Skip to content
DraconiticResearch
Search Sources API
181,716 studies31 sourcesUpdated 20 min ago
Peer-reviewedCC BYDiagnostics (Basel, Switzerland)

A Clinician-Configurable FAHP-Based Ensemble Framework for Multi-Label Chest X-Ray Classification.

  • Hui-Chu Chiu1
  • Yu-Hsiang Tsai2
  • Cheng-Hsuan Juan3,4
  • Chia-Ching Chang2,5
  • Ya-Hui Li2,4
  • Deng-Yiu Chiu6
  • Chun-Jung Juan2,3,7,8,9
9 affiliations
  1. 1Ph.D. Program of Management, Chung-Hua University, Hsinchu 300, Taiwan.
  2. 2Department of Medical Imaging, China Medical University Hsinchu Hospital, Hsinchu 302, Taiwan.
  3. 3Department of Medical Imaging, China Medical University Hospital, Taichung 404, Taiwan.
  4. 4Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei 106, Taiwan.
  5. 5Department of Management Science, National Yang Ming Chiao Tung University, Hsinchu 300, Taiwan.
  6. 6Department of Information Management, Chung-Hua University, Hsinchu 300, Taiwan.
  7. 7Department of Radiology, School of Medicine, College of Medicine, China Medical University, Taichung 406, Taiwan.
  8. 8Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu 300, Taiwan.
  9. 9Department of Computer Science and Information Engineering, National Taiwan University, Taipei 106, Taiwan.
Read full textPDFDOICite

Abstract

Background/Objectives: Deep learning-based chest X-ray (CXR) analysis has shown strong diagnostic performance; however, conventional ensemble methods typically rely on equal-weight aggregation and do not account for disease-specific model behavior or clinician-defined priorities. Methods: We developed a clinician-configurable ensemble framework based on the fuzzy analytic hierarchy process (FAHP) for multi-label CXR classification. The proposed method employs a two-stage design consisting of disease-specific model weighting using multi-criteria decision analysis and utility-guided seed aggregation based on ΔNet Benefit (ΔNB). Five base architectures, including Vision Transformer (ViT), SwinV2 Transformer, DenseNet121, DenseNet121 + ViT and DenseNet121 + SwinV2, were trained on the National Institutes of Health (NIH) ChestX-ray14 dataset using 10 random seeds. Performance was evaluated on a held-out test set using area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), balanced accuracy, and F1 score and compared with single-model baselines, Hard Voting, and Soft Voting. Results: In the seed-level analyses, Soft Voting achieved the highest overall discrimination performance (AUROC, 0.851 ± 0.001; AUPRC, 0.293 ± 0.003), whereas seed-level FAHP showed comparable performance (AUROC, 0.850 ± 0.001; AUPRC, 0.287 ± 0.004). In the final-level analyses, the sensitivity-priority FAHP-ΔNB configuration achieved a macro-averaged AUROC of 0.855 and an AUPRC of 0.299. Hard Voting demonstrated inferior performance, particularly for AUPRC. Per-disease analysis revealed heterogeneous results, with FAHP showing favorable performance in selected conditions, including Effusion, Emphysema, Mass, Nodule, and Pleural Thickening, while Soft Voting remained superior in others. Statistical analysis confirmed that differences between FAHP and Soft Voting were generally modest and disease-dependent. Conclusions: The proposed FAHP-based ensemble maintains performance comparable to strong equal-weight baselines while enabling clinician-configurable, disease-specific model weighting and utility-guided aggregation. This framework provides an aggregation-level transparent approach for incorporating configurable diagnostic priorities into multi-label CXR classification.

