Clinical Data for OHSS Risk Model Development in Assisted Reproductive Technology

Xiangqian Meng published a dataset of clinical and laboratory data from 500 infertility patients undergoing controlled ovarian hyperstimulation at Jinxin Xinan Women & Children Hospital between 2020 and 2023. The data was used to develop and internally validate machine learning models for post-retrieval ovarian hyperstimulation syndrome risk stratification. The dataset includes 40 diagnosed OHSS cases and was used to train models achieving an AUC of 0.81.

Use Cases

• Developing predictive models for OHSS risk based on clinical and laboratory variables.
• Identifying key predictors like oocyte count and hormone levels for risk stratification.
• Training interpretable machine learning models using SHAP analysis for clinical decision support.
• Benchmarking model performance (e.g., Random Forest vs. LASSO) on a specific patient cohort.

Strengths

• Dataset contains 500 patient records with 40 confirmed OHSS cases.
• Models were evaluated using multiple metrics including AUC, accuracy, sensitivity, and specificity.
• Key predictors like number of oocytes retrieved and hormone levels were identified using SHAP.

Limitations

• Row and column counts are unknown, limiting suitability assessment.
• Data is from a single-center retrospective study, which may limit generalizability.
• The authors note findings are exploratory and require external validation.

Provenance

Source

Jinxin Xinan Women & Children Hospital

Collection Method

Retrospective collection of clinical and laboratory data from patients undergoing controlled ovarian hyperstimulation.

Time Range

2020 to 2023

Freshness

Last updated 2026-05-25 04:33:10.

Geography

Data likely originates from China, based on the hospital name.