Disorders of human joints manifest during dynamic movement, yet no objective tools are widely available for clinicians to assess or diagnose abnormal joint motion during functional activity. Machine learning tools have supported advances in many applications for image interpretation and understanding and have the potential to enable clinically and economically practical methods for objective assessment of human joint mechanics. We performed a study using convolutional neural networks to autonomously segment radiographic images of knee replacements and to determine the potential for autonomous measurement of knee kinematics. The autonomously segmented images provided superior kinematic measurements for both femur and tibia implant components. We believe this is an encouraging first step towards realization of a completely autonomous capability to accurately quantify dynamic joint motion using a clinically and economically practical methodology

Aims. Interest in soft tissue Radiostereometric Analysis (RSA) is rising. Previous authors have tried, with varying levels of success, to use this technique to analyse the intra-substance portion of anterior cruciate ligament (ACL) graft constructs. These methods were either prone to large amounts of marker migration, deemed unsuitable for in-vivo use or, where alternative markers such as stainless steel sutures were used, lost the inherent accuracy that made RSA an attractive tool in the first place. We describe a modification of tantalum marker balls that allows for a new method of secure fixation to soft tissue in order to accurately analyse stretch, displacement and, potentially, dynamic movement using RSA. Methods. 1.5 mm tantalum tendon markers were predrilled with 0.3 mm holes, allowing them to be sutured directly to soft tissue. Using a previously described ACL graft model, the amount of marker ball migration was then analysed using RSA after cyclical loading between 20 N and 170 N at 25 Hz for 225,000 cycles. Results. None of 40 balls loosened or became detached from the tendon grafts after 225,000 cycles. Mean migration of the tendon attached marker balls was < 0.1 mm. The maximum imprecision in determining the distance between two markers as assessed by intra- and inter-observer variability was 0.082 mm. Analysis of elastic stretch, plastic stretch, stiffness and localisation of stretch within the graft model was possible due to the low migration levels. Conclusions. This method of attachment of marker beads to soft tissue is simple, reliable and mechanically sound. This technique could potentially be used not only for the post-operative assessment of ACL reconstruction, but in all areas where soft tissue RSA is of interest

Introduction. Shoulder motion results from a complex interaction between the interconnected segments of the shoulder girdle. Coordination is necessary for normal shoulder function and is achieved by synchronous and coordinated muscle activity. During rotational movements, the humeral head translates on the glenoid fossa in the anterior-posterior plane. Tension developed by the rotator cuff muscles compresses the humeral head into the glenoid fossa. This acts to limit the degree of humeral head translation and establishes a stable GH fulcrum about which the arm can be moved. Previous studies have been limited by the use of contrived movement protocols and muscular coordination has not been previously considered with regard to shoulder rotation movements. This study reports the activation profile and coordination of 13 muscles and 4 muscle groups during a dynamic rotational movement task based on activities of daily living. Methods. Eleven healthy male volunteers were included in the study. Electromyography (EMG) was recorded from 13 muscles (10 surface and 3 fine-wire intramuscular electrodes) using a wireless EMG system. EMG was recorded during a movement task in which the shoulder was consecutively rotated internally (phase 1) and externally (phase 2) with a weight in the hand. Muscle group data was calculated by ensemble averaging the activity of the individual component muscles. Mean signal amplitude and Pearson correlation coefficient (PCC) analysed muscle activation and coordination, respectively. Results. The mean length of phase 1 (internal rotation) and phase 2 (external rotation) was 1.1s (SD+0.15) and 1.09s (SD+0.18), respectively with no significant difference between them. Mean signal amplitude was significantly higher during external rotation for the anterior, middle and posterior deltoid, teres major and the rotator cuff muscles (Table 1). Significant positive correlations were identified between the activation patterns of the deltoid and rotator cuff groups (PCC=0.95, p=<0.001), the deltoid and latissimus dorsi-teres major groups (PCC=0.74, p=<0.001) and the latissimus dorsi-teres major and rotator cuff groups (PCC=0.87, p=<0.001) (Figure1). Discussion. The subscapularis is extensively described as an internal rotator of the glenohumeral joint; however, during this study it was primarily active during external rotation. During activities of daily living the subscapularis balances the force generated by the supraspinatus and infraspinatus by contracting eccentrically as external rotation progresses. This balance between the anterior and posterior rotator cuff maintains anterior-posterior stability of the humeral head on the glenoid fossa. There is a highly coordinated and synchronous relationship between all the major muscle groups of the shoulder during rotational activities, which ensures glenohumeral joint stability. The function of the shoulder muscles is task specific. This has important implications when considering the impact of muscle pathology on shoulder dysfunction and the treatment strategies employed