From Real-World Patient Data to Individualized Treatment Effects Using Machine Learning: Current and Future Methods to Address Underlying Challenges

Review of machine learning solutions for eating disorders

BACKGROUND

Eating Disorders (EDs) are one of the most complex psychiatric disorders, with significant impairment of psychological and physical health, and psychosocial functioning, and are associated with low rates of early detection, low recovery and high relapse rates. This underscores the need for better diagnostic and treatment methods.

OBJECTIVE

This narrative review explores current Machine Learning (ML) and Artificial Intelligence (AI) applications in the domain of EDs, with a specific emphasis on clinical management in treatment settings. The primary objective are to (i) decrease the knowledge gap between ED researchers and AI-practitioners, by presenting the current state-of-the-art AI applications (including models for causality) in different ED use-cases; (ii) identify limitations of these existing AI interventions and how to address them.

RESULTS

AI/ML methods have been applied in different ED use-cases, including ED risk factor identification and incidence prediction (including the analysis of social media content in the general population), diagnosis, monitoring patients and treatment response and prognosis in clinical populations. A comparative analysis of AI-techniques deployed in these use-cases have been performed, considering factors such as complexity, flexibility, functionality, explainability and adaptability to healthcare constraints.

CONCLUSION

Multiple restrictions have been identified in the existing methods in ML and Causality in terms of achieving actionable healthcare for ED, like lack of good quality and quantity of data for models to train on, while requiring models to be flexible, high-performing, yet being explainable and producing counterfactual explanations, for ensuring the fairness and trustworthiness of its decisions. We conclude that to overcome these limitations and for future AI research and application in clinical management of ED, (i) careful considerations are required with regards to AI-model selection, and (ii) joint efforts from ED researcher and patient community are essential in building better quality and quantity of dedicated ED datasets and secure AI-solution framework.

Artificial intelligence and headache

BACKGROUND AND METHODS

In this narrative review, we introduce key artificial intelligence (AI) and machine learning (ML) concepts, aimed at headache clinicians and researchers. Thereafter, we thoroughly review the use of AI in headache, based on a comprehensive literature search across PubMed, Embase and IEEExplore. Finally, we discuss limitations, as well as ethical and political perspectives.

RESULTS

We identified six main research topics. First, natural language processing can be used to effectively extract and systematize unstructured headache research data, such as from electronic health records. Second, the most common application of ML is for classification of headache disorders, typically based on clinical record data, or neuroimaging data, with accuracies ranging from around 60% to well over 90%. Third, ML is used for prediction of headache disease trajectories. Fourth, ML shows promise in forecasting of headaches using self-reported data such as triggers and premonitory symptoms, data from wearable sensors and external data. Fifth and sixth, ML can be used for prediction of treatment responses and inference of treatment effects, respectively, aiming to optimize and individualize headache management.

CONCLUSIONS

The potential uses of AI and ML in headache are broad, but, at present, many studies suffer from poor reporting and lack out-of-sample evaluation, and most models are not validated in a clinical setting.

Machine Learning Methods to Estimate Individualized Treatment Effects for Use in Health Technology Assessment

BACKGROUND

Recent developments in causal inference and machine learning (ML) allow for the estimation of individualized treatment effects (ITEs), which reveal whether treatment effectiveness varies according to patients' observed covariates. ITEs can be used to stratify health policy decisions according to individual characteristics and potentially achieve greater population health. Little is known about the appropriateness of available ML methods for use in health technology assessment.

METHODS

In this scoping review, we evaluate ML methods available for estimating ITEs, aiming to help practitioners assess their suitability in health technology assessment. We present a taxonomy of ML approaches, categorized by key challenges in health technology assessment using observational data, including handling time-varying confounding and time-to event data and quantifying uncertainty.

RESULTS

We found a wide range of algorithms for simpler settings with baseline confounding and continuous or binary outcomes. Not many ML algorithms can handle time-varying or unobserved confounding, and at the time of writing, no ML algorithm was capable of estimating ITEs for time-to-event outcomes while accounting for time-varying confounding. Many of the ML algorithms that estimate ITEs in longitudinal settings do not formally quantify uncertainty around the point estimates.

LIMITATIONS

This scoping review may not cover all relevant ML methods and algorithms as they are continuously evolving.

CONCLUSIONS

Existing ML methods available for ITE estimation are limited in handling important challenges posed by observational data when used for cost-effectiveness analysis, such as time-to-event outcomes, time-varying and hidden confounding, or the need to estimate sampling uncertainty around the estimates.

IMPLICATIONS

ML methods are promising but need further development before they can be used to estimate ITEs for health technology assessments.

HIGHLIGHTS

Estimating individualized treatment effects (ITEs) using observational data and machine learning (ML) can support personalized treatment advice and help deliver more customized information on the effectiveness and cost-effectiveness of health technologies.ML methods for ITE estimation are mostly designed for handling confounding at baseline but not time-varying or unobserved confounding. The few models that account for time-varying confounding are designed for continuous or binary outcomes, not time-to-event outcomes.Not all ML methods for estimating ITEs can quantify the uncertainty of their predictions.Future work on developing ML that addresses the concerns summarized in this review is needed before these methods can be widely used in clinical and health technology assessment-like decision making.

Modern approaches for evaluating treatment effect heterogeneity from clinical trials and observational data

In this paper, we review recent advances in statistical methods for the evaluation of the heterogeneity of treatment effects (HTE), including subgroup identification and estimation of individualized treatment regimens, from randomized clinical trials and observational studies. We identify several types of approaches using the features introduced in Lipkovich et al (Stat Med 2017;36: 136-196) that distinguish the recommended principled methods from basic methods for HTE evaluation that typically rely on rules of thumb and general guidelines (the methods are often referred to as common practices). We discuss the advantages and disadvantages of various principled methods as well as common measures for evaluating their performance. We use simulated data and a case study based on a historical clinical trial to illustrate several new approaches to HTE evaluation.

Development, validation, and usability evaluation of machine learning algorithms for predicting personalized red blood cell demand among thoracic surgery patients

INTRODUCTION

Preparing appropriate red blood cells (RBCs) before surgery is crucial for improving both the efficacy of perioperative workflow and patient safety. In particular, thoracic surgery (TS) is a procedure that requires massive transfusion with high variability for each patient. Hence, the precise prediction of RBC requirements for individual patients is becoming increasingly important. This study aimed to 1) develop and validate a machine learning algorithm for personalized RBC predictions for TS patients and 2) assess the usability of a clinical decision support system (CDSS) integrating this artificial intelligence model.

METHODS

Adult patients who underwent TS between January 2016 and October 2021 were included in this study. Multiple models were developed by employing both traditional statistical- and machine-learning approaches. The primary outcome evaluated the model's performance in predicting RBC requirements through root mean square error and adjusted R2. Surgeons and informaticians determined the precision MSBOS-Thoracic Surgery (pMSBOS-TS) algorithm through a consensus process. The usability of the pMSBOS-TS was assessed using the System Usability Scale (SUS) survey with 60 clinicians.

RESULTS

We identified 7,843 cases (6,200 for training and 1,643 for test sets) of TSs. Among the models with variable performance indices, the extreme gradient boosting model was selected as the pMSBOS-TS algorithm. The pMSBOS-TS model showed statistically significant lower root mean square error (mean: 3.203 and 95% confidence interval [CI]: 3.186-3.220) compared to the calculated Maximum Surgical Blood Ordering Schedule (MSBOS) and a higher adjusted R2 (mean: 0.399 and 95% CI: 0.395-0.403) compared to the calculated MSBOS, while requiring approximately 200 fewer packs for RBC preparation compared to the calculated MSBOS. The SUS score of the pMSBOS-TS CDSS was 72.5 points, indicating good acceptability.

