Identification of features associated with dysmenorrhea in women aged 20–50 years using logistic regression and synthetic minority oversampling technique
5 affiliations
- 1Department of Obstetrics and Gynecology, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan
- 2Graduate Institute of Public Health, College of Public Health, National Defense Medical University, Taipei, Taiwan
- 3MJ Health Research Foundation, Taipei, Taiwan
- 4Division of Nephrology, Department of Internal Medicine, Shin Kong Wu Ho-Su Memorial Hospital, No. 95, Wenchang Rd., Shilin Dist., Taipei, 111045 Taiwan
- 5School of Medicine, Fu Jen Catholic University, New Taipei City, Taiwan
Abstract
Background: Dysmenorrhea (Dys) is a prevalent gynecological condition that significantly impairs women’s quality of life. This study investigates the key predictive features associated with Dys in Taiwanese women aged 20–50, using logistic regression (LR) with and without the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance.
Methods: Data were obtained from the MJ Health Database, involving 1,038 women. Dys was defined via self-report. A total of 41 clinical, biochemical, and lifestyle features were analyzed. Performance metrics—including accuracy, sensitivity, specificity, precision, F1 score, and area under the ROC and PR curves—were computed via 5-fold cross-validation. Logistic regression coefficients and odds ratios were calculated to evaluate feature associations.
Results: Without SMOTE, the LR model showed high specificity (97.9%) and accuracy (93.2%) but low sensitivity (51.9%). With SMOTE, sensitivity improved significantly (83.2%), although precision and specificity declined, demonstrating a trade-off between detecting positive cases and generating false positives. Both models maintained high AUC scores (ROC AUC ~ 0.92; PR AUC ~ 0.68). Variable importance analysis, based on standardized coefficients, highlighted six key predictive features: childbirth history, age, glutamic pyruvic transaminase (GPT), glutamic oxaloacetic transaminase (GOT), menstrual flow volume, and diastolic blood pressure (DBP).
Conclusion: SMOTE-enhanced logistic regression offers a highly sensitive model for detecting Dys in imbalanced datasets. The identified predictive features could inform targeted screening interventions in primary care settings, though external validation is required.
Related
- Serum androstenedione levels are associated with dysmenorrhea and pain severity among women without endometriosis: an exploratory cross-sectional study.
- Identification of Risk-Related Prenatal Medication Exposure Profiles for Childhood Asthma Using the Bayesian Profile Regression: A Cohort Study Using the POMME Cohort.
- A machine learning framework for stage-structured predator–prey systems with fear effect: Sparse Identification, Gaussian Process Regression, and bifurcation analysis
- Hysteroscopic polypectomy with vs. without levonorgestrel-releasing intrauterine system in women with endometrial polyps and dysmenorrhea
- Fire dragon cupping therapy for primary dysmenorrhea: a systematic review of available clinical evidence and research gaps
BibTeX
@article{su2026identification,
title = {{Identification of features associated with dysmenorrhea in women aged 20–50 years using logistic regression and synthetic minority oversampling technique}},
author = {Jian-Lun Su and Yu-Huan Weng and Chi-Kang Lin and Ta-Wei Chu and Yu-Chi Wang and Jyh-Gang Leu},
journal = {BMC Women's Health},
year = {2026},
month = aug,
doi = {10.1186/s12905-026-04751-1},
url = {https://doi.org/10.1186/s12905-026-04751-1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Identification of features associated with dysmenorrhea in women aged 20–50 years using logistic regression and synthetic minority oversampling technique
AU - Jian-Lun Su
AU - Yu-Huan Weng
AU - Chi-Kang Lin
AU - Ta-Wei Chu
AU - Yu-Chi Wang
AU - Jyh-Gang Leu
PY - 2026
DA - 2026/08/08
JO - BMC Women's Health
DO - 10.1186/s12905-026-04751-1
UR - https://doi.org/10.1186/s12905-026-04751-1
AB - Background: Dysmenorrhea (Dys) is a prevalent gynecological condition that significantly impairs women’s quality of life. This study investigates the key predictive features associated with Dys in Taiwanese women aged 20–50, using logistic regression (LR) with and without the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance. Methods: Data were obtained from the MJ Health Database, involving 1,038 women. Dys was defined via self-report. A total of 41 clinical, biochemical, and lifestyle features were analyzed. Performance metrics—including accuracy, sensitivity, specificity, precision, F1 score, and area under the ROC and PR curves—were computed via 5-fold cross-validation. Logistic regression coefficients and odds ratios were calculated to evaluate feature associations. Results: Without SMOTE, the LR model showed high specificity (97.9%) and accuracy (93.2%) but low sensitivity (51.9%). With SMOTE, sensitivity improved significantly (83.2%), although precision and specificity declined, demonstrating a trade-off between detecting positive cases and generating false positives. Both models maintained high AUC scores (ROC AUC ~ 0.92; PR AUC ~ 0.68). Variable importance analysis, based on standardized coefficients, highlighted six key predictive features: childbirth history, age, glutamic pyruvic transaminase (GPT), glutamic oxaloacetic transaminase (GOT), menstrual flow volume, and diastolic blood pressure (DBP). Conclusion: SMOTE-enhanced logistic regression offers a highly sensitive model for detecting Dys in imbalanced datasets. The identified predictive features could inform targeted screening interventions in primary care settings, though external validation is required.
SN - 1472-6874
ER -