becoming increasingly urgent to understand, predict, and prevent suicidal behavior. Here, we compare the predictive value of structured and unstructured EHR data for predicting suicide risk. We find that Naive Bayes Classifier (NBC) and Random Forest (RF) models trained on structured EHR data perform better than those based on unstructured EHR data. An NBC model trained on both structured and unstructured data yields similar performance (AUC = 0.743) to an NBC model trained on structured data alone (0.742, p = 0.668), while an RF model trained on both data types yields significantly better results (AUC = 0.903) than an RF model trained on structured data alone (0.887, p < 0.001), likely due to the RF model’s ability to capture interactions between the two data types. To investigate these interactions, we propose and implement a general framework for identifying specific structured-unstructured feature pairs whose interactions differ between case and non-case cohorts, and thus have the potential to improve predictive performance and increase understanding of clinical risk. We find that such feature pairs tend to capture heterogeneous pairs of general concepts, rather than homogeneous pairs of specific concepts. These findings and this framework can be used to improve current and future EHR-based clinical modeling efforts.

Deep learning for temporal data representation in electronic health records: A systematic review of challenges and methodologies

OBJECTIVE

Temporal electronic health records (EHRs) contain a wealth of information for secondary uses, such as clinical events prediction and chronic disease management. However, challenges exist for temporal data representation. We therefore sought to identify these challenges and evaluate novel methodologies for addressing them through a systematic examination of deep learning solutions.

METHODS

We searched five databases (PubMed, Embase, the Institute of Electrical and Electronics Engineers [IEEE] Xplore Digital Library, the Association for Computing Machinery [ACM] Digital Library, and Web of Science) complemented with hand-searching in several prestigious computer science conference proceedings. We sought articles that reported deep learning methodologies on temporal data representation in structured EHR data from January 1, 2010, to August 30, 2020. We summarized and analyzed the selected articles from three perspectives: nature of time series, methodology, and model implementation.

RESULTS

We included 98 articles related to temporal data representation using deep learning. Four major challenges were identified, including data irregularity, heterogeneity, sparsity, and model opacity. We then studied how deep learning techniques were applied to address these challenges. Finally, we discuss some open challenges arising from deep learning.

CONCLUSION

Temporal EHR data present several major challenges for clinical prediction modeling and data utilization. To some extent, current deep learning solutions can address these challenges. Future studies may consider designing comprehensive and integrated solutions. Moreover, researchers should incorporate clinical domain knowledge into study designs and enhance model interpretability to facilitate clinical implementation.

Identifying and evaluating clinical subtypes of Alzheimer's disease in care electronic health records using unsupervised machine learning

BACKGROUND

Alzheimer's disease (AD) is a highly heterogeneous disease with diverse trajectories and outcomes observed in clinical populations. Understanding this heterogeneity can enable better treatment, prognosis and disease management. Studies to date have mainly used imaging or cognition data and have been limited in terms of data breadth and sample size. Here we examine the clinical heterogeneity of Alzheimer's disease patients using electronic health records (EHR) to identify and characterise disease subgroups using multiple clustering methods, identifying clusters which are clinically actionable.

METHODS

We identified AD patients in primary care EHR from the Clinical Practice Research Datalink (CPRD) using a previously validated rule-based phenotyping algorithm. We extracted and included a range of comorbidities, symptoms and demographic features as patient features. We evaluated four different clustering methods (k-means, kernel k-means, affinity propagation and latent class analysis) to cluster Alzheimer's disease patients. We compared clusters on clinically relevant outcomes and evaluated each method using measures of cluster structure, stability, efficiency of outcome prediction and replicability in external data sets.

RESULTS

We identified 7,913 AD patients, with a mean age of 82 and 66.2% female. We included 21 features in our analysis. We observed 5, 2, 5 and 6 clusters in k-means, kernel k-means, affinity propagation and latent class analysis respectively. K-means was found to produce the most consistent results based on four evaluative measures. We discovered a consistent cluster found in three of the four methods composed of predominantly female, younger disease onset (43% between ages 42-73) diagnosed with depression and anxiety, with a quicker rate of progression compared to the average across other clusters.

CONCLUSION

Each clustering approach produced substantially different clusters and K-Means performed the best out of the four methods based on the four evaluative criteria. However, the consistent appearance of one particular cluster across three of the four methods potentially suggests the presence of a distinct disease subtype that merits further exploration. Our study underlines the variability of the results obtained from different clustering approaches and the importance of systematically evaluating different approaches for identifying disease subtypes in complex EHR.

Extracting Disease-Relevant Features with Adversarial Regularization

Extracting hidden phenotypes is essential in medical data analysis because it facilitates disease subtyping, diagnosis, and understanding of disease etiology. Since the hidden phenotype is usually a low-dimensional representation that comprehensively describes the disease, we require a dimensionality-reduction method that captures as much disease-relevant information as possible. However, most unsupervised or self-supervised methods cannot achieve the goal because they learn a holistic representation containing both disease-relevant and disease-irrelevant information. Supervised methods can capture information that is predictive to the target clinical variable only, but the learned representation is usually not generalizable for the various aspects of the disease. Hence, we develop a dimensionality-reduction approach to extract Disease Relevant Features (DRFs) based on information theory. We propose to use clinical variables that weakly define the disease as so-called anchors. We derive a formulation that makes the DRF predictive of the anchors while forcing the remaining representation to be irrelevant to the anchors via adversarial regularization. We apply our method to a large-scale study of Chronic Obstructive Pulmonary Disease (COPD). Our experiment shows: (1) Learned DRFs are as predictive as the original representation in predicting the anchors, although it is in a significantly lower dimension. (2) Compared to supervised representation, the learned DRFs are more predictive to other relevant disease metrics that are not used during the training. (3) The learned DRFs are related to non-imaging biological measurements such as gene expressions, suggesting the DRFs include information related to the underlying biology of the disease.