CONCLUSIONS

We successfully developed the pMSBOS-TS capable of predicting personalized RBC transfusion requirements for perioperative patients undergoing TS.

Machine learning techniques to predict the risk of developing diabetic nephropathy: a literature review

Purpose

Diabetes is a major public health challenge with widespread prevalence, often leading to complications such as Diabetic Nephropathy (DN)-a chronic condition that progressively impairs kidney function. In this context, it is important to evaluate if Machine learning models can exploit the inherent temporal factor in clinical data to predict the risk of developing DN faster and more accurately than current clinical models.

Methods

Three different databases were used for this literature review: Scopus, Web of Science, and PubMed. Only articles written in English and published between January 2015 and December 2022 were included.

Results

We included 11 studies, from which we discuss a number of algorithms capable of extracting knowledge from clinical data, incorporating dynamic aspects in patient assessment, and exploring their evolution over time. We also present a comparison of the different approaches, their performance, advantages, disadvantages, interpretation, and the value that the time factor can bring to a more successful prediction of diabetic nephropathy.

Conclusion

Our analysis showed that some studies ignored the temporal factor, while others partially exploited it. Greater use of the temporal aspect inherent in Electronic Health Records (EHR) data, together with the integration of omics data, could lead to the development of more reliable and powerful predictive models.

Artificial intelligence to predict individualized outcome of acute ischemic stroke patients: The SIBILLA project

INTRODUCTION

Formulating reliable prognosis for ischemic stroke patients remains a challenging task. We aimed to develop an artificial intelligence model able to formulate in the first 24 h after stroke an individualized prognosis in terms of NIHSS.

PATIENTS AND METHODS

Seven hundred ninety four acute ischemic stroke patients were divided into a training (597) and testing (197) cohort. Clinical and instrumental data were collected in the first 24 h. We evaluated the performance of four machine-learning models (Random Forest, K-Nearest Neighbors, Support Vector Machine, XGBoost) in predicting NIHSS at discharge both in terms of variation between discharge and admission (regressor approach) and in terms of severity class namely NIHSS 0-5, 6-10, 11-20, >20 (classifier approach). We used Shapley Additive exPlanations values to weight features impact on predictions.

RESULTS

XGBoost emerged as the best performing model. The classifier and regressor approaches perform similarly in terms of accuracy (80% vs 75%) and f1-score (79% vs 77%) respectively. However, the regressor has higher precision (85% vs 68%) in predicting prognosis of very severe stroke patients (NIHSS > 20). NIHSS at admission and 24 hours, GCS at 24 hours, heart rate, acute ischemic lesion on CT-scan and TICI score were the most impacting features on the prediction.

DISCUSSION

Our approach, which employs an artificial intelligence based-tool, inherently able to continuously learn and improve its performance, could improve care pathway and support stroke physicians in the communication with patients and caregivers.

CONCLUSION

XGBoost reliably predicts individualized outcome in terms of NIHSS at discharge in the first 24 hours after stroke.

Machine learning models reveal distinct disease subgroups and improve diagnostic and prognostic accuracy for individuals with pathogenic SCN8A gain-of-function variants

Distinguishing clinical subgroups for patients suffering with diseases characterized by a wide phenotypic spectrum is essential for developing precision therapies. Patients with gain-of-function (GOF) variants in the SCN8A gene exhibit substantial clinical heterogeneity, viewed historically as a linear spectrum ranging from mild to severe. To test for hidden clinical subgroups, we applied two machine-learning algorithms to analyze a dataset of patient features collected by the International SCN8A Patient Registry. We used two research methodologies: a supervised approach that incorporated feature severity cutoffs based on clinical conventions, and an unsupervised approach employing an entirely data-driven strategy. Both approaches found statistical support for three distinct subgroups and were validated by correlation analyses using external variables. However, distinguishing features of the three subgroups within each approach were not concordant, suggesting a more complex phenotypic landscape. The unsupervised approach yielded strong support for a model involving three partially ordered subgroups rather than a linear spectrum. Application of these machine-learning approaches may lead to improved prognosis and clinical management of individuals with SCN8A GOF variants and provide insights into the underlying mechanisms of the disease.

A real-world exploration into clinical outcomes of direct oral anticoagulant therapy in people with chronic kidney disease: a large hospital-based study

BACKGROUND

There is limited evidence to support definite clinical outcomes of direct oral anticoagulant (DOAC) therapy in chronic kidney disease (CKD). By identifying the important variables associated with clinical outcomes following DOAC administration in patients in different stages of CKD, this study aims to assess this evidence gap.

METHODS

An anonymised dataset comprising 97,413 patients receiving DOAC therapy in a tertiary health setting was systematically extracted from the multidimensional electronic health records and prepared for analysis. Machine learning classifiers were applied to the prepared dataset to select the important features which informed covariate selection in multivariate logistic regression analysis.

RESULTS

For both CKD and non-CKD DOAC users, features such as length of stay, treatment days, and age were ranked highest for relevance to adverse outcomes like death and stroke. Patients with Stage 3a CKD had significantly higher odds of ischaemic stroke (OR 2.45, 95% Cl: 2.10-2.86; p = 0.001) and lower odds of all-cause mortality (OR 0.87, 95% Cl: 0.79-0.95; p = 0.001) on apixaban therapy. In patients with CKD (Stage 5) receiving apixaban, the odds of death were significantly lowered (OR 0.28, 95% Cl: 0.14-0.58; p = 0.001), while the effect on ischaemic stroke was insignificant.

CONCLUSIONS

A positive effect of DOAC therapy was observed in advanced CKD. Key factors influencing clinical outcomes following DOAC administration in patients in different stages of CKD were identified. These are crucial for designing more advanced studies to explore safer and more effective DOAC therapy for the population.

Analyzing population-level trials as N-of-1 trials: An application to gait

Studying individual causal effects of health interventions is important whenever intervention effects are heterogeneous between study participants. Conducting N-of-1 trials, which are single-person randomized controlled trials, is the gold standard for their analysis. As an alternative method, we propose to re-analyze existing population-level studies as N-of-1 trials, and use gait as a use case for illustration. Gait data were collected from 16 young and healthy participants under fatigued and non-fatigued, as well as under single-task (only walking) and dual-task (walking while performing a cognitive task) conditions. As a reference to the N-of-1 trials approach, we first computed standard population-level ANOVA models to evaluate differences in gait parameters (stride length and stride time) across conditions. Then, we estimated the effect of the interventions on gait parameters on the individual level through Bayesian repeated-measures models, viewing each participant as their own trial, and compared the results. The results illustrated that while few overall population-level effects were visible, individual-level analyses revealed differences between participants. Baseline values of the gait parameters varied largely among all participants, and the effects of fatigue and cognitive task were also heterogeneous, with some individuals showing effects in opposite directions. These differences between population-level and individual-level analyses were more pronounced for the fatigue intervention compared to the cognitive task intervention. Following our empirical analysis, we discuss re-analyzing population studies through the lens of N-of-1 trials more generally and highlight important considerations and requirements. Our work encourages future studies to investigate individual effects using population-level data.

