biostudies-literatureUnknown14Corcoran E<h4>Introduction</h4>Plant image datasets have the potential to greatly improve our understanding of the phenotypic response of plants to environmental and genetic factors. However, manual data extraction from such datasets are known to be time-consuming and resource intensive. Therefore, the development of efficient and reliable machine learning methods for extracting phenotype data from plant imagery is crucial.<h4>Methods</h4>In this paper, a current gold standard computed vision method for detecting and segmenting objects in three-dimensional imagery (StartDist-3D) is applied to X-ray micro-computed tomography scans of oilseed rape (<i>Brassica napus</i>) mature pods.<h4>Results</h4>With a relatively minimal training effort, this fine-tuned StarDist-3D model accurately detected (ValidaFrontiers in plant science1120182https://www.ebi.ac.uk/biostudies/studies/S-EPMC9998914biostudies-literatureAutomated extraction of pod phenotype data from micro-computed tomography.PMC9998914Kurup SCorcoran ESiles LAhnert SfalseAutomated extraction of pod phenotype data from micro-computed tomography.<h4>Introduction</h4>Plant image datasets have the potential to greatly improve our understanding of the phenotypic response of plants to environmental and genetic factors. However, manual data extraction from such datasets are known to be time-consuming and resource intensive. Therefore, the development of efficient and reliable machine learning methods for extracting phenotype data from plant imagery is crucial.<h4>Methods</h4>In this paper, a current gold standard computed vision method for detecting and segmenting objects in three-dimensional imagery (StartDist-3D) is applied to X-ray micro-computed tomography scans of oilseed rape (<i>Brassica napus</i>) mature pods.<h4>Results</h4>With a relatively minimal training effort, this fine-tuned StarDist-3D model accurately detected (Valida2023-01-01T00:00:00Z20232025-04-04T18:49:05.315Z2025-04-04T18:49:05.315ZS-EPMC99989143690942510.3389/fpls.2023.1120182