Application of Quality Management System in the Research Process: A Case Study for Plant Phenotyping Research

Abstract

Phenomics research, driven by advancements in imaging and image processing, enables high-throughput measurements of plant traits, providing insights into growth, tissue development, and biochemical states. However, data accuracy is critical to reliable outcomes, especially in complex methods like 3D reconstruction and hyperspectral imaging. This study demonstrates the role of Quality Management Systems (QMS) in enhancing the research process in plant phenotyping. The study emphasizes the importance of a robust data quality assurance pipeline, focusing on error identification and improving data labeling processes through semi-automation. Root Cause Analysis (RCA) was employed to address discrepancies in annotated datasets and identify critical issues, such as misalignment in experimental protocols and operational errors, including the misplacement of irrigation hoses during data collection. Corrective actions, such as photo documentation and procedural revisions, significantly improved data quality. Additionally, algorithmic support streamlined the annotation process, increasing efficiency and data reliability. This integrated approach underscores the necessity of quality control in research, especially for geographically distributed teams working under variable conditions, and highlights the broader applicability of QMS in optimizing research outputs.

Code and data availability

The paper describes a QMS case study using LeasyScan/PlantEye F600 3D scans and Segments.ai annotation, but provides no public deposit of phenotype data, images, code, or models. Segments.ai and Phenospex are third-party tools/platforms, not paper-specific assets; no availability statements or author URLs for data or '

No evidence-backed public reproduction asset is currently recorded.