SECRETS: Subject-efficient clinical randomized controlled trials using synthetic intervention

Background

The parallel-group randomized controlled trial (RCT) is commonly used in Phase-3 clinical trials to establish treatment effectiveness but requires hundreds-to-thousands of subjects, making it difficult to implement, which leads to high Phase-3 trial failure rates. One approach to increasing power of a trial is to augment data collected from an RCT with external data from prospective studies or prior RCTs. However, this requires that external data be comparable to data from the study of interest, a condition that does not hold for new interventions or populations being studied. Another approach is to lower sample size requirements by using the cross-over design, which measures individual treatment effects (ITEs) to remove inter-subject variability; however, this design is only suitable for chronic conditions and interventions with effects that wash out rapidly.

Method

We propose a novel and practical framework called SECRETS (Subject-Efficient Clinical Randomized Controlled Trials using Synthetic Intervention) to increase power of any parallel-group RCT by simulating the cross-over design using only data collected from the study. SECRETS first estimates ITEs across all subjects recruited to the RCT by using a state-of-the-art counterfactual estimation algorithm called synthetic intervention (SI). Since SI induces dependencies among the ITEs, we introduce a novel hypothesis testing strategy to test for treatment effectiveness.

Results

We show that SECRETS can increase the power of an RCT while maintaining comparable significance levels; in particular, on three real-world clinical RCTs (Phase-3 trials), SECRETS increases power over the baseline method by 6 - 54 % (average: 21.5%, standard deviation: 15.8%), thereby reducing the number of subjects needed to obtain a typically desired statistical operating point of 80% power and 5% significance level by 25 - 76 % (10-3,957 fewer subjects per arm). Our analyses show that SECRETS increases power by consistently reducing the variance of the average treatment effect, thereby mimicking the effects of a cross-over design.

Conclusion

SECRETS increases subject efficiency of an RCT by simulating the cross-over design using only data collected from the RCT; therefore, it is a feasible solution for increasing the trial's power, especially under settings where satisfying sample size requirements is difficult.

Artificial Intelligence for Quantitative Modeling in Drug Discovery and Development: An Innovation and Quality Consortium Perspective on Use Cases and Best Practices

Recent breakthroughs in artificial intelligence (AI) and machine learning (ML) have ushered in a new era of possibilities across various scientific domains. One area where these advancements hold significant promise is model-informed drug discovery and development (MID3). To foster a wider adoption and acceptance of these advanced algorithms, the Innovation and Quality (IQ) Consortium initiated the AI/ML working group in 2021 with the aim of promoting their acceptance among the broader scientific community as well as by regulatory agencies. By drawing insights from workshops organized by the working group and attended by key stakeholders across the biopharma industry, academia, and regulatory agencies, this white paper provides a perspective from the IQ Consortium. The range of applications covered in this white paper encompass the following thematic topics: (i) AI/ML-enabled Analytics for Pharmacometrics and Quantitative Systems Pharmacology (QSP) Workflows; (ii) Explainable Artificial Intelligence and its Applications in Disease Progression Modeling; (iii) Natural Language Processing (NLP) in Quantitative Pharmacology Modeling; and (iv) AI/ML Utilization in Drug Discovery. Additionally, the paper offers a set of best practices to ensure an effective and responsible use of AI, including considering the context of use, explainability and generalizability of models, and having human-in-the-loop. We believe that embracing the transformative power of AI in quantitative modeling while adopting a set of good practices can unlock new opportunities for innovation, increase efficiency, and ultimately bring benefits to patients.

Using Machine Learning to Individualize Treatment Effect Estimation: Challenges and Opportunities

The use of data from randomized clinical trials to justify treatment decisions for real-world patients is the current state of the art. It relies on the assumption that average treatment effects from the trial can be extrapolated to patients with personal and/or disease characteristics different from those treated in the trial. Yet, because of heterogeneity of treatment effects between patients and between the trial population and real-world patients, this assumption may not be correct for many patients. Using machine learning to estimate the expected conditional average treatment effect (CATE) in individual patients from observational data offers the potential for more accurate estimation of the expected treatment effects in each patient based on their observed characteristics. In this review, we discuss some of the challenges and opportunities for machine learning to estimate CATE, including ensuring identification assumptions are met, managing covariate shift, and learning without access to the true label of interest. We also discuss the potential applications as well as future work and collaborations needed to further improve identification and utilization of CATE estimates to increase patient benefit.

The Role of AI in Hospitals and Clinics: Transforming Healthcare in the 21st Century

As healthcare systems around the world face challenges such as escalating costs, limited access, and growing demand for personalized care, artificial intelligence (AI) is emerging as a key force for transformation. This review is motivated by the urgent need to harness AI's potential to mitigate these issues and aims to critically assess AI's integration in different healthcare domains. We explore how AI empowers clinical decision-making, optimizes hospital operation and management, refines medical image analysis, and revolutionizes patient care and monitoring through AI-powered wearables. Through several case studies, we review how AI has transformed specific healthcare domains and discuss the remaining challenges and possible solutions. Additionally, we will discuss methodologies for assessing AI healthcare solutions, ethical challenges of AI deployment, and the importance of data privacy and bias mitigation for responsible technology use. By presenting a critical assessment of AI's transformative potential, this review equips researchers with a deeper understanding of AI's current and future impact on healthcare. It encourages an interdisciplinary dialogue between researchers, clinicians, and technologists to navigate the complexities of AI implementation, fostering the development of AI-driven solutions that prioritize ethical standards, equity, and a patient-centered approach.

Using real-world evidence in haematology

Most new drug approvals are based on data from large randomized clinical trials (RCTs). However, there are sometimes contradictory conclusions from seemingly similar trials and generalizability of conclusions from these trials is limited. These considerations explain, in part, the gap between conclusions from data of RCTs and those from registries termed real world data (RWD). Recently, real-world evidence (RWE) from RWD processed by artificial intelligence has received increasing attention. We describe the potential of using RWD in haematology concluding RWE from RWD may complement data from RCTs to support regulatory decisions.

The practice and facilitators of, and barriers to, implementing individualized repositioning frequency: A national cross-sectional survey of critical care nurses

AIMS

To understand the clinical status of implementing individualized repositioning frequency-and its barriers and facilitators-among critical care nurses in China, in view of developing targeted intervention strategies and improving guideline implementation.

DESIGN

A cross-sectional survey.

METHODS

A self-report questionnaire was developed with reference to the Theoretical Domains Framework and administered to critical care nurses in 15 hospitals across eastern, southern, western, northern, and central geographical areas of China from 20 February 2023 to 16 March 2023. Data were collected for personal demographics, clinical practice status, and from Likert-type responses about barriers to and facilitators of implementing individualized repositioning frequency.

RESULTS

In total, 574 effective questionnaires were collected. Only 3.8 % of respondents reported that their hospital/ward uses an individualized repositioning frequency rule. Six facilitator domains identified were: social/professional role and identity; beliefs about capabilities; optimism; beliefs about consequences; reinforcement; and intentions. Seven barrier domains were: knowledge; skills; goals; memory, attention and decision processes; environmental context and resources; social influences; and behavioral regulation. Inferential analysis showed that critical care nurses who had higher degrees, more years of work, more environmental support, and more nursing experience were prone to being more positive in response to the implementation of individualized repositioning frequency.

CONCLUSION

The clinical practice status of implementing individualized repositioning frequency among critical care nurses in China is unsatisfactory. Implementation is essential but complex and is influenced by several factors. Theory-based suggestions for improving this situation are provided on the basis of identified barriers.

