roi_table

roi_table performs computational processing and provenance tracking of large high-throughput biomedical datasets, including next-generation DNA sequencing, to support statistical and machine-learning analyses for applications such as genomics and finger vein biometric recognition.

Key Features:

• Transparency and Reproducibility: Automatically tracks all computational processes and provenance to enable inspection, publication, and reuse of analyses.
• Scalability for Biomedical Analyses: Supports large-scale analysis and interpretation of vast high-throughput datasets generated by technologies such as next-generation DNA sequencing.

Scientific Applications:

• Genomics: Processing and analysis of next-generation DNA sequencing datasets to support statistical and machine-learning workflows.
• Finger Vein Biometric Recognition: Analysis of finger vein data using (2D)² PCA for feature extraction, metric learning for individualized classification, KNN classifiers tailored per individual, and SMOTE to address class imbalance, with reported recognition rates up to 99.17% in studies.

Methodology:

Computational methods explicitly include (2D)² PCA for feature extraction, metric learning for individualized classification, KNN classifiers tailored per individual, SMOTE for class-imbalance mitigation, and general statistical processing of large datasets.