Related

  1. Multi-task deep learning framework for brain tumor classification and segmentation using ensemble feature extraction
    Retinderdeep Singh, Chander Prabha, Shahab Abdulla
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textNeurologyCrossref
  2. A hybrid deep feature extraction and ensemble learning framework for robust multi-class ECG image classification
    Sonam Nagar, Harish Sharma, Karan Verma +2
    Scientific ReportsPeer-reviewedCC BY 4.0Full textCardiovascular medicineCrossref
  3. Multi-Platform Transcriptomic Classification of LUAD Driver Mutation Subtypes via Heterogeneous Ensemble Learning
    Diagnostics (Basel, Switzerland)Peer-reviewedCC BYFull textGeneral medicineEurope PMC
  4. Gallstone Status Classification Using Adaptive Performance-Diversity Hybrid Ensemble with Pareto-Based Multi-Objective Selection (APDHE-Pareto)
    BiomedicinesPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  5. Temporal deep learning framework for multi condition fault classification of three-phase induction motors
    R. Sooraj, Senthil Kumar Ramu, R. Sitharthan +2
    Scientific ReportsPeer-reviewedCC BYFull textGeneral medicineDOAJ

Cite

BibTeXRIS
BibTeX
@article{chiu2026clinicianconfigurable,
  title = {{A Clinician-Configurable FAHP-Based Ensemble Framework for Multi-Label Chest X-Ray Classification.}},
  author = {Hui-Chu Chiu and Yu-Hsiang Tsai and Cheng-Hsuan Juan and Chia-Ching Chang and Ya-Hui Li and Deng-Yiu Chiu and Chun-Jung Juan},
  journal = {Diagnostics (Basel, Switzerland)},
  year = {2026},
  month = sep,
  doi = {10.3390/diagnostics16193132},
  url = {https://europepmc.org/article/MED/42882287},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - A Clinician-Configurable FAHP-Based Ensemble Framework for Multi-Label Chest X-Ray Classification.
AU  - Hui-Chu Chiu
AU  - Yu-Hsiang Tsai
AU  - Cheng-Hsuan Juan
AU  - Chia-Ching Chang
AU  - Ya-Hui Li
AU  - Deng-Yiu Chiu
AU  - Chun-Jung Juan
PY  - 2026
DA  - 2026/09/26
JO  - Diagnostics (Basel, Switzerland)
DO  - 10.3390/diagnostics16193132
UR  - https://europepmc.org/article/MED/42882287
AB  - Background/Objectives: Deep learning-based chest X-ray (CXR) analysis has shown strong diagnostic performance; however, conventional ensemble methods typically rely on equal-weight aggregation and do not account for disease-specific model behavior or clinician-defined priorities. Methods: We developed a clinician-configurable ensemble framework based on the fuzzy analytic hierarchy process (FAHP) for multi-label CXR classification. The proposed method employs a two-stage design consisting of disease-specific model weighting using multi-criteria decision analysis and utility-guided seed aggregation based on ΔNet Benefit (ΔNB). Five base architectures, including Vision Transformer (ViT), SwinV2 Transformer, DenseNet121, DenseNet121 + ViT and DenseNet121 + SwinV2, were trained on the National Institutes of Health (NIH) ChestX-ray14 dataset using 10 random seeds. Performance was evaluated on a held-out test set using area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), balanced accuracy, and F1 score and compared with single-model baselines, Hard Voting, and Soft Voting. Results: In the seed-level analyses, Soft Voting achieved the highest overall discrimination performance (AUROC, 0.851 ± 0.001; AUPRC, 0.293 ± 0.003), whereas seed-level FAHP showed comparable performance (AUROC, 0.850 ± 0.001; AUPRC, 0.287 ± 0.004). In the final-level analyses, the sensitivity-priority FAHP-ΔNB configuration achieved a macro-averaged AUROC of 0.855 and an AUPRC of 0.299. Hard Voting demonstrated inferior performance, particularly for AUPRC. Per-disease analysis revealed heterogeneous results, with FAHP showing favorable performance in selected conditions, including Effusion, Emphysema, Mass, Nodule, and Pleural Thickening, while Soft Voting remained superior in others. Statistical analysis confirmed that differences between FAHP and Soft Voting were generally modest and disease-dependent. Conclusions: The proposed FAHP-based ensemble maintains performance comparable to strong equal-weight baselines while enabling clinician-configurable, disease-specific model weighting and utility-guided aggregation. This framework provides an aggregation-level transparent approach for incorporating configurable diagnostic priorities into multi-label CXR classification.
SN  - 2075-4418
ER  -