John Deck

The paper presents a machine learning pipeline for automating flower detection on herbarium specimens, resulting in 11.1 million labeled records but highlighting the need for full metadata digitization.

PhenoVision-Leaf uses a Vision Transformer model trained on crowdsourced data to generate global-scale leaf phenology observations from iNaturalist records.

PhenoVision automates plant phenology data extraction from field images, achieving high accuracy in detecting flowers and fruits, and expands global phenology data coverage.

The paper quantifies the loss of spatiotemporal metadata in genomic datasets over time and advocates for improved metadata preservation practices.

The paper outlines ten lessons for designing, provisioning, and reusing persistent identifiers to improve data integration in life sciences.