Machine Learning Approaches to Predict Asthma Exacerbations: A Narrative Review

Advances in Therapy, Journal Year: 2023, Volume and Issue: 41(2), P. 534 - 552

Published: Dec. 19, 2023

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

Language: Английский

---

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

Frontiers in Public Health, Journal Year: 2025, Volume and Issue: 13

Published: Jan. 30, 2025

Background Public health policy evaluation is crucial for improving outcomes, optimizing healthcare resource allocation, and ensuring fairness transparency in decision-making. With the rise of big data, traditional methods face new challenges, requiring innovative approaches. Methods This article reviews principles, scope, limitations public explores application machine learning evaluating policies. It analyzes specific steps applying provides practical examples. The challenges discussed include model interpretability, data bias, continuation historical inequities, privacy concerns, while proposing ways to better apply context data. Results Machine techniques hold promise overcoming some methods, offering more precise evaluations However, such as lack perpetuation concerns remain significant. Discussion To address these suggests integrating data-driven theory-driven approaches improve developing multi-level strategies reduce bias mitigate through technical safeguards legal frameworks, employing validation benchmarking enhance robustness reproducibility.

Language: Английский

---

An Interpretable Machine Learning Model to Predict Hospitalizations

Clinical eHealth, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

---

Emerging applications of artificial intelligence in pathogen genomics

Frontiers in Bacteriology, Journal Year: 2024, Volume and Issue: 3

Published: March 6, 2024

The analysis of microbial genomes has long been recognised as a complex and data-rich domain where artificial intelligence (AI) can assist. As AI technologies have matured expanded, pathogen genomics also contended with exponentially larger datasets an expanding role in clinical public health practice. In this mini-review, we discuss examples emerging applications to address challenges for precision medicine health. These include models genotyping whole genome sequences, identifying novel pathogens metagenomic next generation sequencing, modelling genomic information using approaches from computational linguistics, phylodynamic estimation, large language make bioinformatics more accessible non-experts. We examine factors affecting the adoption into routine laboratory practice need renewed vision potential assist

Language: Английский

---

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

Journal of Neural Engineering, Journal Year: 2024, Volume and Issue: 21(3), P. 036054 - 036054

Published: June 1, 2024

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

Language: Английский

---

Assessing the impact of vaccines on COVID-19 efficacy in survival rates: a survival analysis approach for clinical decision support

Frontiers in Public Health, Journal Year: 2024, Volume and Issue: 12

Published: Nov. 18, 2024

Background The global COVID-19 pandemic, caused by the SARS-CoV-2 virus, has presented significant challenges to healthcare systems worldwide. Objective This study, based on an analysis of a cohort from Public Health System Andalusia (Spain), aims evaluate how vaccination affects case-fatality rate in patients hospitalized due infection Andalusia. Methods consists 37,274 individuals after applying inclusion criteria. We conducted survival analyses employing Cox proportional hazards models and generated adjusted curves examine outcomes. were performed three perspectives: vaccinated vs. unvaccinated patients, grouped age, stratified status. Results indicate substantial correlation between 20% reduction risk case-fatality. Age-specific effects reveal varying degrees protection across different age groups. Conclusion These findings emphasize pivotal role status assessment, supporting development clinical decision support system for accurate predictions optimizing management at admission.

Language: Английский

---

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

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Oct. 21, 2023

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

Language: Английский

---

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

International Journal of Public Health, Journal Year: 2024, Volume and Issue: 69

Published: Oct. 1, 2024

Language: Английский

---

Machine Learning Based in Quantum Mechanics and Theorem of Bayes

2021 International Conference on Electrical, Computer and Energy Technologies (ICECET), Journal Year: 2024, Volume and Issue: unknown, P. 1 - 6

Published: July 25, 2024

Language: Английский

---

Reply to: Microvascular hypertensive disease, long COVID, and end-organ pathology

Hypertension Research, Journal Year: 2023, Volume and Issue: 46(9), P. 2249 - 2250

Published: July 13, 2023

Language: Английский

---