Predicting 5-Year Mortality in Non–Small-Cell Lung Cancer Using the Korean Central Cancer Registry: Model Development and Validation Study - Scorecard - MDSpire
Advertisement
Predicting 5-Year Mortality in Non–Small-Cell Lung Cancer Using the Korean Central Cancer Registry: Model Development and Validation Study
Clinical Scorecard: Developing and Validating a Model to Forecast 5-Year Survival in Non–Small-Cell Lung Cancer Utilizing Data from the Korean Central Cancer Registry
At a Glance
Category
Detail
Condition
Non-small-cell lung cancer (NSCLC)
Key Mechanisms
Influenced by stage at diagnosis and molecular biomarkers (EGFR, ALK)
Target Population
Patients diagnosed with NSCLC in South Korea (2014-2017)
Care Setting
Multicenter cohort from the Korea Central Cancer Registry
Key Highlights
Developed a deep learning model for predicting 5-year mortality in NSCLC
Utilized data from over 50 medical centers in South Korea
Incorporated demographic and clinical variables for robust analysis
Emphasized interpretability and reproducibility in model design
Addressed challenges in data preprocessing and hyperparameter tuning
Guideline-Based Recommendations
Diagnosis
Utilize comprehensive demographic and clinical data for accurate staging
Management
Tailor treatment strategies based on individual patient prognostic predictions
Monitoring & Follow-up
Regularly assess clinical features and biomarkers for ongoing patient evaluation
Risks
Consider potential biases introduced by data exclusion criteria
Patient & Prescribing Data
Patients with NSCLC from the Korea Central Cancer Registry
Focus on integrating clinical features and biomarkers for personalized treatment
Clinical Best Practices
Employ a complete-case approach for core variables to minimize bias
Use stratified data splitting for model training and validation
Ensure consistent preprocessing across clinical feature domains
