Due to its complexity, biomechanical analysis of upper limbs (UL) often plays a limited role in clinical practice. Clinical assessment tests are typically preferred for evaluating improvements from interventions, but they mainly focus on execution time or reflect the patient’s subjective perception, while compensatory movements are generally not considered. Another challenge in this field is the lack of a standardized methods for locating joint centers, which would be essential for collecting accurate biomechanical data. A key objective of this study was to investigate a 3D freehand ultrasound (3D USMO) approach to estimate the shoulder joint center (GJC), and comparing it to other established methods. Additionally, the study aimed to explore clinical assessments and emphasize the importance of biomechanical analysis in the UL.The 3D USMO combines a 2D ultrasound system with a Vicon 3D motion capture system, transferring the coordinates of bony landmarks captured by the ultrasound into 3D space. The GJC was determined with this approach in non-disabled subjects and patients in sitting and supine positions, as well as with commonly used predictive and functional methods. For patients, MRI data was also obtained for validation. The results showed that 3D USMO was more accurate than other methods, with an average difference of the 3D position of the joint center of 12 mm compared to MRI data. The functional method, however, had the lowest accuracy and repeatability, with a deviation of around 40 mm from the MRI joint center. Inter- and intra-rater reliability were assessed on different days in both in-vitro and in-vivo settings, with higher errors observed in the in-vivo conditions. The largest standard error of measurement (around 3 mm) and minimal detectable change (around 8 mm) were found between test days. 3D USMO was also tested for measuring socket-stump displacements in trans-radial amputees (TR), though it lacked the precision to detect shifts below 2 mm.To enhance the need for biomechanical analysis, various clinical assessments were examined to assess movement strategies. These included a self-designed ADL-Assessment (Activities of Daily Living), the Southampton Assessment Procedure (SHAP), a dynamic range of motion (dROM) assessment, and the SHELF-Test. Data of all participants were captured with a Vicon 3D motion analysis system using an upper limb model.The ADL-Assessment based on eight bimanual and unilateral daily tasks. The focus was on analyzing execution times and compensatory strategies (kinematics summarized by the Arm Profile Score – APS, and Dynamic Time Warping parameters) using motion capturing. Seven TR patients using different prosthetic hand types (MichelAngelo, SensorHand Speed, and Axon Hook) were enrolled. Center of pressure (COP) movement, represented by a 95% confidence ellipse, was also analyzed during each task. Compensation strategy was needed using the Hook prosthesis, particularly in the elbow joint, as it required a different positioning for grasping compared to the other prosthetic hands. However, its precise grip reduced execution times. Generally, limited wrist flexibility and grip options led to compensation movements and longer execution times. Across most tasks, the SensorHand Speed performed the worst in terms of the parameters. This assessment can be used as a helpful tool for clinicians and certified prosthetists to differentiate between the performance of various prosthetic hand types. However, COP analysis was only partially effective in detecting compensatory movements in three of the eight tasks. Future research should focus on bilateral tasks, as unilateral tasks were primarily designed for laboratory settings, or on tasks with higher degrees of freedom needed.The SHAP is a clinically validated 26-item test to evaluate hand function, primarily based on time measurements, without assessing execution quality. Since the total score is calculated using a proprietary reference database, one objective was to establish a new reference database with 20 non-disabled individuals. The study also included 12 patients (5 TR and 7 children who underwent single-bone surgeries) to investigate the correlation between the SHAP’s total score (Linear Index of Function - LIF) and the biomechanical summary index APS. Significant differences in execution times were observed compared to published data, but a strong correlation was found between the LIF and the SHAP’s original reference total score. Additionally, notable differences between left- and right-handed individuals were detected in the reference group. No significant correlation was observed between the LIF, which reflects execution times, and the APS, which reflects compensatory movements. Single-bone patients primarily experienced elbow limitations, while TR patients faced significant restrictions in wrist mobility and elbow flexion, leading to increased compensation in the proximal joints. Consequently, fast SHAP performance does not necessarily indicate high-quality execution, but summary indices can aid in decision-making. The new reference dataset, despite differences from the published data, is suitable for clinical practice and allows for extracting specific items for targeted clinical questions.The dROM assessment, designed to detect range of motion limitations, involved 12 non-disabled individuals as a reference and 11 patients. It successfully distinguished between patient populations by clearly identifying limitations, making it highly recommended for patients without a definitive diagnosis or for tracking upper extremity range of motion progress after interventions.The SHELF-Test, an assessment for trans-radial prosthetic users, requires a high degree of freedom. This part aimed to standardize the test and capture reference data from 20 non-disabled individuals. Seven patients (3 TR and 4 with shoulder pathologies) also participated. The SHELF Test proved effective for TR patients, showing clear compensation, though it was less sensitive for shoulder patients. Future research should investigate additional patient groups to better define the target population for this assessment. These results emphasize the importance of incorporating biomechanical analysis into clinical practice for better evaluation of movement strategies and prosthetic performance.
