Score level multibiometrics fusion approach for healthcare

Abstract

Biometric traits such as iris and face can help for the elementary assessment of human diseases and healthcare monitoring. It has several advantages such as increased patient and staff (doctors and nurses) safety, the accuracy and quality of the healthcare system, reduction of the healthcare fraud. In addition, it provides a secure way to detect the inhabitant’s mood and ocular pathologies in order to treat them. The paper introduces a prototype for biometric-based healthcare monitoring. In the proposed prototype, the patient/user seeking for healthcare assistance can send a request by his/her biometric traits. The biometric traits are processed in the cloud management. The caregiver with valid identification/verification can receive the request and analyze it in order for further treatment. This paper also introduces an efficient multibiometrics fusion framework based on Aczel-Alsina triangular norm. The proposed approach utilizes the 1D-log Gabor iris features, two-directional two-dimensional modified fisher principal component analysis ( $$(2\hbox {D})^{2}$$ MFPCA) and Complex Gabor Jet Descriptor face features to be used for healthcare monitoring. Results show that the multibiometrics fusion approach has better performance compared with the previous fusion approaches.