Machine learning for predicting readmission risk among the frail: Explainable AI for healthcare

Machine Learning Models for Frailty Classification of Older Adults in Northern Thailand: Model Development and Validation Study

Background

Frailty is defined as a clinical state of increased vulnerability due to the age-associated decline of an individual's physical function resulting in increased morbidity and mortality when exposed to acute stressors. Early identification and management can reverse individuals with frailty to being robust once more. However, we found no integration of machine learning (ML) tools and frailty screening and surveillance studies in Thailand despite the abundance of evidence of frailty assessment using ML globally and in Asia.

Objective

We propose an approach for early diagnosis of frailty in community-dwelling older individuals in Thailand using an ML model generated from individual characteristics and anthropometric data.

Methods

Datasets including 2692 community-dwelling Thai older adults in Lampang from 2016 and 2017 were used for model development and internal validation. The derived models were externally validated with a dataset of community-dwelling older adults in Chiang Mai from 2021. The ML algorithms implemented in this study include the k-nearest neighbors algorithm, random forest ML algorithms, multilayer perceptron artificial neural network, logistic regression models, gradient boosting classifier, and linear support vector machine classifier.

Results

Logistic regression showed the best overall discrimination performance with a mean area under the receiver operating characteristic curve of 0.81 (95% CI 0.75-0.86) in the internal validation dataset and 0.75 (95% CI 0.71-0.78) in the external validation dataset. The model was also well-calibrated to the expected probability of the external validation dataset.

Conclusions

Our findings showed that our models have the potential to be utilized as a screening tool using simple, accessible demographic and explainable clinical variables in Thai community-dwelling older persons to identify individuals with frailty who require early intervention to become physically robust.

Predicting Readmission Among High-Risk Discharged Patients Using a Machine Learning Model With Nursing Data: Retrospective Study

Background

Unplanned readmissions increase unnecessary health care costs and reduce the quality of care. It is essential to plan the discharge care from the beginning of hospitalization to reduce the risk of readmission. Machine learning-based readmission prediction models can support patients' preemptive discharge care services with improved predictive power.

Objective

This study aimed to develop a readmission early prediction model utilizing nursing data for high-risk discharge patients.

Methods

This retrospective study included the electronic medical records of 12,977 patients with 1 of the top 6 high-risk readmission diseases at a tertiary hospital in Seoul from January 2018 to January 2020. We used demographic, clinical, and nursing data to construct a prediction model. We constructed unplanned readmission prediction models by dividing them into Model 1 and Model 2. Model 1 used early hospitalization data (up to 1 day after admission), and Model 2 used all the data. To improve the performance of the machine learning method, we performed 5-fold cross-validation and utilized adaptive synthetic sampling to address data imbalance. The 6 algorithms of logistic regression, random forest, decision tree, XGBoost, CatBoost, and multiperceptron layer were employed to develop predictive models. The analysis was conducted using Python Language Reference, version 3.11.3. (Python Software Foundation).

Results

In Model 1, among the 6 prediction model algorithms, the random forest model had the best result, with an area under the receiver operating characteristic (AUROC) curve of 0.62. In Model 2, the CatBoost model had the best result, with an AUROC of 0.64. BMI, systolic blood pressure, and age consistently emerged as the most significant predictors of readmission risk across Models 1 and 2. Model 1, which enabled early readmission prediction, showed a higher proportion of nursing data variables among its important predictors compared to Model 2.

Conclusions

Machine learning-based readmission prediction models utilizing nursing data provide basic data for evidence-based clinical decision support for high-risk discharge patients with complex conditions and facilitate early intervention. By integrating nursing data containing diverse patient information, these models can provide more comprehensive risk assessment and improve patient outcomes.