Flexible Machine Learning Estimation of Conditional Average Treatment Effects: A Blessing and a Curse

Causal inference from observational data requires untestable identification assumptions. If these assumptions apply, machine learning methods can be used to study complex forms of causal effect heterogeneity. Recently, several machine learning methods were developed to estimate the conditional average treatment effect (ATE). If the features at hand cannot explain all heterogeneity, the individual treatment effects can seriously deviate from the conditional ATE. In this work, we demonstrate how the distributions of the individual treatment effect and the conditional ATE can differ when a causal random forest is applied. We extend the causal random forest to estimate the difference in conditional variance between treated and controls. If the distribution of the individual treatment effect equals that of the conditional ATE, this estimated difference in variance should be small. If they differ, an additional causal assumption is necessary to quantify the heterogeneity not captured by the distribution of the conditional ATE. The conditional variance of the individual treatment effect can be identified when the individual effect is independent of the outcome under no treatment given the measured features. Then, in the cases where the individual treatment effect and conditional ATE distributions differ, the extended causal random forest can appropriately estimate the variance of the individual treatment effect distribution, whereas the causal random forest fails to do so.

Demystification of artificial intelligence for respiratory clinicians managing patients with obstructive lung diseases

INTRODUCTION

Asthma and chronic obstructive pulmonary disease (COPD) are leading causes of morbidity and mortality worldwide. Despite all available diagnostics and treatments, these conditions pose a significant individual, economic and social burden. Artificial intelligence (AI) promises to support clinical decision-making processes by optimizing diagnosis and treatment strategies of these heterogeneous and complex chronic respiratory diseases. Its capabilities extend to predicting exacerbation risk, disease progression and mortality, providing healthcare professionals with valuable insights for more effective care. Nevertheless, the knowledge gap between respiratory clinicians and data scientists remains a major constraint for wide application of AI and may hinder future progress. This narrative review aims to bridge this gap and encourage AI deployment by explaining its methodology and added value in asthma and COPD diagnosis and treatment.

AREAS COVERED

This review offers an overview of the fundamental concepts of AI and machine learning, outlines the key steps in building a model, provides examples of their applicability in asthma and COPD care, and discusses barriers to their implementation.

EXPERT OPINION

Machine learning can advance our understanding of asthma and COPD, enabling personalized therapy and better outcomes. Further research and validation are needed to ensure the development of clinically meaningful and generalizable models.

A Tutorial Introduction to Heterogeneous Treatment Effect Estimation with Meta-learners

Psychotherapy has been proven to be effective on average, though patients respond very differently to treatment. Understanding which characteristics are associated with treatment effect heterogeneity can help to customize therapy to the individual patient. In this tutorial, we describe different meta-learners, which are flexible algorithms that can be used to estimate personalized treatment effects. More specifically, meta-learners decompose treatment effect estimation into multiple prediction tasks, each of which can be solved by any machine learning model. We begin by reviewing necessary assumptions for interpreting the estimated treatment effects as causal, and then give an overview over key concepts of machine learning. Throughout the article, we use an illustrative data example to show how the different meta-learners can be implemented in R. We also point out how current popular practices in psychotherapy research fit into the meta-learning framework. Finally, we show how heterogeneous treatment effects can be analyzed, and point out some challenges in the implementation of meta-learners.

Bridging the Worlds of Pharmacometrics and Machine Learning

Precision medicine requires individualized modeling of disease and drug dynamics, with machine learning-based computational techniques gaining increasing popularity. The complexity of either field, however, makes current pharmacological problems opaque to machine learning practitioners, and state-of-the-art machine learning methods inaccessible to pharmacometricians. To help bridge the two worlds, we provide an introduction to current problems and techniques in pharmacometrics that ranges from pharmacokinetic and pharmacodynamic modeling to pharmacometric simulations, model-informed precision dosing, and systems pharmacology, and review some of the machine learning approaches to address them. We hope this would facilitate collaboration between experts, with complementary strengths of principled pharmacometric modeling and flexibility of machine learning leading to synergistic effects in pharmacological applications.

Causal inference in drug discovery and development

To discover new drugs is to seek and to prove causality. As an emerging approach leveraging human knowledge and creativity, data, and machine intelligence, causal inference holds the promise of reducing cognitive bias and improving decision-making in drug discovery. Although it has been applied across the value chain, the concepts and practice of causal inference remain obscure to many practitioners. This article offers a nontechnical introduction to causal inference, reviews its recent applications, and discusses opportunities and challenges of adopting the causal language in drug discovery and development.

Predicting non-improvement of symptoms in daily mental healthcare practice using routinely collected patient-level data: a machine learning approach

Objectives

Anxiety and mood disorders greatly affect the quality of life for individuals worldwide. A substantial proportion of patients do not sufficiently improve during evidence-based treatments in mental healthcare. It remains challenging to predict which patients will or will not benefit. Moreover, the limited research available on predictors of treatment outcomes comes from efficacy RCTs with strict selection criteria which may limit generalizability to a real-world context. The current study evaluates the performance of different machine learning (ML) models in predicting non-improvement in an observational sample of patients treated in routine specialized mental healthcare.

Methods

In the current longitudinal exploratory prediction study diagnosis-related, sociodemographic, clinical and routinely collected patient-reported quantitative outcome measures were acquired during treatment as usual of 755 patients with a primary anxiety, depressive, obsessive compulsive or trauma-related disorder in a specialized outpatient mental healthcare center. ML algorithms were trained to predict non-response (< 0.5 standard deviation improvement) in symptomatic distress 6 months after baseline. Different models were trained, including models with and without early change scores in psychopathology and well-being and models with a trimmed set of predictor variables. Performance of trained models was evaluated in a hold-out sample (30%) as a proxy for unseen data.

Results

ML models without early change scores performed poorly in predicting six-month non-response in the hold-out sample with Area Under the Curves (AUCs) < 0.63. Including early change scores slightly improved the models' performance (AUC range: 0.68-0.73). Computationally-intensive ML models did not significantly outperform logistic regression (AUC: 0.69). Reduced prediction models performed similar to the full prediction models in both the models without (AUC: 0.58-0.62 vs. 0.58-0.63) and models with early change scores (AUC: 0.69-0.73 vs. 0.68-0.71). Across different ML algorithms, early change scores in psychopathology and well-being consistently emerged as important predictors for non-improvement.

Conclusion

Accurately predicting treatment outcomes in a mental healthcare context remains challenging. While advanced ML algorithms offer flexibility, they showed limited additional value compared to traditional logistic regression in this study. The current study confirmed the importance of taking early change scores in both psychopathology and well-being into account for predicting longer-term outcomes in symptomatic distress.

Mining the contribution of intensive care clinical course to outcome after traumatic brain injury

Existing methods to characterise the evolving condition of traumatic brain injury (TBI) patients in the intensive care unit (ICU) do not capture the context necessary for individualising treatment. Here, we integrate all heterogenous data stored in medical records (1166 pre-ICU and ICU variables) to model the individualised contribution of clinical course to 6-month functional outcome on the Glasgow Outcome Scale -Extended (GOSE). On a prospective cohort (n = 1550, 65 centres) of TBI patients, we train recurrent neural network models to map a token-embedded time series representation of all variables (including missing values) to an ordinal GOSE prognosis every 2 h. The full range of variables explains up to 52% (95% CI: 50-54%) of the ordinal variance in functional outcome. Up to 91% (95% CI: 90-91%) of this explanation is derived from pre-ICU and admission information (i.e., static variables). Information collected in the ICU (i.e., dynamic variables) increases explanation (by up to 5% [95% CI: 4-6%]), though not enough to counter poorer overall performance in longer-stay (>5.75 days) patients. Highest-contributing variables include physician-based prognoses, CT features, and markers of neurological function. Whilst static information currently accounts for the majority of functional outcome explanation after TBI, data-driven analysis highlights investigative avenues to improve the dynamic characterisation of longer-stay patients. Moreover, our modelling strategy proves useful for converting large patient records into interpretable time series with missing data integration and minimal processing.

