Search

Abstract Glaucoma is an eye condition that leads to loss of vision and blindness if not diagnosed in time. Diagnosis requires human experts to estimate in a limited time subtle changes in the shape of the optic disc from retinal fundus images. Deep learning methods have been satisfactory in classifying and segmenting diseases in retinal fundus images, assisting in analyzing the increasing amount of images. Model training requires extensive annotations to achieve successful generalization, which can be highly problematic given the costly expert annotations. This work aims at designing and training a novel multi-task deep learning model that leverages the similarities of related eye-fundus tasks and measurements used in glaucoma diagnosis. The model simultaneously learns different segmentation and classification tasks, thus benefiting from their similarity. The evaluation of the method in a retinal fundus glaucoma challenge dataset, including 1200 retinal fundus images from different cameras and medical centers, obtained a $$96.76 \pm 0.96$$ 96.76 ± 0.96 AUC performance compared to an $$93.56 \pm 1.48$$ 93.56 ± 1.48 obtained by the same backbone network trained to detect glaucoma. Our approach outperforms other multi-task learning models, and its performance pairs with trained experts using $$~\sim 3.5$$ ∼ 3.5 times fewer parameters than training each task separately. The data and the code for reproducing our results are publicly available.

Equitation in the French tradition is a school of riding that emphasizes harmonious relations between humans and horses. The best-known community is the Cadre Noir of Saumur, whose specialty is the air above the ground (AAG). No study has yet looked at the horse–rider interaction in this specific context. The purpose of this study was to identify and quantify indicators of AAGs based on the empirical perception of the écuyers expressed by a method of self-confrontation interviews. Fourteen training sessions were the subject of phenomenological and biomechanical approaches. Contact, balance, and hoof-beat, decisive for performance quality, were characterized for 49 AAGs, performed by five horses trained by two expert écuyers, with rein tension meters integrated in their double bridle (curb and snaffle reins) and six inertial measurement units fixed on the limbs, sternum, and croup. Their action was characterized by a peak of 65 ± 39 N on the inside curb rein. They considered that their horse was in balance (forehand inclined 13 ± 7° and −12 ± 9° for the hind hand). After the peak, during the 3.3 ± 2 s the horse’s trunk was stable and the écuyers released the contact until the AAG was perceived as satisfactory by the écuyer. The mixed approach allowed a pattern of action to be envisaged for the écuyer based on contact, balance, and hoof-beat in the execution of AAGs. The quantification of rein tension, trunk movements, and acceleration of the four limbs objectified the expert écuyers’ feeling of developing aptitudes for their actions in the human–horse interactions for improved transmission to young écuyers. The mixed approach used in this study has given rise to new training methods that are transferable to other equestrian activities.