Author(s): RUBERS, J., OBIOL FIBLA, J.2, PAZOS, R.2, MARTÍNEZ VARELA, F.2, ARRIAZA, R.2, MICHAUD, F.1, Institution: UNIVERSIDADE DA CORUÑA, Country: SPAIN, Abstract-ID: 1448

INTRODUCTION:
Low back pain (LBP) is one of the most common musculoskeletal disorders and the leading cause of disability worldwide. Lifetime prevalence ranges from 65-84 %, and in 2020 about 619 million people were affected globally, with projections of 843 million by 2050 [1-2]. LBP is also common in sports, especially weightlifting, where spinal loading is high [3]. Maintaining a neutral spine during lifting is widely recommended to reduce lumbar injury risk. However, achieving this posture during dynamic exercises is challenging, as it requires core stability, proper hip hinge mechanics, and constant posture awareness, which are difficult to monitor in real training settings. This study proposes a wearable system using two inertial measurement units (IMUs) placed on the pelvis and thorax to provide real-time visual and auditory biofeedback when unsafe lumbar flexion occurs. Previous validation confirmed system accuracy and showed that nearly half of participants lifted with poor spinal posture [4]. Therefore, this study evaluates its effectiveness in reducing lumbar injury risk and improving hip hinge mechanics.
METHODS:
Twenty-one healthy adults (7 males, 14 females; 37 +/- 19 years) without injuries affecting performance participated. Two IMUs (MinIMU-9, Pololu Robotics and Electronics, Las Vegas, USA, 26 Hz) were placed at S1 and T10 to estimate pelvis and trunk orientation. After a 5 minute warm-up and calibration of neutral spine posture, participants performed five deadlift repetitions with a self-selected load without biofeedback to assess baseline lumbar flexion. Following instruction from a certified trainer, participants completed a 15 minute training session with real-time visual and auditory biofeedback. Finally, five additional repetitions without biofeedback were performed to assess improvements.
RESULTS:
Pre-training mean lumbar flexion was 35 degrees, while 62 % of participants exceeded the safe threshold of 30 degrees. After the 15 minute biofeedback training session, peak lumbar flexion was significantly reduced (95 % CI 12.72 to 24.30 degrees), with a mean decrease of 18.51 degrees. All participants improved, and 86 % remained below the 30 degrees threshold. All female participants achieved average peak angles below 30 degrees, compared with 57 % of the male participants. These differences may reflect training adaptations and individual anatomical or biomechanical factors. Future research will examine long-term effects and comparison with traditional coaching.
CONCLUSION:
The dual IMU system with real-time biofeedback reduced excessive lumbar flexion and improved hip hinge awareness. This approach may support lower back injury prevention and promote neutral spine control during lifting.

[1] Ferreira, Manuela L et al. The Lancet Rheumatology. 2023, Volume 5, Issue 6, e316 - e329.
[2] Manchikanti L. Pain Physician. 2000. 3(2):167-92. PMID: 16906196.
[3] Strömbäck, E. et al. Orthop. J. Sports Med. 2018, 6, 5.
[4] Michaud F et al. Sensors (Basel). 2021. 21(16):5487.