Introduction. Effective recovery of neuromuscular control after Anterior Cruciate Ligament Reconstruction (ACL-R) is crucial for regaining appropriate mechanical function of the knee joint and preventing further injuries. Muscle onset timing immediately before unplanned Change-of-Direction (CoD) might represent a key parameter for assessing neuromuscular control strategies in Return-to-Sport evaluation. With the growing use of artificial intelligence for electromyography (EMG) analysis, Long Short-Term Memory recurrent neural network for Muscle Activity Detection (LSTM-MAD) [1] proved successful in detecting muscle onsets from EMG signals. This study aims to compare muscle activation onset time in ACL-R and control athletes during an unplanned 90° CoD task, adopting LSTM-MAD for EMG-onset detection. Methods. Ten ACL-R and 17 control athletes performed 3 unplanned CoD for each leg with speed gates indicating the CoD direction. Foot contact was detected by two force plates [2]. EMG signals of Vastus Lateralis (VL), Vastus Medialis (VM), Semitendinosus (ST), and Biceps Femoris (BF) muscles were acquired bilaterally, band-pass filtered (20–450 Hz) and standardized with zero mean and unit variance. Muscle activation onset times prior to CoD (Tonset) were normalized by the duration of the swing phase (Tswing) (see Fig. 1.A) and then multiplied by 100 to obtain Tnorm. The median value across three trials was considered for the analysis. For each muscle, a 1-way ANOVA with Bonferroni adjustment for multiple comparisons was performed to test differences in Tnorm between ACL-R athletes and controls. Results. Statistically significant differences in Tnorm between populations were found for the VM (ACL-R: 84±5 % (mean±SE); controls: 95±2 %; p=0.018), VL (ACL-R: 85±5 %; controls: 98±1 %; p<0.001) and BF (ACL-R: 80±4 %; controls: 91±3 %; p=0.013) muscles (see Fig. 1.B). Discussion. Despite being cleared to Return to Sport, ACL-R athletes still exhibit alterations in neuromuscular control during an unplanned task compared to controls, thought to predispose to either re-injury or to contralateral ACL injury. Notably, early muscle activation onset may reflect increased pre-tension, serving as a protective mechanism by increasing joint stiffness in preparation for CoD movement.

Delayed muscle onset in athletes after anterior cruciate ligament reconstruction during unplanned change-of-directions / Russo, F.; Ghislieri, M.; Baldazzi, A.; Borzuola, R.; Rum, L.; Bergamini, E.; Agostini, V.. - In: GAIT & POSTURE. - ISSN 0966-6362. - ELETTRONICO. - 122:(2025). (Intervento presentato al convegno 25th National Congress of the Italian Society of Clinical Movement Analysis(SIAMOC) tenutosi a Cagliari (Ita) nel 1-4 October 2025) [10.1016/j.gaitpost.2025.08.009].

Delayed muscle onset in athletes after anterior cruciate ligament reconstruction during unplanned change-of-directions

Ghislieri, M.;Agostini, V.
2025

Abstract

Introduction. Effective recovery of neuromuscular control after Anterior Cruciate Ligament Reconstruction (ACL-R) is crucial for regaining appropriate mechanical function of the knee joint and preventing further injuries. Muscle onset timing immediately before unplanned Change-of-Direction (CoD) might represent a key parameter for assessing neuromuscular control strategies in Return-to-Sport evaluation. With the growing use of artificial intelligence for electromyography (EMG) analysis, Long Short-Term Memory recurrent neural network for Muscle Activity Detection (LSTM-MAD) [1] proved successful in detecting muscle onsets from EMG signals. This study aims to compare muscle activation onset time in ACL-R and control athletes during an unplanned 90° CoD task, adopting LSTM-MAD for EMG-onset detection. Methods. Ten ACL-R and 17 control athletes performed 3 unplanned CoD for each leg with speed gates indicating the CoD direction. Foot contact was detected by two force plates [2]. EMG signals of Vastus Lateralis (VL), Vastus Medialis (VM), Semitendinosus (ST), and Biceps Femoris (BF) muscles were acquired bilaterally, band-pass filtered (20–450 Hz) and standardized with zero mean and unit variance. Muscle activation onset times prior to CoD (Tonset) were normalized by the duration of the swing phase (Tswing) (see Fig. 1.A) and then multiplied by 100 to obtain Tnorm. The median value across three trials was considered for the analysis. For each muscle, a 1-way ANOVA with Bonferroni adjustment for multiple comparisons was performed to test differences in Tnorm between ACL-R athletes and controls. Results. Statistically significant differences in Tnorm between populations were found for the VM (ACL-R: 84±5 % (mean±SE); controls: 95±2 %; p=0.018), VL (ACL-R: 85±5 %; controls: 98±1 %; p<0.001) and BF (ACL-R: 80±4 %; controls: 91±3 %; p=0.013) muscles (see Fig. 1.B). Discussion. Despite being cleared to Return to Sport, ACL-R athletes still exhibit alterations in neuromuscular control during an unplanned task compared to controls, thought to predispose to either re-injury or to contralateral ACL injury. Notably, early muscle activation onset may reflect increased pre-tension, serving as a protective mechanism by increasing joint stiffness in preparation for CoD movement.
2025
File in questo prodotto:
File Dimensione Formato  
1-s2.0-S096663622500606X-main.pdf

accesso riservato

Descrizione: Fuul-text
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 488.75 kB
Formato Adobe PDF
488.75 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
RUSSO_F_1.pdf

embargo fino al 18/10/2026

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Creative commons
Dimensione 233.61 kB
Formato Adobe PDF
233.61 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/3004394