Increased Pre-Activation and Co-Contraction in ACL-Reconstructed Athletes: Insights from AI-based EMG Analysis

INTRODUCTION: Anterior Cruciate Ligament (ACL) tears disrupt the neural structures within the ligament, impairing the neuromuscular control of knee-stabilizing muscles. Consequently, muscle activity patterns are a crucial area of research in return-to-sport evaluation after ACL Reconstruction (ACL-R). Artificial intelligence-based methods may help identify and extract meaningful parameters from these patterns. This study aims at accurately and reliably quantifying knee muscle pre-activation and timing-based co-contraction indexes in athletes with and without ACL-R during sports-specific landing tasks using an artificial intelligence approach. METHODS: Eleven athletes with ACL-R and 18 Control Athletes (CA) performed two landing tasks: single-leg hop and single-leg cross drop landing. EMG signals were recorded bilaterally from four knee-stabilizing muscles: Biceps Femoris (BF), SemiTendinosus (ST), Vastus Lateralis (VL), and Vastus Medialis (VM). Muscle pre-activation onsets prior to landing and timing-based co-contraction indexes were computed through a pre-trained deep learning-based muscle activity detector (LSTM-MAD). To ensure comparability with state-of-the-art approaches, amplitude-based EMG co-contraction indexes were also computed. RESULTS: LSTM-MAD estimated muscle pre-activation onset with an error under 23 ms compared to manual segmentations by three experts. During single-leg hops, ACL-R athletes exhibited significantly greater BF (ACL-R: -193±12 ms; CA: -152±5 ms; p=0.002) and ST (ACL-R: -189±11 ms; CA: -140±4 ms; p<0.001) pre-activations and longer co-contraction durations (ACL-R: 69±2%; CA: 54±2%; p<0.001) compared to CA. During single-leg cross drop landings, ACL-R athletes showed greater BF (ACL-R: -234±25 ms; CA: -150±11 ms; p=0.02) pre-activations and longer BF and ST co-contractions (ACL-R: 55±5%; CA: 35±2%; p=0.003) compared to CA. At return-to-sport, ACL-R athletes demonstrated greater BF and ST pre-activation and co-contraction during landing tasks. CONCLUSIONS: Integrating artificial intelligence-based methods to assess neuromuscular control could improve the effectiveness of rehabilitation protocols, facilitating safer return-to-sport decisions.