Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Reza Ahmadi has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Comparison of Regression Models for Predicting Lower-Limb Joint Angles During Stationary Cycling

    by Reza Ahmadi, Shahram Rasoulian, Reza Kakavand, Peyman Tahghighi, Walter Herzog, Amin Komeili

    Accurate measurement of lower-limb joint kinematics is integral for developing effective rehabilitation programs and understanding musculoskeletal biomechanics. While previous studies have achieved good accuracy in predicting primary joint angles in gait, few studies have explored machine learning (ML) models tailored specifically to assess cycling biomechanics, and comparative evaluations for inter-subject generalization remain sparse. This study addresses these gaps by comparing the performance of three ML models, Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Transformer, for predicting lower-limb joint angles during stationary cycling. Data were collected from 15 participants using a motion capture system synchronized with pedal encoders, capturing kinematics under varying saddle positions and cadences. The models were trained to predict pelvis, hip, knee, and ankle joint kinematics using features such as crank angles, saddle positions, and anthropometric data. The Transformer model achieved the lowest prediction errors, with inter-subject root mean square error (RMSE) values ranging from 2.7° ± 0.7° to 4.4° ± 0.4°, but was 8 times more computationally intensive than the CNN. Significant differences in nRMSE were observed among models across joint movements in both intra-subject and inter-subject evaluations. The CNN model offered a favorable trade-off, with slightly higher errors (3.7° ± 0.7° to 6.1° ± 1.3°) but much faster processing times compared to the Transformer model, making it suitable for real-time applications. The SVM model exhibited the highest errors (4.8° ± 1.2° to 7.0° ± 1.4°) and limited predictive performance. This study highlights the potential of ML-based approaches to advance biomechanical cycling analysis, offering insights into joint motions and supporting applications in rehabilitation and performance monitoring.

  • Open Access

    Muscle synergy adaptation with fatigue development in constant-power cycling

    by Shahram Rasoulian, Reza Ahmadi, Samira Fazeli Veisari, Thomas K. Uchida, Walter Herzog, Amin Komeili

  • Open Access

    Effect of muscle fatigue on cycling asymmetry during a constant-power test

    by Shahram Rasoulian, Seyed Hamidreza Heidary, Reza Ahmadi, Samira Fazeli Veisari, Walter Herzog, Amin Komeili

    Asymmetry in cycling may reduce mechanical efficiency and increase the risk of overuse injuries, particularly as fatigue develops. While asymmetry is often assessed using global power output, analyzing directional force components may reveal biomechanical patterns more relevant to performance and injury. In this study, we investigated the effect of progressive fatigue on bilateral asymmetry in power and pedal force components during a submaximal constant-power cycling test. Twenty healthy recreational participants performed prolonged cycling at 70% of their individual peak power until exhaustion while three-dimensional pedal forces were recorded using an instrumented ergometer. Asymmetry was quantified using the Normalized Symmetry Index (NSI) for power and for tangential, radial, and mediolateral force components, and evaluated at five time points representing progressive fatigue. Fatigue did not significantly alter mean relative interlimb asymmetry in power output or pedal force components as quantified by NSI, nor did it significantly change mean absolute interlimb force or power differences. However, cycle-to-cycle variability of the asymmetry measures and absolute interlimb differences changed across time, suggesting that fatigue influenced variability in limb coordination rather than producing a systematic shift in mean left-right force sharing. Asymmetry also differed significantly between force directions, with mediolateral forces showing the largest NSI range (-10% to 8%) compared to radial (-3% to 6%) and tangential (-7% to 3%) forces. These findings suggest that, in healthy individuals performing seated constant-power cycling, fatigue-related changes may be expressed more strongly through variability and direction-specific force behavior than through systematic changes in mean relative or absolute interlimb asymmetry.