Machine learning in predicting outcomes for stroke patients following rehabilitation treatment: A systematic review

OBJECTIVE

This review aimed to summarize the use of machine learning for predicting the potential benefits of stroke rehabilitation treatments, to evaluate the risk of bias of predictive models, and to provide recommendations for future models.

MATERIALS AND METHODS

This systematic review was conducted in accordance with the PRISMA statement and the CHARMS checklist. The PubMed, Embase, Cochrane Library, Scopus, and CNKI databases were searched up to April 08, 2023. The PROBAST tool was used to assess the risk of bias of the included models.

RESULTS

Ten studies within 32 models met our inclusion criteria. The optimal AUC value of the included models ranged from 0.63 to 0.91, and the optimal R2 value ranged from 0.64 to 0.91. All of the included models were rated as having a high or unclear risk of bias, and most of them were downgraded due to inappropriate data sources or analysis processes.

DISCUSSION AND CONCLUSION

There remains much room for improvement in future modeling studies, such as high-quality data sources and model analysis. Reliable predictive models should be developed to improve the efficacy of rehabilitation treatment by clinicians.

Predictive Models for Health Deterioration: Understanding Disease Pathways for Personalized Medicine

Artificial intelligence (AI) and machine learning (ML) methods are currently widely employed in medicine and healthcare. A PubMed search returns more than 100,000 articles on these topics published between 2018 and 2022 alone. Notwithstanding several recent reviews in various subfields of AI and ML in medicine, we have yet to see a comprehensive review around the methods' use in longitudinal analysis and prediction of an individual patient's health status within a personalized disease pathway. This review seeks to fill that gap. After an overview of the AI and ML methods employed in this field and of specific medical applications of models of this type, the review discusses the strengths and limitations of current studies and looks ahead to future strands of research in this field. We aim to enable interested readers to gain a detailed impression of the research currently available and accordingly plan future work around predictive models for deterioration in health status.

Inferring Personalized Treatment Effect of Antihypertensives on Alzheimer's Disease Using Deep Learning

Alzheimer's disease (AD) is one of the leading causes of death in the United States, especially among the elderly. Recent studies have shown how hypertension is related to cognitive decline in elderly patients, which in turn leads to increased mortality as well as morbidity. There have been various studies that have looked at the effect of antihypertensive drugs in reducing cognitive decline, and their results have proved inconclusive. However, most of these studies assume the treatment effect is similar for all patients, thus considering only the average treatment effects of antihypertensive drugs. In this paper, we assume that the effect of antihypertensives on the onset of AD depends on patient characteristics. We develop a deep learning method called LASSO-Dragonnet to estimate the individualized treatment effects of each patient. We considered six antihypertensive drugs, and each of the six models considered one of the drugs as the treatment and the remaining as control. Our studies showed that although many antihypertensives have a positive impact in delaying AD onset on average, the impact varies from individual to individual, depending on their various characteristics. We also analyzed the importance of various covariates in such an estimation. Our results showed that the individualized treatment effects of each patient could be estimated accurately using a deep learning method, and that the importance of various covariates could be determined.

Strategies to Address Current Challenges in Real-World Evidence Generation in Japan

The generation of real-world evidence (RWE), which describes patient characteristics or treatment patterns using real-world data (RWD), is rapidly growing more popular as a tool for decision-making in Japan. The aim of this review was to summarize challenges to RWE generation in Japan related to pharmacoepidemiology, and to propose strategies to address some of these challenges. We first focused on data-related issues, including the lack of transparency of RWD sources, linkage across different care settings, definitions of clinical outcomes, and the overall assessment framework of RWD when used for research purposes. Next the study reviewed methodology-related challenges. As lack of design transparency impairs study reproducibility, transparent reporting of study design is critical for stakeholders. For this review, we considered different sources of biases and time-varying confounding, along with potential study design and methodological solutions. Additionally, the implementation of robust assessment of definition uncertainty, misclassification, and unmeasured confounders would enhance RWE credibility in light of RWD source-related limitations, and is being strongly considered by task forces in Japan. Overall, the development of guidance for best practices on data source selection, design transparency, and analytical methods to address different sources of biases and robustness in the process of RWE generation will enhance credibility for stakeholders and local decision-makers.

Assessing Heterogeneity of Treatment Effect in Real-World Data

Increasing availability of real-world data (RWD) generated from patient care enables the generation of evidence to inform clinical decisions for subpopulations of patients and perhaps even individuals. There is growing opportunity to identify important heterogeneity of treatment effects (HTE) in these subgroups. Thus, HTE is relevant to all with interest in patients' responses to interventions, including regulators who must make decisions about products when signals of harms arise postapproval and payers who make coverage decisions based on expected net benefit to their beneficiaries. Prior work discussed HTE in randomized studies. Here, we address methodological considerations when investigating HTE in observational studies. We propose 4 primary goals of HTE analyses and the corresponding approaches in the context of RWD: to confirm subgroup effects, to describe the magnitude of HTE, to discover clinically important subgroups, and to predict individual effects. We discuss other possible goals including exploring prognostic score- and propensity score-based treatment effects, and testing the transportability of trial results to populations different from trial participants. Finally, we outline methodological needs for enhancing real-world HTE analysis.

Machine learning-based prediction of 1-year mortality in hypertensive patients undergoing coronary revascularization surgery

BACKGROUND

Machine learning (ML) has shown promising results in all fields of medicine, including preventive cardiology. Hypertensive patients are at higher risk of mortality after coronary artery bypass graft (CABG) surgery; thus, we aimed to design and evaluate five ML models to predict 1-year mortality among hypertensive patients who underwent CABG.

HYOTHESIS

ML algorithms can significantly improve mortality prediction after CABG.

METHODS

Tehran Heart Center's CABG data registry was used to extract several baseline and peri-procedural characteristics and mortality data. The best features were chosen using random forest (RF) feature selection algorithm. Five ML models were developed to predict 1-year mortality: logistic regression (LR), RF, artificial neural network (ANN), extreme gradient boosting (XGB), and naïve Bayes (NB). The area under the curve (AUC), sensitivity, and specificity were used to evaluate the models.

RESULTS

Among the 8,493 hypertensive patients who underwent CABG (mean age of 68.27 ± 9.27 years), 303 died in the first year. Eleven features were selected as the best predictors, among which total ventilation hours and ejection fraction were the leading ones. LR showed the best prediction ability with an AUC of 0.82, while the least AUC was for the NB model (0.79). Among the subgroups, the highest AUC for LR model was for two age range groups (50-59 and 80-89 years), overweight, diabetic, and smoker subgroups of hypertensive patients.

CONCLUSIONS

All ML models had excellent performance in predicting 1-year mortality among CABG hypertension patients, while LR was the best regarding AUC. These models can help clinicians assess the risk of mortality in specific subgroups at higher risk (such as hypertensive ones).

