Related Experiment Video
Updated: Apr 28, 2026
Murine Model of Leukemia Relapse to Induction Chemotherapy for Acute Lymphoblastic Leukemia
Published on: October 17, 2025
Development of an interpretable machine learning model to predict complete remission and first adverse event in
Ningshu Huang1, Weiwei Wang2, Rui Yang2
1Department of Clinical Laboratory, Children's Hospital of Chongqing Medical University, National Clinical Research Center for Children and Adolescents' Health and Diseases, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Pediatrics Metabolism and Inflammatory Diseases, Laboratory for clinical diagnostic and translational research of CHCQMU, Chongqing, China.
Insights
Machine learning models accurately predict complete remission and adverse events in pediatric acute myeloid leukemia (pAML) using routine clinical data. This aids in personalized treatment strategies for pAML patients.
Area of Science:
- Hematology
- Oncology
- Machine Learning
Background:
- Pediatric acute myeloid leukemia (pAML) is a severe hematologic malignancy.
- Predicting complete remission (CR) and adverse events (AE) is crucial for tailoring pAML treatment.
- Interpretable machine learning (ML) models using routine clinical features are needed.
Purpose of the Study:
- Develop and evaluate ML models for predicting CR and AE in pediatric AML patients.
- Identify key clinical features that predict CR and AE.
- Assess model interpretability for clinical application.
Main Methods:
- Utilized data from 206 pediatric AML patients, excluding acute promyelocytic leukemia.
- Developed seven supervised ML algorithms for CR and AE prediction.
- Employed feature selection methods (random forest, stepwise regression, JMIM) and interpretability techniques (feature importance, ALE, SHAP).
Main Results:
- Random forest models achieved high performance for both CR (test AUC 0.79) and AE prediction (test AUC 0.78).
- Key predictors for CR included age, blasts, platelet count, and genetic markers.
- Key predictors for AE included WBC, blasts, platelet count, hemoglobin, age, and genetic markers.
- Higher platelet count predicted increased CR probability and reduced AE risk; elevated WBC and blasts predicted higher AE incidence.
Conclusions:
- A random forest model using routine hematological parameters shows significant potential for predicting CR and AE in pAML.
- The model facilitates early risk stratification and personalized treatment approaches for pediatric AML.
- Interpretable ML models can effectively leverage routine clinical data for improved pAML management.
Background:
Pediatric acute myeloid leukemia (pAML) is a rapidly progressive myeloid malignancy characterized by malignant clonal expansion of hematopoietic stem and progenitor cells. The prediction of complete remission (CR) and first adverse event (AE) is critical for personalizing pAML treatment; however, interpretable machine learning (ML) models that utilize only routine clinical features for this purpose are lacking.
Methods:
A total of 206 de novo pediatric AML patients (excluding acute promyelocytic leukemia) were randomly split into training (80%) and test (20%) sets. Seven supervised ML algorithms were constructed for predicting CR and AE, and their performance was evaluated by accuracy, specificity, F1-score, and area under the receiver operating characteristic curve (AUC). Model interpretability was assessed using feature importance, accumulated local effect (ALE) plots, and SHAP values.
Results:
To identify optimal predictors of CR, three feature selection methods-random forest, stepwise regression, and joint mutual information maximization (JMIM)-were employed. Their intersection revealed seven key features: age, bone marrow blasts, peripheral blood blasts, platelet count (PLT), t (8;21), TP53 and del7/del7q. The random forest model demonstrated optimal performance, with a training AUC of 0.90 (95% CI: 0.86-0.97) and a test AUC of 0.79 (95% CI: 0.72-0.86). The similar machine learning pipeline was applied to predict the first adverse event (AE). Nine features were selected as optimal predictors through the intersection of the same three algorithms: white blood cell count (WBC), peripheral blood blasts, PLT, bone marrow blasts, hemoglobin, age, t (8;21), NPM1 and KIT. For AE prediction, the random forest algorithm also exhibited optimal performance, with a training AUC of 0.92 (95% CI: 0.85-0.97) and a test AUC of 0.78 (95% CI: 0.66-0.84). Interpretability analysis of the random forest models revealed that a higher platelet count at diagnosis was predictive of an increased probability of CR and a reduced risk of AE. In contrast, elevated WBC and peripheral blood blast percentage were associated with a higher incidence of AE.
Conclusion:
Our random forest model, built on routine hematological parameters, demonstrated strong potential for predicting CR and AE in pAML, thereby facilitating early risk stratification and guiding personalized treatment strategies.
More Related Videos
09:57Comprehensive Protocol to Sample and Process Bone Marrow for Measuring Measurable Residual Disease and Leukemic Stem Cells in Acute Myeloid Leukemia
Published on: March 5, 2018
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018