Relation of computerized tomography Hounsfield unit measurements and internal components of fresh chestnuts (Castanea spp.)

Little is known about Chestnut (Castanea spp.) in vivo cultivar characterization and morphology, the effect of mechanical harvesting, peelability, postharvest mold, physiological kernel decay, and quality. Thus, there is a need to develop a nondestructive technique able to assess fresh chestnut internal components, especially disorders. In this study, a medical grade computed tomography (CT) was used to obtain transversal two-dimensional (2D) images from decayed and healthy fresh chestnuts, from the hybrid cultivar ‘Colossal’ and ‘Chinese seedlings’. Attenuation coefficients, referred to as Hounsfield-units (HU) or CT numbers, were acquired from different 2D CT image regions including air, and several chestnut components containing decayed tissue, healthy tissue, various imperfections such as pellicle invagination into healthy kernel, and void spaces. Results offer an in vivo accurate insight of fresh intact chestnuts, and suggest that CT technology is appropriate for inline sorting. HU measurements can be used as a nondestructive predictor of fresh chestnut internal components with a 90.6% overall accuracy rate.

► Computer tomography (CT) technology is examined as a novel tool to study fresh in vivo chestnut components.
► Hounsfield unit (HU) values obtained from CT images are used to determine the presence of internal components of chestnut with 90.6% accuracy.
► Essential information to develop classification algorithms to sort chestnuts based on their internal quality is offered.
► CT technology is able to detect effects of post- and pre-harvest handling of chestnuts and may be appropriate for inline sorting.