Color-Coding Method Reveals Enhancement of Stereotypic Locomotion by Phenazepam in Rat Open Field Test

One of the most important tasks in neuroscience is the search for theoretical foundations for the development of methods for diagnosing and treating neurological pathology, and for assessing the effect of pharmacological drugs on the nervous system. Specific behavioral changes associated with exposure to systemic influences have been invisible to the human eye for a long time. A similar pattern of changes is characteristic of phenazepam, a drug with a wide range of effects on the brain. In this study, we used a color-coding method, which consists of combining three time positions in one image, the present (0 s), the near future (0.33 s) and the far future (1.6 s). This method made it possible to identify movement patterns, such as the initialization of ahead movements, side turns and 180° turns (back), and also to determine the degree of predictability of future movements. The obtained data revealed a decrease in the number of turns to the sides while maintaining ahead movement, as well as an increase in the predictability of movements in rats under the influence of phenazepam. Thus, sedative doses of phenazepam do not exhibit general depression of brain functions, but the inhibition of specific centers, including the medial prefrontal cortex and postsubiculum, which are involved in stereotypic locomotive behavior.

Variational Temporal Deconfounder for Individualized Treatment Effect Estimation with Longitudinal Observational Data

Purpose

This paper proposes a new approach, Variational Temporal Deconfounder (VTD), for estimating individualized treatment effects (ITE) from longitudinal observational data, where we address the hidden confounding issues by using proxies (i.e., surrogate variables that serve for unobservable variables).

Methods

We build VTD by incorporating a variational recurrent autoencoder that learns the latent encodings of hidden confounders from observed proxies and an ITE estimation network that takes the learned hidden encodings to predict the probability of receiving treatments and potential outcomes.

Results

We test VTD on both synthetic and real-world clinical data, and the results from synthetic data experiments demonstrate VTD's effectiveness in deconfounding by outperforming existing methods, while results from two real-world datasets (i.e., Medical Information Mart for Intensive Care version III [MIMIC-III] and the National Alzheimer's Coordinating Center [NACC] database) suggest that the performance of the VTD model outperforms existing baseline models, however, varies depending on the assumptions of underlying causal structures and availability of proxies for hidden confounders.

Conclusion

The VTD offers a unique solution to address the confounding bias without the "unconfoundedness" assumption when estimating the ITE from longitudinal observational data. The elimination of the requirement for the "unconfoundedness" assumption makes the VTD more versatile and practical in real-world clinical applications of personalized medicine.

Challenges and Possible Solutions to Direct-Acting Oral Anticoagulants (DOACs) Dosing in Patients with Extreme Bodyweight and Renal Impairment

This article aims to highlight the dosing issues of direct oral anticoagulants (DOACs) in patients with renal impairment and/or obesity in an attempt to develop solutions employing advanced data-driven techniques. DOACs have become widely accepted by clinicians worldwide because of their superior clinical profiles, more predictable pharmacokinetics, and hence more convenient dosing relative to other anticoagulants. However, the optimal dosing of DOACs in extreme bodyweight  patients and patients with renal impairment is difficult to achieve using the conventional dosing approach. The standard dosing approach (fixed-dose) is based on limited data from clinical studies. The existing formulae (models) for determining the appropriate doses for these patient groups leads to suboptimal dosing. This problem of mis-dosing is worsened by the lack of standardized laboratory parameters for monitoring the exposure to DOACs in renal failure and extreme bodyweight patients. Model-informed precision dosing (MIPD) encompasses a range of techniques like machine learning and pharmacometrics modelling, which could uncover key variables and relationships as well as shed more light on the pharmacokinetics and pharmacodynamics of DOACs in patients with extreme bodyweight or renal impairment. Ultimately, this individualized approach-if implemented in clinical practice-could optimise dosing for the DOACs for better safety and efficacy.

Emulate randomized clinical trials using heterogeneous treatment effect estimation for personalized treatments: Methodology review and benchmark

Big data and (deep) machine learning have been ambitious tools in digital medicine, but these tools focus mainly on association. Intervention in medicine is about the causal effects. The average treatment effect has long been studied as a measure of causal effect, assuming that all populations have the same effect size. However, no "one-size-fits-all" treatment seems to work in some complex diseases. Treatment effects may vary by patient. Estimating heterogeneous treatment effects (HTE) may have a high impact on developing personalized treatment. Lots of advanced machine learning models for estimating HTE have emerged in recent years, but there has been limited translational research into the real-world healthcare domain. To fill the gap, we reviewed and compared eleven recent HTE estimation methodologies, including meta-learner, representation learning models, and tree-based models. We performed a comprehensive benchmark experiment based on nationwide healthcare claim data with application to Alzheimer's disease drug repurposing. We provided some challenges and opportunities in HTE estimation analysis in the healthcare domain to close the gap between innovative HTE models and deployment to real-world healthcare problems.

The potential role of machine learning in modelling advanced chronic liver disease

The use of artificial intelligence is rapidly increasing in medicine to support clinical decision making mostly through diagnostic and prediction models. Such models derive from huge databases (big data) including a large variety of health-related individual patient data (input) and the corresponding diagnosis and/or outcome (labels). Various types of algorithms (e.g. neural networks) based on powerful computational ability (machine), allow to detect the relationship between input and labels (learning). More complex algorithms, like recurrent neural network can learn from previous as well as actual input (deep learning) and are used for more complex tasks like imaging analysis and personalized (bespoke) medicine. The prompt availability of big data makes that artificial intelligence can provide rapid answers to questions that would require years of traditional clinical research. It may therefore be a key tool to overcome several major gaps in the model of advanced chronic liver disease, mostly transition from mild to clinically significant portal hypertension, the impact of acute decompensation and the role of further decompensation and treatment efficiency. However, several limitations of artificial intelligence should be overcome before its application in clinical practice. Assessment of the risk of bias, understandability of the black boxes developing the models and models' validation are the most important areas deserving clarification for artificial intelligence to be widely accepted from physicians and patients.

The Use of Machine Learning to Reduce Overtreatment of the Axilla in Breast Cancer: Retrospective Cohort Study

Background

Patients with early breast cancer undergoing primary surgery, who have low axillary nodal burden, can safely forego axillary node clearance (ANC). However, routine use of axillary ultrasound (AUS) leads to 43% of patients in this group having ANC unnecessarily, following a positive AUS. The intersection of machine learning with medicine can provide innovative ways to understand specific risks within large patient data sets, but this has not yet been trialed in the arena of axillary node management in breast cancer.

Objective

The objective of this study was to assess if machine learning techniques could be used to improve preoperative identification of patients with low and high axillary metastatic burden.

Methods

A single-center retrospective analysis was performed on patients with breast cancer who had a preoperative AUS, and the specificity and sensitivity of AUS were calculated. Standard statistical methods and machine learning methods, including artificial neural network, naive Bayes, support vector machine, and random forest, were applied to the data to see if they could improve the accuracy of preoperative AUS to better discern high and low axillary burden.

Results

The study included 459 patients; 142 (31%) had a positive AUS; among this group, 88 (62%) had 2 or fewer macrometastatic nodes at ANC. Logistic regression outperformed AUS (specificity 0.950 vs 0.809). Of all the methods, the artificial neural network had the highest accuracy (0.919). Interestingly, AUS had the highest sensitivity of all methods (0.777), underlining its utility in this setting.

Conclusions

We demonstrated that machine learning improves identification of the important subgroup of patients with no palpable axillary disease, positive ultrasound, and more than 2 metastatically involved nodes. A negative ultrasound in patients with no palpable lymphadenopathy is highly indicative of low axillary burden, and it is unclear whether sentinel node biopsy adds value in this situation. Further studies with larger patient numbers focusing on specific breast cancer subgroups are required to refine these techniques in this setting.

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.

