Initial Health Education for Participants Initially, all participants received identical health education in the form of thematic lectures once per week for 40 minutes each. The content included the following aspects: emphasising the importance of close cooperation with physicians as key to maintaining health; the process of the occurrence and development of KOA, along with its treatment methods; the role of family and social support in improving the condition; establishing a reasonable daily activity routine, such as protecting the affected knee joint, avoiding long-distance fatigue, mountain climbing, ascending and descending high-rise stairs and various adverse body postures (prolonged standing, kneeling and squatting); and advising those who are overweight to reduce their body weight. On this foundation, the four groups of participants underwent the following interventions.

The PG and JG were injected with autologous PRP at baseline. The injection was performed by the same clinician with 10 years of experience in knee joint injections. Participants were seated with their knees extended, and the needle was inserted through the joint space into the intra-articular cavity of the knee, administering approximately 2.5 mL of PRP into each knee joint. Thirty millilitres of venous blood were drawn from the participant's elbow vein, and approximately 5 mL of PRP was obtained for bilateral knee joint injection by using a TD5A tabletop low-speed centrifuge manufactured in China. The first centrifugation was performed at 2500 revolutions per minute (rpm) for 3 minutes, followed by a second centrifugation at 2300 rpm for 6 minutes.

The IG and JG exercises followed the methods of our previous research, with participants undergoing an 16-week ICQ intervention five times per week. Each session consisted of isometric contractions and relaxations of the quadriceps muscle for 5 seconds each, in sets of 30 repetitions, for a total of 10 sets, with a 30-second rest between sets. During this period, researchers conducted weekly follow-up calls or face-to-face interviews with participants to understand their living conditions, ensuring that they did not change their original lifestyle habits and did not engage in any other form of regular fitness activities. Participants who did not comply were not included in the final data analysis.

The PG and JG were injected with autologous PRP at baseline. The injection was performed by the same clinician with 10 years of experience in knee joint injections. Participants were seated with their knees extended, and the needle was inserted through the joint space into the intra-articular cavity of the knee, administering approximately 2.5 mL of PRP into each knee joint. Thirty millilitres of venous blood were drawn from the participant's elbow vein, and approximately 5 mL of PRP was obtained for bilateral knee joint injection by using a TD5A tabletop low-speed centrifuge manufactured in China. The first centrifugation was performed at 2500 revolutions per minute (rpm) for 3 minutes, followed by a second centrifugation at 2300 rpm for 6 minutes.

Symptoms of KOA participants were evaluated using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). These questions included assessments of pain (5 questions), stiffness (2 questions) and physical function (17 questions), totalling 24 questions. A 0-100 mm visual analogue scale was used for scoring, which was measured and recorded by a professional by using a ruler. The scores for pain (maximum 500), stiffness (maximum 200) and physical function (maximum 1700) were based on the measured values, with higher scores indicating greater severity.

The static and dynamic balance of all participants was tested using a BioDex Balance System (BioDex Medical Systems, USA). The test session required participants to stand with both feet, with their hands placed on their sides. The static balance tests included firm surface with eyes closed (FiEC), foam surface with eyes open (FoEO) and foam surface with eyes closed (FoEC) tests.

The isokinetic strength of both knees (60°/s for 5 repetitions and 180°/s for 5 repetitions) was tested in all participants by using a Germany-manufactured IsoMed 2000 isokinetic testing device. The knee was positioned in a sitting posture with a range of motion of 80°. Participants were required to warm their lower limbs for 10 minutes before the test. The testers provided verbal encouragement throughout the process to ensure reliable test data. The index was the peak torque.

After completing the isokinetic muscle strength test while maintaining the test posture, proprioceptive testing was conducted on the dominant side. The angular velocity was set at 1°/s. The isokinetic device's actuator automatically moved to preset angles of 30° and 60° of knee flexion (with the fully extended knee position at 0°). After each movement to an angle, the actuator returned to the starting position, the participant actively moved to the designated position (the movement stopped automatically after pressing a button), and the difference between the actual and preset angles was recorded. Smaller differences indicate better proprioception, and the average of three trials for each angle was taken.