Estimating redundancy in clinical text

The current mode of use of Electronic Health Records (EHR) elicits text redundancy. Clinicians often populate new documents by duplicating existing notes, then updating accordingly. Data duplication can lead to propagation of errors, inconsistencies and misreporting of care. Therefore, measures to quantify information redundancy play an essential role in evaluating innovations that operate on clinical narratives. This work is a quantitative examination of information redundancy in EHR notes. We present and evaluate two methods to measure redundancy: an information-theoretic approach and a lexicosyntactic and semantic model. Our first measure trains large Transformer-based language models using clinical text from a large openly available US-based ICU dataset and a large multi-site UK based Hospital. By comparing the information-theoretic efficient encoding of clinical text against open-domain corpora, we find that clinical text is ∼1.5× to ∼3× less efficient than open-domain corpora at conveying information. Our second measure, evaluates automated summarisation metrics Rouge and BERTScore to evaluate successive note pairs demonstrating lexicosyntactic and semantic redundancy, with averages from ∼43 to ∼65%.

A cluster analysis of patients with axial spondyloarthritis using tumour necrosis factor alpha inhibitors based on clinical characteristics

Background

This study aimed to classify the distinct group of patients with axial spondyloarthritis (SpA) on tumour necrosis factor alpha inhibitors (TNFi) according to the baseline characteristics using a clustering algorithm.

Methods

The clinical characteristics and demographic data of patients with axial SpA included in the Korean College of Rheumatology Biologics and Targeted Therapy registry were investigated. The patterns of disease manifestations were examined using divisive hierarchical cluster analysis. After clustering, we compared the clinical characteristics of patients and the drug survival of TNFi between the classified groups.

Results

A total of 1042 patients were analysed. The cluster analysis classified patients into two groups: axial group predominantly showing isolated axial manifestations (n = 828) and extra-axial group more frequently showing extra-axial symptoms (n = 214). Almost all extra-axial symptoms (peripheral arthritis, enthesitis, uveitis, and psoriasis) were more frequently observed in the extra-axial group than in the axial group. Moreover, patients in the extra-axial group had shorter disease duration, later disease onset, and higher disease activity than those in the axial group. The disease activity was comparable between the two groups after 1 year of treatment with TNFi. Interestingly, the extra-axial group had a lower drug survival with TNFi than the axial group (p = 0.001).

Conclusions

Cluster analysis of patients with axial SpA using TNFi classified two distinct clinical phenotypes. These clusters had different TNFi drug survival, clinical characteristics, and disease activity.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13075-021-02647-z.

Contrastive Learning Improves Critical Event Prediction in COVID-19 Patients

Deep Learning (DL) models typically require large-scale, balanced training data to be robust, generalizable, and effective in the context of healthcare. This has been a major issue for developing DL models for the coronavirus-disease 2019 (COVID-19) pandemic where data are highly class imbalanced. Conventional approaches in DL use cross-entropy loss (CEL) which often suffers from poor margin classification. We show that contrastive loss (CL) improves the performance of CEL especially in imbalanced electronic health records (EHR) data for COVID-19 analyses. We use a diverse EHR data set to predict three outcomes: mortality, intubation, and intensive care unit (ICU) transfer in hospitalized COVID-19 patients over multiple time windows. To compare the performance of CEL and CL, models are tested on the full data set and a restricted data set. CL models consistently outperform CEL models with differences ranging from 0.04 to 0.15 for AUPRC and 0.05 to 0.1 for AUROC.

Machine Learning Techniques for Personalised Medicine Approaches in Immune-Mediated Chronic Inflammatory Diseases: Applications and Challenges

In the past decade, the emergence of machine learning (ML) applications has led to significant advances towards implementation of personalised medicine approaches for improved health care, due to the exceptional performance of ML models when utilising complex big data. The immune-mediated chronic inflammatory diseases are a group of complex disorders associated with dysregulated immune responses resulting in inflammation affecting various organs and systems. The heterogeneous nature of these diseases poses great challenges for tailored disease management and addressing unmet patient needs. Applying novel ML techniques to the clinical study of chronic inflammatory diseases shows promising results and great potential for precision medicine applications in clinical research and practice. In this review, we highlight the clinical applications of various ML techniques for prediction, diagnosis and prognosis of autoimmune rheumatic diseases, inflammatory bowel disease, autoimmune chronic kidney disease, and multiple sclerosis, as well as ML applications for patient stratification and treatment selection. We highlight the use of ML in drug development, including target identification, validation and drug repurposing, as well as challenges related to data interpretation and validation, and ethical concerns related to the use of artificial intelligence in clinical research.

Discovering Key Topics From Short, Real-World Medical Inquiries via Natural Language Processing

Millions of unsolicited medical inquiries are received by pharmaceutical companies every year. It has been hypothesized that these inquiries represent a treasure trove of information, potentially giving insight into matters regarding medicinal products and the associated medical treatments. However, due to the large volume and specialized nature of the inquiries, it is difficult to perform timely, recurrent, and comprehensive analyses. Here, we combine biomedical word embeddings, non-linear dimensionality reduction, and hierarchical clustering to automatically discover key topics in real-world medical inquiries from customers. This approach does not require ontologies nor annotations. The discovered topics are meaningful and medically relevant, as judged by medical information specialists, thus demonstrating that unsolicited medical inquiries are a source of valuable customer insights. Our work paves the way for the machine-learning-driven analysis of medical inquiries in the pharmaceutical industry, which ultimately aims at improving patient care.