Background and Objectives: We investigated the effects of sling-suspension-based active shoulder joint exercise training on shoulder joint subluxation, pain, muscle strength, and upper extremity function in patients with subacute stroke. Materials and Methods: Twenty-eight patients with subacute stroke were randomly assigned to either the sling-suspension-based active shoulder joint exercise (SASE) group (n = 14) or the motorized upper extremity exercise (MUEE) group (n = 14). The SASE group actively performed shoulder joint flexion, extension, abduction, adduction, external and internal rotation, and horizontal abduction and adduction using a sling suspension system, whereas the MUEE group underwent an exercise program using a motorized upper extremity exercise machine. All participants underwent a 4-week intervention with 30 min of exercise once a day for 5 days a week. Additionally, both groups received general physical therapy and functional electrical stimulation for 30 min twice a day for 5 days a week. Shoulder joint subluxation was measured by radiographic examination before and after training, and pain was evaluated in the splenius, upper trapezius, and infraspinatus muscles using pressure parameters. In addition, a manual muscle tester was used to assess the muscle strength of the shoulder joint flexors, extensors, abductors, adductors, and external and internal rotators, and the Fugl–Mayer Assessment (FMA) and Manual Functional Test (MFT) were used to evaluate upper extremity function. Results: A significant group–time interaction was observed for pain, with F-values of F(1, 26) = 7.470, p < 0.011 for the splenius and F(1, 26) = 9.623, p < 0.005 for the upper trapezius. A significant time–group interaction was observed for the muscle strength of the shoulder, with F-values of F(1, 26) = 13.211, p < 0.001; F(1, 26) = 4.974, p = 0.035 and F(1, 26) = 9.674, p = 0.004 for flexors, abductors, and external rotators, respectively. A significant time–group interaction was observed in the FMA, with F-values of F(1, 26) = 13.243, p < 0.001. When comparing the interaction effects between time and group for MFT scores, a significant difference was observed, with F-values of F(1, 26) = 32.386, p < 0.001. Conclusions: This study confirmed that sling-suspension-based active shoulder joint exercises are effective in improving shoulder joint subluxation, pain, muscle strength, and upper extremity function in patients with subacute stroke. [ABSTRACT FROM AUTHOR]