Integrating real-world data and machine learning: A framework to assess covariate importance in real-world use of alternative intravenous dosing regimens for atezolizumab

The increase in the availability of real-world data (RWD), in combination with advances in machine learning (ML) methods, provides a unique opportunity for the integration of the two to explore complex clinical pharmacology questions. Here we present a recently developed RWD/ML framework that utilizes ML algorithms to understand the influence and importance of various covariates on the use of a given dose and schedule for drugs that have multiple approved dosing regimens. To demonstrate the application of this framework, we present atezolizumab as a use case on account of its three approved alternative intravenous (IV) dosing regimens. As expected, the real-world use of atezolizumab has generally been increasing since 2016 for the 1200 mg every 3 weeks regimen and since 2019 for the 1680 mg every 4 weeks regimen. Out of the ML algorithms evaluated, XGBoost performed the best, as measured by the area under the precision-recall curve, with an emphasis on the under-sampled class given the imbalance in the data. The importance of features was measured by Shapley Additive exPlanations (SHAP) values and showed metastatic breast cancer and use of protein-bound paclitaxel as the most correlated with the use of 840 mg every 2 weeks. Although patient usage data for alternative IV dosing regimens are still maturing, these analyses provide initial insights on the use of atezolizumab and set up a framework for the re-analysis of atezolizumab (at a future data cut) as well as application to other molecules with approved alternative dosing regimens.

Multimodal Machine Learning for Prediction of 30-Day Readmission Risk in Elderly Population

BACKGROUND

Readmission within 30 days is a prevalent issue among elderly patients, linked to unfavorable health outcomes. Our objective was to develop and validate multimodal machine learning models for predicting 30-day readmission risk in elderly patients discharged from internal medicine departments.

METHODS

This was a retrospective cohort study which included elderly patients aged 75 or older, who were hospitalized at the Hadassah Medical Center internal medicine departments between 2014 and 2020. Three machine learning algorithms were developed and employed to predict 30-day readmission risk. The primary measures were predictive model performance scores, specifically area under the receiver operator curve (AUROC), and average precision.

RESULTS

This study included 19,569 admissions. Of them, 3258 (16.65%) resulted in 30-day readmission. Our 3 proposed models demonstrated high accuracy and precision on an unseen test set, with AUROC values of 0.87, 0.89, and 0.93, respectively, and average precision values of 0.76, 0.78, and 0.81. Feature importance analysis revealed that the number of admissions in the past year, history of 30-day readmission, Charlson score, and admission length were the most influential variables. Notably, the natural language processing score, representing the probability of readmission according to a textual-based model trained on social workers' assessment letters during hospitalization, ranked among the top 10 contributing factors.

CONCLUSIONS

Leveraging multimodal machine learning offers a promising strategy for identifying elderly patients who are at high risk for 30-day readmission. By identifying these patients, machine learning models may facilitate the effective execution of preventive actions to reduce avoidable readmission incidents.

How artificial intelligence could transform emergency care

Artificial intelligence (AI) in healthcare is the ability of a computer to perform tasks typically associated with clinical care (e.g. medical decision-making and documentation). AI will soon be integrated into an increasing number of healthcare applications, including elements of emergency department (ED) care. Here, we describe the basics of AI, various categories of its functions (including machine learning and natural language processing) and review emerging and potential future use-cases for emergency care. For example, AI-assisted symptom checkers could help direct patients to the appropriate setting, models could assist in assigning triage levels, and ambient AI systems could document clinical encounters. AI could also help provide focused summaries of charts, summarize encounters for hand-offs, and create discharge instructions with an appropriate language and reading level. Additional use cases include medical decision making for decision rules, real-time models that predict clinical deterioration or sepsis, and efficient extraction of unstructured data for coding, billing, research, and quality initiatives. We discuss the potential transformative benefits of AI, as well as the concerns regarding its use (e.g. privacy, data accuracy, and the potential for changing the doctor-patient relationship).

