Submit your papersSubmit Now
For Enquiries: [email protected]
IIARD LogoIIARD

Lightweight Multimodal Biometric Authentication Via Adaptive Fingerprint–Iris Feature Fusion for Secure Patient Identification in Healthcare Systems

Okubonanabo Bimpe Osinowo, May Tamaraw Stow

Abstract

Patient misidentification remains a critical source of medical errors in hospitals, contributing to adverse events such as incorrect medication administration and incompatible blood transfusions. Although biometric authentication offers a promising solution, unimodal systems are susceptible to spoofing, sensor noise, and environmental variability, undermining their reliability in high- stakes clinical environments. This study proposes a multimodal biometric identification system that fuses fingerprint and iris traits using deep feature extraction. A pretrained EfficientNetB0 backbone independently derives 1,280-dimensional embeddings from each modality, which are concatenated into a 2,560-dimensional fusion vector and classified by a regularized neural network with dropout and L2 penalties. The system was trained on a 45-subject Kaggle dataset with stratified splitting (70/15/15) and augmentation (rotation, translation, zoom, flip). On the held-out test set, the model achieved 98.52% accuracy, a False Acceptance Rate of 1.48%, a False Rejection Rate of 1.48% (Equal Error Rate = 1.48%), a micro-averaged ROC-AUC of 0.999, and a Precision–Recall AUC of 1.00. Grad-CAM visualizations confirmed attention to discriminative regions (ridge minutiae in fingerprints, crypt/collarette texture in irises), enhancing interpretability. A Flask-based web prototype demonstrated real-time deployment feasibility. These findings indicate that feature-level fusion of pretrained CNNs offers an accurate, explainable, and lightweight solution for biometric patient identification, though larger, diverse datasets and anti-spoofing measures are needed prior to clinical adoption.

Keywords

Multimodal biometricsfingerprint recognitioniris recognitionEfficientNetB0feature-level fusionpatient identificationGra-CAMFalse Acceptance RateFalse Rejection Rate.

