Longitudinal multisource clinical model for early lung cancer risk stratification and screening

Objectives Lung cancer is the leading cause of cancer-related mortality worldwide, with poor prognosis largely due to late-stage diagnosis. Current screening methods such as low-dose CT face accessibility and cost barriers in resource-limited settings. This study develops a lightweight multichannel convolutional neural network for lung cancer screening support through longitudinal risk stratification using routine pre-diagnostic healthcare data. Methods We conducted a retrospective cohort study using Taiwan’s National Health Insurance Research Database, comprising 99 615 individuals (575 lung cancer cases; 99 040 non-cancer controls). Diagnostic codes, medication records and medical orders within a 36-month observation window were extracted. Log-likelihood ratio feature selection was implemented to reduce dimensionality, achieving 99.8% reduction in computational requirements while retaining clinical relevance. A multichannel Convolutional Neural Network (CNN) architecture was designed to process these heterogeneous data modalities simultaneously. Results The proposed method achieved an F₁-score of 0.5738, precision of 0.7149, Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.8316 and Area Under the Precision-Recall Curve (AUPRC) of 0.1617, outperforming baseline methods in precision and F₁-score. Ablation studies confirmed that medical orders provide primary predictive value, while medication features contribute limited discriminative signal in the pre-diagnostic phase. SHapley Additive exPlanations analysis revealed that routine healthcare utilisation patterns, rather than cancer-specific features, drive risk stratification. Discussion The lightweight architecture enables deployment in resource-constrained clinical environments while maintaining robust performance, offering potential as a preliminary screening tool to identify high-risk individuals for further diagnostic examination. Conclusion Efficient deep learning models using routine clinical data can facilitate lung cancer risk stratification and screening, providing a scalable solution for clinical implementation.

• Publication year

  2026

• Venue

  BMJ Health & Care Informatics

• Publication date

  2026-02-01

• Fields of study

  Medicine

• Identifiers
  DOI 10.1136/bmjhci-2025-101989PMID 41734977PMCID PMC12933799
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• The role of explainable artificial intelligence in disease prediction: a systematic literature review and future research directions

  2025cited by this paper
• Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries

  2024cited by this paper
• Machine Learning and Real-World Data to Predict Lung Cancer Risk in Routine Care

  2022cited by this paper
• Identification of robust deep neural network models of longitudinal clinical measurements

  2022cited by this paper
• Screening for Lung Cancer With Low-Dose Computed Tomography: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force.

  2021cited by this paper
• Does changing healthcare use signal opportunities for earlier detection of cancer? A review of studies using information from electronic patient records

  2021cited by this paper
• Lung cancer LDCT screening and mortality reduction — evidence, pitfalls and future perspectives

  2020cited by this paper
• ConvBERT: Improving BERT with Span-based Dynamic Convolution

  2020cited by this paper
• High-performance medicine: the convergence of human and artificial intelligence

  2019cited by this paper
• Taiwan’s National Health Insurance Research Database: past and future

  2019cited by this paper
• Temporal convolutional networks allow early prediction of events in critical care

  2019cited by this paper
• Opportunities and challenges in developing deep learning models using electronic health records data: a systematic review

  2018cited by this paper
• Data resource profile: the National Health Insurance Research Database (NHIRD)

  2018cited by this paper
• Dermatologist-level classification of skin cancer with deep neural networks

  2017cited by this paper
• A survey on deep learning in medical image analysis

  2017cited by this paper
• A Unified Approach to Interpreting Model Predictions

  2017cited by this paper
• Machine Learning and Prediction in Medicine - Beyond the Peak of Inflated Expectations.

  2017influential reference
• Impact of a False-Positive Screening Mammogram on Subsequent Screening Behavior and Stage at Breast Cancer Diagnosis

  2017cited by this paper
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction

  2017cited by this paper
• Risk factors for lung cancer worldwide

  2016cited by this paper
• Global cancer statistics, 2012

  2015cited by this paper
• Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening

  2012cited by this paper
• Technology

  1999cited by this paper
• A Comparative Study on Feature Selection in Text Categorization

  1997cited by this paper

• No citing papers are available for this paper.