Application of Electronic Health Record Text Mining: Real-World Tolerability, Safety, and Efficacy of Adjuvant Melanoma Treatments

Introduction: Nivolumab (N), pembrolizumab (P), and dabrafenib plus trametinib (D + T) have been registered as adjuvant treatments for resected stage III and IV melanoma since 2018. Electronic health records (EHRs) are a real-world data source that can be used to review treatments in clinical practice. In this study, we applied EHR text-mining software to evaluate the real-world tolerability, safety, and efficacy of adjuvant melanoma treatments. Methods: Adult melanoma patients receiving adjuvant treatment between January 2019 and October 2021 at the Leiden University Medical Center, the Netherlands, were included. CTcue text-mining software (v3.1.0, CTcue B.V., Amsterdam, The Netherlands) was used to construct rule-based queries and perform context analysis for patient inclusion and data collection from structured and unstructured EHR data. Results: In total, 122 patients were included: 54 patients treated with nivolumab, 48 with pembrolizumab, and 20 with D + T. Significantly more patients discontinued treatment due to toxicity on D + T (N: 16%, P: 6%, D + T: 40%), and X2 (6, n = 122) = 14.6 and p = 0.024. Immune checkpoint inhibitors (ICIs) mainly showed immune-related treatment-limiting adverse events (AEs), and chronic thyroid-related AE occurred frequently (hyperthyroidism: N: 15%, P: 13%, hypothyroidism: N: 20%, P: 19%). Treatment-limiting toxicity from D + T was primarily a combination of reversible AEs, including pyrexia and fatigue. The 1-year recurrence-free survival was 70.3% after nivolumab, 72.4% after pembrolizumab, and 83.0% after D + T. Conclusions: Text-mining EHR is a valuable method to collect real-world data to evaluate adjuvant melanoma treatments. ICIs were better tolerated than D + T, in line with RCT results. For BRAF+ patients, physicians must weigh the higher risk of reversible treatment-limiting AEs of D + T against the risk of long-term immune-related AEs.

Model enhanced reinforcement learning to enable precision dosing: A theoretical case study with dosing of propofol

Extending the potential of precision dosing requires evaluating methodologies offering more flexibility and higher degree of personalization. Reinforcement learning (RL) holds promise in its ability to integrate multidimensional data in an adaptive process built toward efficient decision making centered on sustainable value creation. For general anesthesia in intensive care units, RL is applied and automatically adjusts dosing through monitoring of patient's consciousness. We further explore the problem of optimal control of anesthesia with propofol by combining RL with state-of-the-art tools used to inform dosing in drug development. In particular, we used pharmacokinetic-pharmacodynamic (PK-PD) modeling as a simulation engine to generate experience from dosing scenarios, which cannot be tested experimentally. Through simulations, we show that, when learning from retrospective trial data, more than 100 patients are needed to reach an accuracy within the range of what is achieved with a standard dosing solution. However, embedding a model of drug effect within the RL algorithm improves accuracy by reducing errors to target by 90% through learning to take dosing actions maximizing long-term benefit. Data residual variability impacts accuracy while the algorithm efficiently coped with up to 50% interindividual variability in the PK and 25% in the PD model's parameters. We illustrate how extending the state definition of the RL agent with meaningful variables is key to achieve high accuracy of optimal dosing policy. These results suggest that RL constitutes an attractive approach for precision dosing when rich data are available or when complemented with synthetic data from model-based tools used in model-informed drug development.

OnAI-Comp: An Online AI Experts Competing Framework for Early Sepsis Detection

Sepsis is a major public concern due to its high mortality, morbidity, and financial cost. There are many existing works of early sepsis prediction using different machine learning models to mitigate the outcomes brought by sepsis. In the practical scenario, the dataset grows dynamically as new patients visit the hospital. Most existing models, being "offline" models and having used retrospective observational data, cannot be updated and improved dynamically using the new observational data. Incorporating the new data to improve the offline models requires retraining the model, which is very computationally expensive. To solve the challenge mentioned above, we propose an Online Artificial Intelligence Experts Competing Framework (OnAI-Comp) for early sepsis detection using an online learning algorithm called Multi-armed Bandit. We selected several machine learning models as the artificial intelligence experts and used average regret to evaluate the performance of our model. The experimental analysis demonstrated that our model would converge to the optimal strategy in the long run. Meanwhile, our model can provide clinically interpretable predictions using existing local interpretable model-agnostic explanation technologies, which can aid clinicians in making decisions and might improve the probability of survival.

Inferring heterogeneous treatment effects of work zones on crashes

The increasing number of work zone crashes has been a significant concern for road users, transportation agencies, and researchers. Crashes can be caused by work zones, and this effect changes across different work zone configurations, traffic volumes, roadway functional classifications, and weather conditions. This is typically represented by Crash Modification Functions (CMFunctions). However, current methods for developing work zone CMFunctions have two major limitations: (1) They focus on analyzing statistical associations and fail to mitigate the confounding bias due to possible unobservable roadway characteristics; and (2) They cannot address CMFunctions of multiple variables simultaneously, such as weather and traffic conditions, since they are represented using mixed data types (continuous and categorical) that could potentially affect the causal effect of work zones on crashes. In this study, we develop a method that utilizes causal forest with fixed-effect modeling to mitigate the confounding bias while identifying CMFunctions conditioning on various environmental characteristics, including work zone configurations, traffic volume, roadway functional classification, and weather conditions. The developed method was applied to 3378 work zones that occurred in Pennsylvania between 2015 and 2017. The results were validated via a series of robustness tests. The validations demonstrate that this method can mitigate the confounding bias and identify CMFunctions of multiple variables. The results also show that the causal effect of a work zone on crash occurrence is significantly positive (p<0.05) on roadways with high traffic volumes (e.g., > 20,000 vehicles per day) and on medium length (e.g., 2000 to 5000 m) work zones. It appears that having medium-long (e.g., between 6000 and 8000 m) work zones or long duration (e.g., longer than 4 h) work zones do not necessarily lead to extra crashes.

Machine learning algorithms for predicting mortality after coronary artery bypass grafting

Background

As the era of big data analytics unfolds, machine learning (ML) might be a promising tool for predicting clinical outcomes. This study aimed to evaluate the predictive ability of ML models for estimating mortality after coronary artery bypass grafting (CABG).

Materials and methods

Various baseline and follow-up features were obtained from the CABG data registry, established in 2005 at Tehran Heart Center. After selecting key variables using the random forest method, prediction models were developed using: Logistic Regression (LR), Support Vector Machine (SVM), Naïve Bayes (NB), K-Nearest Neighbors (KNN), Extreme Gradient Boosting (XGBoost), and Random Forest (RF) algorithms. Area Under the Curve (AUC) and other indices were used to assess the performance.

Results

A total of 16,850 patients with isolated CABG (mean age: 67.34 ± 9.67 years) were included. Among them, 16,620 had one-year follow-up, from which 468 died. Eleven features were chosen to train the models. Total ventilation hours and left ventricular ejection fraction were by far the most predictive factors of mortality. All the models had AUC &gt; 0.7 (acceptable performance) for 1-year mortality. Nonetheless, LR (AUC = 0.811) and XGBoost (AUC = 0.792) outperformed NB (AUC = 0.783), RF (AUC = 0.783), SVM (AUC = 0.738), and KNN (AUC = 0.715). The trend was similar for two-to-five-year mortality, with LR demonstrating the highest predictive ability.

Conclusion

Various ML models showed acceptable performance for estimating CABG mortality, with LR illustrating the highest prediction performance. These models can help clinicians make decisions according to the risk of mortality in patients undergoing CABG.

