The role of machine learning in health policies during the COVID-19 pandemic and in long COVID management

Machine learning and public health policy evaluation: research dynamics and prospects for challenges

Background

Public health policy evaluation is crucial for improving health outcomes, optimizing healthcare resource allocation, and ensuring fairness and transparency in decision-making. With the rise of big data, traditional evaluation methods face new challenges, requiring innovative approaches.

Methods

This article reviews the principles, scope, and limitations of traditional public health policy evaluation methods and explores the application of machine learning in evaluating public health policies. It analyzes the specific steps for applying machine learning and provides practical examples. The challenges discussed include model interpretability, data bias, the continuation of historical health inequities, and data privacy concerns, while proposing ways to better apply machine learning in the context of big data.

Results

Machine learning techniques hold promise in overcoming some limitations of traditional methods, offering more precise evaluations of public health policies. However, challenges such as lack of model interpretability, the perpetuation of health inequities, data bias, and privacy concerns remain significant.

Discussion

To address these challenges, the article suggests integrating data-driven and theory-driven approaches to improve model interpretability, developing multi-level data strategies to reduce bias and mitigate health inequities, ensuring data privacy through technical safeguards and legal frameworks, and employing validation and benchmarking strategies to enhance model robustness and reproducibility.

Precision Epidemiology: A Computational Analysis of the Impact of Algorithmic Prediction on the Relationship Between Population Epidemiology and Clinical Epidemiology

Objectives

Precision Medicine (PM) uses advanced Machine Learning (ML) techniques and big data to develop personalized treatments, but healthcare still relies on traditional statistical procedures not targeted on individuals. This study investigates the impact of ML on epidemiology.

Methods

A quantitative analysis of the articles in PubMed for the years 2000-2019 was conducted to investigate the use of statistical methods and ML in epidemiology. Using structural topic modelling, two groups of topics were identified and analysed over time: topics closer to the clinical side of epidemiology and topics closer to the population side.

Results

The curve of the prevalence of topics associated with population epidemiology basically corresponds to the curve of the relative statistical methods, while the more dynamic curve of clinical epidemiology broadly reproduces the trend of algorithmic methods.

Conclusion

The findings suggest that a renewed separation between clinical epidemiology and population epidemiology is emerging, with clinical epidemiology taking more advantage of recent developments in algorithmic techniques and moving closer to bioinformatics, whereas population epidemiology seems to be slower in this innovation.

NeurostimML: a machine learning model for predicting neurostimulation-induced tissue damage

Objective. The safe delivery of electrical current to neural tissue depends on many factors, yet previous methods for predicting tissue damage rely on only a few stimulation parameters. Here, we report the development of a machine learning approach that could lead to a more reliable method for predicting electrical stimulation-induced tissue damage by incorporating additional stimulation parameters.Approach. A literature search was conducted to build an initial database of tissue response information after electrical stimulation, categorized as either damaging or non-damaging. Subsequently, we used ordinal encoding and random forest for feature selection, and investigated four machine learning models for classification: Logistic Regression, K-nearest Neighbor, Random Forest, and Multilayer Perceptron. Finally, we compared the results of these models against the accuracy of the Shannon equation.Main Results. We compiled a database with 387 unique stimulation parameter combinations collected from 58 independent studies conducted over a period of 47 years, with 195 (51%) categorized as non-damaging and 190 (49%) categorized as damaging. The features selected for building our model with a Random Forest algorithm were: waveform shape, geometric surface area, pulse width, frequency, pulse amplitude, charge per phase, charge density, current density, duty cycle, daily stimulation duration, daily number of pulses delivered, and daily accumulated charge. The Shannon equation yielded an accuracy of 63.9% using akvalue of 1.79. In contrast, the Random Forest algorithm was able to robustly predict whether a set of stimulation parameters was classified as damaging or non-damaging with an accuracy of 88.3%.Significance. This novel Random Forest model can facilitate more informed decision making in the selection of neuromodulation parameters for both research studies and clinical practice. This study represents the first approach to use machine learning in the prediction of stimulation-induced neural tissue damage, and lays the groundwork for neurostimulation driven by machine learning models.

Machine Learning Approaches to Predict Asthma Exacerbations: A Narrative Review

The implementation of artificial intelligence (AI) and machine learning (ML) techniques in healthcare has garnered significant attention in recent years, especially as a result of their potential to revolutionize personalized medicine. Despite advances in the treatment and management of asthma, a significant proportion of patients continue to suffer acute exacerbations, irrespective of disease severity and therapeutic regimen. The situation is further complicated by the constellation of factors that influence disease activity in a patient with asthma, such as medical history, biomarker phenotype, pulmonary function, level of healthcare access, treatment compliance, comorbidities, personal habits, and environmental conditions. A growing body of work has demonstrated the potential for AI and ML to accurately predict asthma exacerbations while also capturing the entirety of the patient experience. However, application in the clinical setting remains mostly unexplored, and important questions on the strengths and limitations of this technology remain. This review presents an overview of the rapidly evolving landscape of AI and ML integration into asthma management by providing a snapshot of the existing scientific evidence and proposing potential avenues for future applications.

NeurostimML: A machine learning model for predicting neurostimulation-induced tissue damage

Objective

The safe delivery of electrical current to neural tissue depends on many factors, yet previous methods for predicting tissue damage rely on only a few stimulation parameters. Here, we report the development of a machine learning approach that could lead to a more reliable method for predicting electrical stimulation-induced tissue damage by incorporating additional stimulation parameters.

Approach

A literature search was conducted to build an initial database of tissue response information after electrical stimulation, categorized as either damaging or non-damaging. Subsequently, we used ordinal encoding and random forest for feature selection, and investigated four machine learning models for classification: Logistic Regression, K-nearest Neighbor, Random Forest, and Multilayer Perceptron. Finally, we compared the results of these models against the accuracy of the Shannon equation.

Main Results

We compiled a database with 387 unique stimulation parameter combinations collected from 58 independent studies conducted over a period of 47 years, with 195 (51%) categorized as non-damaging and 190 (49%) categorized as damaging. The features selected for building our model with a Random Forest algorithm were: waveform shape, geometric surface area, pulse width, frequency, pulse amplitude, charge per phase, charge density, current density, duty cycle, daily stimulation duration, daily number of pulses delivered, and daily accumulated charge. The Shannon equation yielded an accuracy of 63.9% using a k value of 1.79. In contrast, the Random Forest algorithm was able to robustly predict whether a set of stimulation parameters was classified as damaging or non-damaging with an accuracy of 88.3%.

Significance

This novel Random Forest model can facilitate more informed decision making in the selection of neuromodulation parameters for both research studies and clinical practice. This study represents the first approach to use machine learning in the prediction of stimulation-induced neural tissue damage, and lays the groundwork for neurostimulation driven by machine learning models.