Hemorrhage Risk Nomogram for Acute Leukemia Patients, 2013-2023

468 acute leukemia patient records from two Chinese hospitals between January 2013 and December 2023 were used to develop a predictive nomogram. The model, created by Xinxin Hu, uses five clinical variables to predict World Health Organization grade 2 or higher hemorrhagic events related to anti-tumor therapy. The nomogram achieved an area under the ROC curve of 0.741 in the training group and 0.718 in the test group.

Use Cases

• Predicting hemorrhage risk in acute leukemia patients based on infection status and platelet count.
• Evaluating clinical utility of a predictive model using decision curve analysis based on the described methodology.
• Training logistic regression models for medical risk assessment using the identified predictors like hematocrit and transfusion history.

Strengths

• Model performance is quantitatively reported with area under the ROC curve values of 0.741 and 0.718.
• Dataset includes 468 patient records from a 10-year time range (2013-2023).
• Study design includes a 7:3 random split for training and validation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect geographic bias inherent to the two specific hospitals in Shanghai and Nanyang.
• Row count for the underlying clinical data is unknown, which may limit suitability assessment.

Provenance

Source

The Shanghai Fifth People’s Hospital and Nanyang Municipal Central Hospital.

Collection Method

Retrospective cohort study.

Time Range

January 2013 to December 2023.

Freshness

Last updated 2026-03-18 07:34:17; freshness should be verified.

Geography

China (Shanghai and Nanyang).