Search

Injuries to the shoulder are very common in sports that involve overhead arm or throwing movements. Strength training of the chest muscles has the potential to protect the shoulder from injury. Kinematic and kinetic data were acquired in 20 healthy subjects (age: 24.9 ± 2.7 years) using motion capture, force plates for the bench press exercises and load cells in the cable for the cable pulley exercises with 15% and 30% of body weight (BW). Joint ranges of motion (RoM) and joint moments at the shoulder, elbow and wrist were derived using an inverse dynamics approach. The maximum absolute moments at the shoulder joint were significantly larger for the cable pulley exercises than for the bench press exercises. The cable cross-over exercise resulted in substantially different joint angles and loading patterns compared to most other exercises, with higher fluctuations during the exercise cycle. The present results indicate that a combination of bench press and cable pulley exercises are best to train the full RoM and, thus, intra-muscular coordination across the upper limbs. Care has to be taken when performing cable cross-over exercises to ensure proper stabilisation of the joints during exercise execution and avoid joint overloading.

Measuring joint kinematics is a key requirement for a plethora of biomechanical research and applications. While x-ray based systems avoid the soft-tissue artefacts arising in skinbased measurement systems, extracting the object's pose (translation and rotation) from the x-ray images is a time-consuming and expensive task. Based on about 106'000 annotated images of knee implants, collected over the last decade with our moving fluoroscope during activities of daily living, we trained a deep-learning model to automatically estimate the 6D poses for the femoral and tibial implant components. By pretraining a single stage of our architecture using renderings of the implant geometries, our approach offers personalised predictions of the implant poses, even for unseen subjects. Our approach predicted the pose of both implant components better than about 0.75 mm (in-plane translation), 25 mm (out-of-plane translation), and 2° (all Euler-angle rotations) over 50% of the test samples. When evaluating over 90% of test samples, which included heavy occlusions and low contrast images, translation performance was better than 1.5 mm (in-plane) and 30 mm (out-ofplane), while rotations were predicted better than 3-4°. Importantly, this approach now allows for pose estimation in a fully automated manner., PLoS ONE, 17 (6 6), ISSN:1932-6203