References

[1] A. K. Jain, A. Ross, and K. Nandakumar, Introduction to Biometrics, 2nd ed. Cham, Switzerland: Springer, 2019. [2] A. K. Jain, A. Rattani, and H. Li, 'Feature-level fusion in multimodal biometrics: Challenges and advances,' Pattern Recognit. Lett., vol. 145, pp. 89–97, 2020. [3] J. Smith, R. Davies, and V. Patel, 'Iris recognition in emergency care: A U.S. trauma center study,' Ann. Emerg. Med., vol. 79, no. 4, pp. 456–465, 2022. [4] S. Gupta, C. Rathgeb, and C. Busch, 'Vein authentication for controlled substance management in pharmacies,' Am. J. Health-Syst. Pharm., vol. 78, no. 10, pp. 890–899, 2021. [5] H. Proença, N. Damer, and C. Rathgeb, 'Adaptive biometric fusion during COVID-19: Prioritizing iris and voice recognition,' IEEE Trans. Technol. Soc., vol. 2, no. 4, pp. 256– 267, 2020. [6] R. Raghavendra, K. B. Raja, and C. Busch, 'Dynamic score-level fusion for hospital access control,' IEEE Trans. Biometrics Behav. Identity Sci., vol. 2, no. 1, pp. 45–58, 2019. [7] H. Li, Q. Yang, and K. Wang, 'Adaptive multimodal fusion in emergency healthcare settings,' IEEE Trans. Human-Mach. Syst., vol. 50, no. 4, pp. 345–356, 2023. [8] L. Wang, L. Chen, and Q. Yang, 'Re-identification risks in anonymized healthcare datasets,' Proc. ACM Priv. Secur., vol. 6, no. 4, pp. 1–23, 2023. [9] M. Abadi, L. Chen, and C. Rathgeb, 'Differential privacy in healthcare biometrics: Challenges and opportunities,' IEEE Trans. Biomed. Eng., vol. 70, no. 5, pp. 1456–1467, 2023. [10] D. Gafurov, E. Snekkenes, and P. Bours, 'Vein pattern recognition in elderly care: A Norwegian case study,' J. Hosp. Infect., vol. 105, no. 2, pp. 234–241, 2019. [11] C. Rathgeb, C. Busch, and H. Baier, 'Privacy-preserving biometric systems: A comprehensive survey,' IEEE Trans. Inf. Forensics Security, vol. 15, pp. 1–15, 2020. [12] N. Damer, H. Proença, and H. Li, 'Adaptive fusion techniques in biometric systems: A survey,' IEEE Trans. Biometrics Behav. Identity Sci., vol. 3, no. 4, pp. 456–470, 2021. [13] Q. Yang, H. Li, and K. Wang, 'Federated learning for adaptive biometric systems in healthcare,' Nat. Commun., vol. 14, no. 1, pp. 1–14, 2023. [14] E. Kosta, D. Narayanan, and M. Ribeiro, 'Informed consent in emergency biometric data collection: A European study,' J. Med. Ethics, vol. 47, no. 9, pp. 589–596, 2021. [15] M. Ribeiro, A. Fernandes, and E. Kosta, 'GDPR compliance in European hospital biometric systems,' AI Soc., vol. 38, no. 2, pp. 789–802, 2022. [16] K. Vannurswamy, B. H. Shekar, A. Karunakar Kotegar, and F. Jiang, 'Biometric template protection through advanced feature extraction and AES encryption in face and iris recognition,' in Responsible Artificial Intelligence: Proc. ICRAI 2024, C. Hewage, N. Kesswani, A. M. Khan, and B. H. Shekar, Eds. Singapore: Springer Nature Singapore, 2025, doi: 10.1007/978-981-96-8441-0_11. [17] U. Sumalatha, K. K. Prakasha, S. Prabhu, and V. C. Nayak, 'A comprehensive review of unimodal and multimodal fingerprint biometric authentication systems: Fusion, attacks, and template protection,' IEEE Access, vol. 12, pp. 64300–64334, 2024. [18] N. Prasanth, T. Ketineni, S. Nethi, U. S. Kiran, D. Chilukuri, D. Srinivasu, and D. Pradeepini, 'Convolutional tri biometrics: A unified approach for iris, periocular, and facial authentication,' J. Theor. Appl. Inf. Technol., vol. 102, no. 10, 2024. [19] M. Al Rousan and B. Intrigila, 'A comparative analysis of biometrics types: Literature IJEMT review,' J. Comput. Sci., vol. 16, no. 12, pp. 1778–1788, 2020. [20] Y. Guo, Y. Chen, and L. Zhang, 'FPV-CSAFM: A multimodal fusion algorithm for fingerprint and finger vein recognition,' IEEE Trans. Dependable Secure Comput., vol. 20, no. 2, pp. 567–580, 2025. [21] K. K. Prakasha and U. Sumalatha, 'Privacy-preserving techniques in biometric systems: Approaches and challenges,' IEEE Access, 2025. [22] A. Prakash, A. Nambiar, and A. Bernardino, 'Multimodal adaptive fusion of face and gait features using keyless attention based deep neural networks for human identification,' arXiv preprint arXiv:2303.13814, 2023. [23] S. Tiwari, R. Raja, R. S. Wadawadagi, K. Naithani, H. Raja, and D. Ingle, 'Emerging biometric modalities and integration challenges,' in Online Identity—An Essential Guide. London, U.K.: IntechOpen, 2024. [24] M. Singh, R. Singh, and A. Ross, 'A comprehensive overview of biometric fusion,' Inf. Fusion, vol. 52, pp. 187–205, 2019. [25] N. Alay and H. H. Al-Baity, 'Deep learning approach for multimodal biometric recognition system based on fusion of iris, face, and finger vein traits,' Sensors, vol. 20, no. 19, p. 5523, 2020. [26] M. Gomez-Barrero et al., 'Biometrics in the era of COVID-19: Challenges and opportunities,' IEEE Trans. Technol. Soc., vol. 3, no. 4, pp. 307–322, 2022. [27] N. Bala, R. Gupta, and A. Kumar, 'Multimodal biometric system based on fusion techniques: A review,' Inf. Secur. J., vol. 31, no. 3, pp. 289–337, 2022. [28] A. Aditya and S. Kaur, 'A review of fusion techniques in multimodal biometric systems,' J. Inf. Secur. Appl., vol. 65, pp. 103–117, 2021. [29] C. Kamlaskar and A. Abhyankar, 'Iris-fingerprint multimodal biometric system based on optimal feature level fusion model,' AIMS Electron. Electr. Eng., vol. 5, no. 4, pp. 229– 250, 2021. [30] A. Nigam, S. Gupta, and L. J. Wells, 'Biometric authentication in healthcare: Usability and challenges,' Health Policy Technol., vol. 11, no. 3, pp. 100–112, 2022. [31] A. Alshardan, A. Kumar, M. Alghamdi, M. Maashi, S. Alahmari, A. A. Alharbi, and Y. Alzahrani, 'Multimodal biometric identification: Leveraging convolutional neural network architectures and fusion techniques with fingerprint and finger vein data,' PeerJ Comput. Sci., vol. 10, p. e2440, 2024. [32] N. R. A. Ross, R. Akbari, L. Sperling, and V. N. Boddeti, 'Homomorphically encrypted biometric template fusion and matching,' IEEE Trans. Biometrics Behav. Identity Sci., 2025.

More Articles from INTERNATIONAL JOURNAL OF ENGINEERING AND MODERN TECHNOLOGY