Causal machine learning for healthcare and precision medicine

Causal machine learning (CML) has experienced increasing popularity in healthcare. Beyond the inherent capabilities of adding domain knowledge into learning systems, CML provides a complete toolset for investigating how a system would react to an intervention (e.g. outcome given a treatment). Quantifying effects of interventions allows actionable decisions to be made while maintaining robustness in the presence of confounders. Here, we explore how causal inference can be incorporated into different aspects of clinical decision support systems by using recent advances in machine learning. Throughout this paper, we use Alzheimer's disease to create examples for illustrating how CML can be advantageous in clinical scenarios. Furthermore, we discuss important challenges present in healthcare applications such as processing high-dimensional and unstructured data, generalization to out-of-distribution samples and temporal relationships, that despite the great effort from the research community remain to be solved. Finally, we review lines of research within causal representation learning, causal discovery and causal reasoning which offer the potential towards addressing the aforementioned challenges.

StudyU: A Platform for Designing and Conducting Innovative Digital N-of-1 Trials

N-of-1 trials are the gold standard study design to evaluate individual treatment effects and derive personalized treatment strategies. Digital tools have the potential to initiate a new era of N-of-1 trials in terms of scale and scope, but fully functional platforms are not yet available. Here, we present the open source StudyU platform, which includes the StudyU Designer and StudyU app. With the StudyU Designer, scientists are given a collaborative web application to digitally specify, publish, and conduct N-of-1 trials. The StudyU app is a smartphone app with innovative user-centric elements for participants to partake in trials published through the StudyU Designer to assess the effects of different interventions on their health. Thereby, the StudyU platform allows clinicians and researchers worldwide to easily design and conduct digital N-of-1 trials in a safe manner. We envision that StudyU can change the landscape of personalized treatments both for patients and healthy individuals, democratize and personalize evidence generation for self-optimization and medicine, and can be integrated in clinical practice.

Artificial Neural Network vs. Pharmacometric Model for Population Prediction of Plasma Concentration in Real-World Data: A Case Study on Valproic Acid

We compared the predictive performance of an artificial neural network to traditional pharmacometric modeling for population prediction of plasma concentrations of valproate in real-world data. We included individuals aged 65 years or older with epilepsy who redeemed their first prescription of valproate after the diagnosis of epilepsy and had at least one valproate plasma concentration measured. A long short-term memory neural network (LSTM) was developed using the training data set to fit the LSTM and the test data set to validate the model. Predictions from the LSTM were compared with those obtained from the population predictions from a pharmacometric model by Birnbaum et al. which had the best predictive performance for population predictions of valproate concentrations in Danish databases. We used the cutoff of ± 20 mg/L of prediction error to define good predictions. A total of 1,252 individuals were included in the study. The LSTM fitted using the training data set had poor predictive performance in the test data set, but better than that of the pharmacometric model. The proportion of individuals with at least one predicted concentration within ± 20 mg/L of observed concentration was largest in case of the LSTM (64.4%, 95% confidence interval (CI): 58.4-70.2%) compared with the pharmacometric model by Birnbaum et al. (49.8%, 95% CI: 47.0-52.6%). LSTM shows better predictive performance to predict valproate plasma concentrations compared with a traditional pharmacometric model in the investigated setting with real-world data in older patients with epilepsy where information on exact timepoints for both dosing and plasma concentration measurement are missing.

Real world data and data science in medical research: present and future

Real world data (RWD) are generating greater interest in recent times despite being not new. There are various purposes of the RWD analytics in medical research as follows: effectiveness and safety of medical treatment, epidemiology such as incidence and prevalence of disease, burden of disease, quality of life and activity of daily living, medical costs, etc. The RWD research in medicine is a mixture of digital transformation, statistics or data science, public health, and regulatory science. Most of the articles describing the RWD or real-world evidence (RWE) in medical research cover only a portion of these specializations, which might lead to an incomplete understanding of the RWD. This article summarizes the overview and challenges of the RWD analysis in medical fields from methodological perspectives. As the first step for the RWD analysis, data source of the RWD should be comprehended. The progress of the RWD is closely related to the digitization, especially of medical administrative data and medical records. Second, the selection of appropriate statistical and epidemiological methods is highly critical for an RWD analysis than those for randomized clinical trials. This is because it contains greater varieties of bias, which should be controlled by balancing the underlying risk between treatment groups. Last, the future of the RWD is discussed in terms of overcoming limited data by proxy confounders, using unstructured text data, linking of multiple databases, using the RWD or RWE for a regulatory purpose, and evaluating values and new aspects in medical research brought by the RWD.

Implementation of prognostic machine learning algorithms in paediatric chronic respiratory conditions: a scoping review

Machine learning (ML) holds great potential for predicting clinical outcomes in heterogeneous chronic respiratory diseases (CRD) affecting children, where timely individualised treatments offer opportunities for health optimisation. This paper identifies rate-limiting steps in ML prediction model development that impair clinical translation and discusses regulatory, clinical and ethical considerations for ML implementation. A scoping review of ML prediction models in paediatric CRDs was undertaken using the PRISMA extension scoping review guidelines. From 1209 results, 25 articles published between 2013 and 2021 were evaluated for features of a good clinical prediction model using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) guidelines.

Most of the studies were in asthma (80%), with few in cystic fibrosis (12%), bronchiolitis (4%) and childhood wheeze (4%). There were inconsistencies in model reporting and studies were limited by a lack of validation, and absence of equations or code for replication. Clinician involvement during ML model development is essential and diversity, equity and inclusion should be assessed at each step of the ML pipeline to ensure algorithms do not promote or amplify health disparities among marginalised groups. As ML prediction studies become more frequent, it is important that models are rigorously developed using published guidelines and take account of regulatory frameworks which depend on model complexity, patient safety, accountability and liability.

Translating pharmacology models effectively to predict therapeutic benefit

Many in vitro and in vivo models are used in pharmacological research to evaluate the role of targeted proteins in a disease. Understanding the translational relevance and limitation of these models for analyzing the disposition, pharmacokinetic/pharmacodynamic (PK/PD) profile, mechanism, and efficacy of a drug, is essential when selecting the most appropriate model of the disease of interest and predicting clinically efficacious doses of the investigational drug. Here, we review selected animal models used in ophthalmology, infectious diseases, oncology, autoimmune diseases, and neuroscience. Each area has specific challenges around translatability and determination of an efficacious dose: new patient-specific dosing methods could help overcome these limitations.

Diffused responsibility: attributions of responsibility in the use of AI-driven clinical decision support systems

Good decision-making is a complex endeavor, and particularly so in a health context. The possibilities for day-to-day clinical practice opened up by AI-driven clinical decision support systems (AI-CDSS) give rise to fundamental questions around responsibility. In causal, moral and legal terms the application of AI-CDSS is challenging existing attributions of responsibility. In this context, responsibility gaps are often identified as main problem. Mapping out the changing dynamics and levels of attributing responsibility, we argue in this article that the application of AI-CDSS causes diffusions of responsibility with respect to a causal, moral, and legal dimension. Responsibility diffusion describes the situation where multiple options and several agents can be considered for attributing responsibility. Using the example of an AI-driven ‘digital tumor board’, we illustrate how clinical decision-making is changed and diffusions of responsibility take place. Not denying or attempting to bridge responsibility gaps, we argue that dynamics and ambivalences are inherent in responsibility, which is based on normative considerations such as avoiding experiences of disregard and vulnerability of human life, which are inherently accompanied by a moment of uncertainty, and is characterized by revision openness. Against this background and to avoid responsibility gaps, the article concludes with suggestions for managing responsibility diffusions in clinical decision-making with AI-CDSS.