Automated Patient Identity Verification Framework for Multimodal Medical Imaging Using Deep Metric Learning and Domain Adaptation
- 1The University of Osaka
Abstract
Purpose: Patient identity management is fundamental in healthcare information systems, as identification inconsistencies can compromise patient safety, data integrity, and clinical workflow efficiency. Reliable linkage of medical images acquired across different imaging modalities remains challenging because of variations in image appearance, acquisition geometry, and imaging characteristics. In this study, we developed an automated patient identity verification framework for multimodal medical imaging using deep metric learning and data-augmented domain adaptation (DADA). Methods: The proposed framework learned modality-invariant patient representations from labeled source-domain data while leveraging unlabeled target-domain data to mitigate cross-modality distribution shifts. Although patient identity annotations were available within the research dataset for evaluation purposes, target-domain labels were intentionally excluded from the DADA optimization process. Chest radiographs and computed tomography (CT) scout images obtained under routine clinical conditions were retrospectively collected and used for evaluation. Verification performance was assessed using receiver operating characteristic (ROC) analysis, with the area under the ROC curve (AUC) used as the primary performance metric. Results: The proposed framework achieved consistently high verification performance across all evaluation conditions, with AUC values ranging from 0.9989 to 0.9999. Similarity-score distributions demonstrated distinct separation between same-patient and different-patient image pairs despite substantial differences between imaging modalities. Conclusion: These findings indicate that patient-specific anatomical representations can be preserved across heterogeneous imaging domains through metric learning and domain adaptation. The proposed framework may serve as a practical infrastructure component for patient identity management, multimodal data integration, quality assurance, and patient safety applications within healthcare information systems.
Related
- Citation reliability of frontier large language models in medical writing and its automated verification
- Hybrid retrieval–verification framework for hallucination reduction in medical LLMs
- Professional Change and Identity in Contemporary Physiotherapy: A Six‐Domain Conceptual Framework
- Incorporating Childhood Identity-Based Discrimination into the Adverse Childhood Experiences Framework
- Multimodal deep learning with particle swarm guided feature refinement for automated pain detection
BibTeX
@misc{ueda2026automated,
title = {{Automated Patient Identity Verification Framework for Multimodal Medical Imaging Using Deep Metric Learning and Domain Adaptation}},
author = {Y. Ueda and T. Ishida},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.11.26360177},
url = {https://www.medrxiv.org/content/10.64898/2026.08.11.26360177v2},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Automated Patient Identity Verification Framework for Multimodal Medical Imaging Using Deep Metric Learning and Domain Adaptation
AU - Y. Ueda
AU - T. Ishida
PY - 2026
DA - 2026/08/11
PB - medRxiv
DO - 10.64898/2026.08.11.26360177
UR - https://www.medrxiv.org/content/10.64898/2026.08.11.26360177v2
AB - Purpose: Patient identity management is fundamental in healthcare information systems, as identification inconsistencies can compromise patient safety, data integrity, and clinical workflow efficiency. Reliable linkage of medical images acquired across different imaging modalities remains challenging because of variations in image appearance, acquisition geometry, and imaging characteristics. In this study, we developed an automated patient identity verification framework for multimodal medical imaging using deep metric learning and data-augmented domain adaptation (DADA). Methods: The proposed framework learned modality-invariant patient representations from labeled source-domain data while leveraging unlabeled target-domain data to mitigate cross-modality distribution shifts. Although patient identity annotations were available within the research dataset for evaluation purposes, target-domain labels were intentionally excluded from the DADA optimization process. Chest radiographs and computed tomography (CT) scout images obtained under routine clinical conditions were retrospectively collected and used for evaluation. Verification performance was assessed using receiver operating characteristic (ROC) analysis, with the area under the ROC curve (AUC) used as the primary performance metric. Results: The proposed framework achieved consistently high verification performance across all evaluation conditions, with AUC values ranging from 0.9989 to 0.9999. Similarity-score distributions demonstrated distinct separation between same-patient and different-patient image pairs despite substantial differences between imaging modalities. Conclusion: These findings indicate that patient-specific anatomical representations can be preserved across heterogeneous imaging domains through metric learning and domain adaptation. The proposed framework may serve as a practical infrastructure component for patient identity management, multimodal data integration, quality assurance, and patient safety applications within healthcare information systems.
ER -