Abstract

The medical biometric dataset comprises 10,000 records collected across 23 patients spanning different demographics, biometric profiles, and temporal variations between 2022 and 2023. It is accumulated from various hospitals, digital health records, and biometric-enabled healthcare security systems. The dataset includes real-world biometric authentication and clinical profiling scenarios while ensuring compliance with standard medical and biometric data regulations. The dataset aligns with the best practices for secure biometric authentication in electronic health records (EHRs), patient identity verification in telemedicine, and forensic biometric applications. The structured representation of multi-modal biometrics ensures that this dataset is viable for clinical validation. The attributes encompass core biometric identifiers, medical conditions influencing biometric readings, and authentication success rates, making it a comprehensive dataset for AI-driven biometric modeling in healthcare applications. The dataset is structured to enable both supervised and unsupervised learning investigations, catering to model training, fairness evaluations, and security-focused research.

Instructions:

The dataset includes demographic attributes such as Age, Gender, and Ethnicity, which are crucial in evaluating algorithmic fairness and bias detection in biometric systems. The biometric attributes include Fingerprint Ridge Density (lines/mm), Iris Texture Complexity (scale of 0-1), Facial Landmark Accuracy (%), ECG Wave Pattern Variance (signal fluctuations), Pupil Dilation Response (mm/s), and Voice Pitch Variability (Hz)—all of which contribute to assessing individual identity verification through multimodal biometric authentication. Additionally, Medical Conditions (e.g., Hypertension, Diabetes, Neuropathy, Cardiac Arrhythmia) are included to analyze their impact on biometric feature variability and AI model adaptability. The Authentication Success variable (0 or 1) provides a binary classification target, enabling the evaluation of biometric recognition system performance across diverse patient profiles. The inclusion of Timestamp (2022-2023) allows for temporal analysis of biometric variations, making it valuable for longitudinal biometric modeling.

Futuristic Significance for Technical Investigations

This dataset holds significant promise for technical investigations in AI-driven biometric security, medical identity authentication, and forensic applications. Future research can leverage this dataset to design and validate fairness-aware deep learning models, ensuring that biometric authentication is robust against demographic biases. It also enables investigations into biometric signal variations due to medical conditions, offering new insights into how diseases like neuropathy or arrhythmia influence biometric signals. The dataset’s time-series nature (2022-2023) makes it applicable for temporal biometric analysis, aiding in the development of adaptive AI-driven authentication models that can dynamically adjust to biometric changes over time. Furthermore, researchers can explore multi-modal biometric fusion strategies to improve authentication success rates, ensuring high-security standards in medical and forensic biometric applications. Ultimately, this dataset serves as a cornerstone for AI-driven, mathematically enhanced, and ethically transparent biometric research in healthcare and security applications.