The application of explainable artificial intelligence (XAI) in electronic health record research: A scoping review

Machine Learning (ML) and Deep Learning (DL) models show potential in surpassing traditional methods including generalised linear models for healthcare predictions, particularly with large, complex datasets. However, low interpretability hinders practical implementation. To address this, Explainable Artificial Intelligence (XAI) methods are proposed, but a comprehensive evaluation of their effectiveness is currently limited. The aim of this scoping review is to critically appraise the application of XAI methods in ML/DL models using Electronic Health Record (EHR) data. In accordance with PRISMA scoping review guidelines, the study searched PUBMED and OVID/MEDLINE (including EMBASE) for publications related to tabular EHR data that employed ML/DL models with XAI. Out of 3220 identified publications, 76 were included. The selected publications published between February 2017 and June 2023, demonstrated an exponential increase over time. Extreme Gradient Boosting and Random Forest models were the most frequently used ML/DL methods, with 51 and 50 publications, respectively. Among XAI methods, Shapley Additive Explanations (SHAP) was predominant in 63 out of 76 publications, followed by partial dependence plots (PDPs) in 11 publications, and Locally Interpretable Model-Agnostic Explanations (LIME) in 8 publications. Despite the growing adoption of XAI methods, their applications varied widely and lacked critical evaluation. This review identifies the increasing use of XAI in tabular EHR research and highlights a deficiency in the reporting of methods and a lack of critical appraisal of validity and robustness. The study emphasises the need for further evaluation of XAI methods and underscores the importance of cautious implementation and interpretation in healthcare settings.

Biomarkers of aging in frailty and age-associated disorders: State of the art and future perspective

According to the Geroscience concept that organismal aging and age-associated diseases share the same basic molecular mechanisms, the identification of biomarkers of age that can efficiently classify people as biologically older (or younger) than their chronological (i.e. calendar) age is becoming of paramount importance. These people will be in fact at higher (or lower) risk for many different age-associated diseases, including cardiovascular diseases, neurodegeneration, cancer, etc. In turn, patients suffering from these diseases are biologically older than healthy age-matched individuals. Many biomarkers that correlate with age have been described so far. The aim of the present review is to discuss the usefulness of some of these biomarkers (especially soluble, circulating ones) in order to identify frail patients, possibly before the appearance of clinical symptoms, as well as patients at risk for age-associated diseases. An overview of selected biomarkers will be discussed in this regard, in particular we will focus on biomarkers related to metabolic stress response, inflammation, and cell death (in particular in neurodegeneration), all phenomena connected to inflammaging (chronic, low-grade, age-associated inflammation). In the second part of the review, next-generation markers such as extracellular vesicles and their cargos, epigenetic markers and gut microbiota composition, will be discussed. Since recent progresses in omics techniques have allowed an exponential increase in the production of laboratory data also in the field of biomarkers of age, making it difficult to extract biological meaning from the huge mass of available data, Artificial Intelligence (AI) approaches will be discussed as an increasingly important strategy for extracting knowledge from raw data and providing practitioners with actionable information to treat patients.

Predicting drug toxicity at the intersection of informatics and biology: DTox builds a foundation

Hao et al. (2022) present DTox (deep learning for toxicology), a neural network designed to predict and probe the sites and potential mechanisms underlying chemical toxicity; results provide a map to facilitate modular testing and improvements across multiple disparate applications.

The Effectiveness of a Hybrid Exercise Program on the Physical Fitness of Frail Elderly

BACKGROUND

Frailty is a serious physical disorder affecting the elderly all over the world. However, the frail elderly have low physical fitness, which limits the effectiveness of current exercise programs. Inspired by this, we attempted to integrate Baduanjin and strength and endurance exercises into an exercise program to improve the physical fitness and alleviate frailty among the elderly. Additionally, to achieve the goals of personalized medicine, machine learning simulations were performed to predict post-intervention frailty.

METHODS

A total of 171 frail elderly individuals completed the experiment, including a Baduanjin group (BDJ), a strength and endurance training group (SE), and a combination of Baduanjin and strength and endurance training group (BDJSE), which lasted for 24 weeks. Physical fitness was evaluated by 10-meter maximum walk speed (10 m MWS), grip strength, the timed up-and-go test (TUGT), and the 6 min walk test (6 min WT). A one-way analysis of variance (ANOVA), chi-square test, and two-way repeated-measures ANOVA were carried out to analyze the experimental data. In addition, nine machine learning models were utilized to predict the frailty status after the intervention.

