Advancing Pediatric Growth Assessment with Machine Learning: Overcoming Challenges in Early Diagnosis and Monitoring DOI Creative Commons
Mauro Rodriguez-Marin, Luis Gustavo Orozco-Alatorre

Children, Год журнала: 2025, Номер 12(3), С. 317 - 317

Опубликована: Фев. 28, 2025

Background: Pediatric growth assessment is crucial for early diagnosis and intervention in disorders. Traditional methods often lack accuracy real-time decision-making capabilities This study explores the application of machine learning (ML), particularly logistic regression, to improve diagnostic precision timeliness pediatric assessment. Logistic regression a reliable easily interpretable model detecting abnormalities children. Unlike complex models, it offers parsimony transparency, efficiency, reproducibility, making ideal clinical settings where explainable, data-driven decisions are essential. Methods: A was developed using R analyze biometric demographic data from cross-sectional dataset, including real-world public institucions. The employed bibliometric analysis identify key trends incorporated preprocessing techniques such as cleaning, imputation, feature selection enhance performance. Performance metrics, accuracy, sensitivity, Receiver Operating Characteristic (ROC) curve, were utilized evaluation. Results: demonstrated an 94.65% sensitivity 91.03%, significantly improving identification anomalies compared conventional methods. model’s ROC curve showed area under (AUC) 0.96, indicating excellent predictive capability. Findings highlight ML’s potential automating monitoring supporting decision-making, can be very simple highly practice. Conclusions: ML, promising tool healthcare by enhancing operational efficiency. Despite these advancements, challenges remain regarding quality, integration, privacy concerns. Future research should focus on expanding dataset diversity, interpretability, conducting external validation facilitate broader adoption.

Язык: Английский

Advancing Pediatric Growth Assessment with Machine Learning: Overcoming Challenges in Early Diagnosis and Monitoring DOI Creative Commons
Mauro Rodriguez-Marin, Luis Gustavo Orozco-Alatorre

Children, Год журнала: 2025, Номер 12(3), С. 317 - 317

Опубликована: Фев. 28, 2025

Background: Pediatric growth assessment is crucial for early diagnosis and intervention in disorders. Traditional methods often lack accuracy real-time decision-making capabilities This study explores the application of machine learning (ML), particularly logistic regression, to improve diagnostic precision timeliness pediatric assessment. Logistic regression a reliable easily interpretable model detecting abnormalities children. Unlike complex models, it offers parsimony transparency, efficiency, reproducibility, making ideal clinical settings where explainable, data-driven decisions are essential. Methods: A was developed using R analyze biometric demographic data from cross-sectional dataset, including real-world public institucions. The employed bibliometric analysis identify key trends incorporated preprocessing techniques such as cleaning, imputation, feature selection enhance performance. Performance metrics, accuracy, sensitivity, Receiver Operating Characteristic (ROC) curve, were utilized evaluation. Results: demonstrated an 94.65% sensitivity 91.03%, significantly improving identification anomalies compared conventional methods. model’s ROC curve showed area under (AUC) 0.96, indicating excellent predictive capability. Findings highlight ML’s potential automating monitoring supporting decision-making, can be very simple highly practice. Conclusions: ML, promising tool healthcare by enhancing operational efficiency. Despite these advancements, challenges remain regarding quality, integration, privacy concerns. Future research should focus on expanding dataset diversity, interpretability, conducting external validation facilitate broader adoption.

Язык: Английский

Процитировано

0