Statistical models versus machine learning for competing risks: development and validation of prognostic models

Comparison of machine learning and conventional statistical modeling for predicting readmission following acute heart failure hospitalization

INTRODUCTION

Developing accurate models for predicting the risk of 30-day readmission is a major healthcare interest. Evidence suggests that models developed using machine learning (ML) may have better discrimination than conventional statistical models (CSM), but the calibration of such models is unclear.

OBJECTIVES

To compare models developed using ML with those developed using CSM to predict 30-day readmission for cardiovascular and noncardiovascular causes in HF patients.

METHODS

We retrospectively enrolled 10,919 patients with HF (> 18 years) discharged alive from a hospital or emergency department (2004-2007) in Ontario, Canada. The study sample was randomly divided into training and validation sets in a 2:1 ratio. CSMs to predict 30-day readmission were developed using Fine-Gray subdistribution hazards regression (treating death as a competing risk), and the ML algorithm employed random survival forests for competing risks (RSF-CR). Models were evaluated in the validation set using both discrimination and calibration metrics.

RESULTS

In the validation sample of 3602 patients, RSF-CR (c-statistic=0.620) showed similar discrimination to the Fine-Gray competing risk model (c-statistic=0.621) for 30-day cardiovascular readmission. In contrast, for 30-day noncardiovascular readmission, the Fine-Gray model (c-statistic=0.641) slightly outperformed the RSF-CR model (c-statistic=0.632). For both outcomes, The Fine-Gray model displayed better calibration than RSF-CR using calibration plots of observed vs predicted risks across the deciles of predicted risk.

CONCLUSIONS

Fine-Gray models had similar discrimination but superior calibration to the RSF-CR model, highlighting the importance of reporting calibration metrics for ML-based prediction models. The discrimination was modest in all readmission prediction models regardless of the methods used.

Risk factor-targeted abdominal aortic aneurysm screening: systematic review of risk prediction for abdominal aortic aneurysm

BACKGROUND

This systematic review aimed to investigate the current state of risk prediction for abdominal aortic aneurysm in the literature, identifying and comparing published models and describing their performance and applicability to a population-based targeted screening strategy.

METHODS

Electronic databases MEDLINE (via Ovid), Embase (via Ovid), MedRxiv, Web of Science, and the Cochrane Library were searched for papers reporting or validating risk prediction models for abdominal aortic aneurysm. Studies were included only if they were developed on a cohort or study group derived from the general population and used multiple variables with at least one modifiable risk factor. Risk of bias was assessed using the Prediction model Risk Of Bias ASsessment Tool. A synthesis and comparison of the identified models was undertaken.

RESULTS

The search identified 4813 articles. After full-text review, 37 prediction models were identified, of which 4 were unique predictive models that were reported in full. Applicability was poor when considering targeted screening strategies using electronic health record-based populations. Common risk factors used for the predictive models were explored across all 37 models; the most common risk factors in predictive models for abdominal aortic aneurysm were: age, sex, biometrics (such as height, weight, or BMI), smoking, hypertension, hypercholesterolaemia, and history of heart disease. Few models had undergone standardized model development, adequate external validation, or impact evaluation.

CONCLUSION

This study identified four risk models that can be replicated and used to predict abdominal aortic aneurysm with acceptable levels of discrimination. None of the models have been validated externally.

Clinical Deployment of Machine Learning Tools in Transplant Medicine: What Does the Future Hold?

Medical applications of machine learning (ML) have shown promise in analyzing patient data to support clinical decision-making and provide patient-specific outcomes. In transplantation, several applications of ML exist which include pretransplant: patient prioritization, donor-recipient matching, organ allocation, and posttransplant outcomes. Numerous studies have shown the development and utility of ML models, which have the potential to augment transplant medicine. Despite increasing efforts to develop robust ML models for clinical use, very few of these tools are deployed in the healthcare setting. Here, we summarize the current applications of ML in transplant and discuss a potential clinical deployment framework using examples in organ transplantation. We identified that creating an interdisciplinary team, curating a reliable dataset, addressing the barriers to implementation, and understanding current clinical evaluation models could help in deploying ML models into the transplant clinic setting.

Integrating Omics Data and AI for Cancer Diagnosis and Prognosis

Cancer is one of the leading causes of death, making timely diagnosis and prognosis very important. Utilization of AI (artificial intelligence) enables providers to organize and process patient data in a way that can lead to better overall outcomes. This review paper aims to look at the varying uses of AI for diagnosis and prognosis and clinical utility. PubMed and EBSCO databases were utilized for finding publications from 1 January 2020 to 22 December 2023. Articles were collected using key search terms such as "artificial intelligence" and "machine learning." Included in the collection were studies of the application of AI in determining cancer diagnosis and prognosis using multi-omics data, radiomics, pathomics, and clinical and laboratory data. The resulting 89 studies were categorized into eight sections based on the type of data utilized and then further subdivided into two subsections focusing on cancer diagnosis and prognosis, respectively. Eight studies integrated more than one form of omics, namely genomics, transcriptomics, epigenomics, and proteomics. Incorporating AI into cancer diagnosis and prognosis alongside omics and clinical data represents a significant advancement. Given the considerable potential of AI in this domain, ongoing prospective studies are essential to enhance algorithm interpretability and to ensure safe clinical integration.

Machine Learning-Assisted Decision Making in Orthopaedic Oncology

» Artificial intelligence is an umbrella term for computational calculations that are designed to mimic human intelligence and problem-solving capabilities, although in the future, this may become an incomplete definition. Machine learning (ML) encompasses the development of algorithms or predictive models that generate outputs without explicit instructions, assisting in clinical predictions based on large data sets. Deep learning is a subset of ML that utilizes layers of networks that use various inter-relational connections to define and generalize data.» ML algorithms can enhance radiomics techniques for improved image evaluation and diagnosis. While ML shows promise with the advent of radiomics, there are still obstacles to overcome.» Several calculators leveraging ML algorithms have been developed to predict survival in primary sarcomas and metastatic bone disease utilizing patient-specific data. While these models often report exceptionally accurate performance, it is crucial to evaluate their robustness using standardized guidelines.» While increased computing power suggests continuous improvement of ML algorithms, these advancements must be balanced against challenges such as diversifying data, addressing ethical concerns, and enhancing model interpretability.

A Critical Review of Data Science Applications in Resource Recovery and Carbon Capture from Organic Waste

Municipal and agricultural organic waste can be treated to recover energy, nutrients, and carbon through resource recovery and carbon capture (RRCC) technologies such as anaerobic digestion, struvite precipitation, and pyrolysis. Data science could benefit such technologies by improving their efficiency through data-driven process modeling along with reducing environmental and economic burdens via life cycle assessment (LCA) and techno-economic analysis (TEA), respectively. We critically reviewed 616 peer-reviewed articles on the use of data science in RRCC published during 2002-2022. Although applications of machine learning (ML) methods have drastically increased over time for modeling RRCC technologies, the reviewed studies exhibited significant knowledge gaps at various model development stages. In terms of sustainability, an increasing number of studies included LCA with TEA to quantify both environmental and economic impacts of RRCC. Integration of ML methods with LCA and TEA has the potential to cost-effectively investigate the trade-off between efficiency and sustainability of RRCC, although the literature lacked such integration of techniques. Therefore, we propose an integrated data science framework to inform efficient and sustainable RRCC from organic waste based on the review. Overall, the findings from this review can inform practitioners about the effective utilization of various data science methods for real-world implementation of RRCC technologies.