RESULTS

In 10 m MWS and TUGT, there was a significant interactive influence between group and time. When comparing the BDJ group and the SE group, participants in the BDJSE group demonstrated the maximum gains in 10 m MWS and TUGT after 24 weeks of intervention. The stacking model surpassed other algorithms in performance. The accuracy and precision rates were 75.5% and 77.1%, respectively.

CONCLUSION

The hybrid exercise program that combined Baduanjin with strength and endurance training proved more effective at improving fitness and reversing frailty in elderly individuals. Based on the stacking model, it is possible to predict whether an elderly person will exhibit reversed frailty following an exercise program.

Human-Centered Explainable Artificial Intelligence: Automotive Occupational Health Protection Profiles in Prevention Musculoskeletal Symptoms

In automotive and industrial settings, occupational physicians are responsible for monitoring workers’ health protection profiles. Workers’ Functional Work Ability (FWA) status is used to create Occupational Health Protection Profiles (OHPP). This is a novel longitudinal study in comparison with previous research that has predominantly relied on the causality and explainability of human-understandable models for industrial technical teams like ergonomists. The application of artificial intelligence can support the decision-making to go from a worker’s Functional Work Ability to explanations by integrating explainability into medical (restriction) and support in contexts of individual, work-related, and organizational risk conditions. A sample of 7857 for the prognosis part of OHPP based on Functional Work Ability in the Portuguese language in the automotive industry was taken from 2019 to 2021. The most suitable regression models to predict the next medical appointment for the workers’ body parts protection were the models based on CatBoost regression, with an RMSLE of 0.84 and 1.23 weeks (mean error), respectively. CatBoost algorithm is also used to predict the next body part severity of OHPP. This information can help our understanding of potential risk factors for OHPP and identify warning signs of the early stages of musculoskeletal symptoms and work-related absenteeism.

Machine Learning Approaches for the Frailty Screening: A Narrative Review

Frailty characterizes a state of impairments that increases the risk of adverse health outcomes such as physical limitation, lower quality of life, and premature death. Frailty prevention, early screening, and management of potential existing conditions are essential and impact the elderly population positively and on society. Advanced machine learning (ML) processing methods are one of healthcare’s fastest developing scientific and technical areas. Although research studies are being conducted in a controlled environment, their translation into the real world (clinical setting, which is often dynamic) is challenging. This paper presents a narrative review of the procedures for the frailty screening applied to the innovative tools, focusing on indicators and ML approaches. It results in six selected studies. Support vector machine was the most often used ML method. These methods apparently can identify several risk factors to predict pre-frail or frailty. Even so, there are some limitations (e.g., quality data), but they have enormous potential to detect frailty early.

Navigating an early career in genomics and data science, written from the perspective of a current PhD student

Rebecca Tooze, a PhD student in the Oxford Clinical Genetics’ group, discusses the importance and application for data science in her field. Using bioinformatic approaches, she analyzes whole-genome sequencing data from patients with craniosynostosis. In this paper, she comments on her current research and her opportunity as an editorial intern with Patterns.

Frailty in Aging and the Search for the Optimal Biomarker: A Review

In the context of accelerated aging of the population worldwide, frailty has emerged as one of the main risk factors that can lead to loss of self-sufficiency in older people. This syndrome is defined as a reduced state of physiological reserve and functional capacity. The main diagnostic tools for frailty are based on scales that show deficits compared to their clinical application, such as the Fried frailty phenotype, among others. In this context, it is important to have one or more biomarkers with clinical applicability that can objectively and precisely determine the degree or risk of frailty in older people. The objective of this review was to analyze the biomarkers associated with frailty, classified according to the pathophysiological components of this syndrome (inflammation, coagulation, antioxidants, and liver function, among others). The evidence demonstrates that biomarkers associated with inflammation, oxidative stress, skeletal/cardiac muscle function, and platelet function represent the most promising markers of frailty due to their pathophysiological association with this syndrome. To a lesser extent but with the possibility of greater innovation, biomarkers associated with growth factors, vitamins, amino acids, and miRNAs represent alternatives as markers of this geriatric syndrome. Likewise, the incorporation of artificial intelligence represents an interesting approach to strengthening the diagnosis of frailty by biomarkers.