  • Open Access

    Variation of Lower-Limb Muscle Activation Asymmetry in Step Incremental and Constant-Power Pedaling Exercise

    by Seyed Hamidreza Heidary, Reza Ahmadi, Shahram Rasoulian, Samira Fazeli Veisari, David Auslander, Saied Jalal Aboodarda, Amin Komeili

    Asymmetry, defined as unequal neuromuscular activation or mechanical performance between contralateral limbs, plays a critical role in cycling efficiency and injury risk. While kinematic and kinetic measures are commonly used to assess asymmetry, surface electromyography (EMG) signals offer an additional perspective on neuromuscular asymmetry. This study evaluated muscle activation asymmetry during cycling using the Normalized Symmetry Index (NSI), a metric that quantifies differences in kinematics and kinetics between limbs, where higher values indicate greater asymmetry. NSI was calculated from EMG recordings of seven lower-limb muscles under two test conditions: step incremental and constant-power cycling to task failure. Twenty recreationally active participants performed both tests on a stationary ergometer while EMG data were collected bilaterally. Step incremental cycling resulted in a significant reduction in NSI for key muscles in the quadriceps group: vastus medialis (from 44% to 21%, p < 0.001), vastus lateralis (from 45% to 22%, p = 0.002), rectus femoris (from 54% to 24%, p < 0.001), and biceps femoris (from 52% to 29%, p = 0.003). No significant changes were observed for the tibialis anterior, soleus, or gastrocnemius medialis. In contrast, under constant-power conditions, NSI values remained unchanged over time for all muscles (all p > 0.05), with average NSI values ranging from 12% to 30%, indicating consistent bilateral activation. These findings highlight the sensitivity of surface EMG in detecting workload-dependent changes in muscle activation asymmetry and suggest that higher cycling intensities, compared to lower ones, may promote more balanced engagement of primary cycling muscles.


  • Open Access

    Reproducibility of Cycling Kinetics on an Ergometer Designed to Quantify Asymmetry

    by Sierra Sweeney, Shahram Rasoulian, Atousa Parsaei, Hamidreza Heidary, Reza Ahmadi, Samira Fazeli Veisari, Saied Jalal Aboodarda, Amin Komeili

    Cycling-based rehabilitation is a non-invasive intervention for individuals with lower limb asymmetries. However, current cycling devices lack comprehensive biomechanical feedback and cannot assess asymmetry. Our lab has developed a novel cycle ergometer equipped with three-dimensional force pedals, a seat post and handlebar force sensors, which allow for a comprehensive analysis of asymmetry across a fatiguing task. This study assessed the reproducibility of the cycling kinetics and asymmetry index derived from this device during incremental and constant load cycling tasks to volitional failure. Eighteen participants completed incremental and constant-load tests, each across two identical sessions. Pedal forces and power were analyzed for each leg individually, and handlebar forces and seat post mediolateral sway were recorded during cycling. Normalized symmetry index (NSI), a metric quantifying the degree of asymmetry between limbs, was calculated for each variable. The reproducibility of the device was assessed using repeated measures analysis of variance and intraclass correlation coefficients (ICC). No significant session or interaction effects were found for pedal, handlebar, and seat post measures (all p > 0.05). Time effects were observed for pedal force and power in the incremental test (all p < 0.001). NSI values were reproducible with high ICC values (≥0.70) for force and power. The results suggest that this ergometer offers reproducible cycling kinetics and asymmetry measures across a fatiguing task. The findings support the application of this ergometer in research and rehabilitation settings.


  • Open Access

    Quantifying lower-limb muscle coordination during cycling using electromyography-informed muscle synergies

    by Reza Ahmadi, Shahram Rasoulian, Hamidreza Heidary, Saied Jalal Aboodarda, Thomas K. Uchida, Walter Herzog, Amin Komeili

    Assessment of muscle coordination during cycling may provide insight into motor control strategies and movement efficiency. This study evaluated muscle synergies and coactivation patterns as indicators of neuromuscular coordination in lower-limb across three power levels of cycling. Twenty recreational cyclists performed a graded cycling test on a stationary bicycle ergometer. Electromyography was recorded bilaterally from seven lower-limb muscles and muscle synergies were extracted using non-negative matrix factorization. The Coactivation Index (CI), Synergy Index (SI), and Synergy Coordination Index (SCI) were calculated to assess muscle coordination patterns. Four muscle synergies were identified consistently across power levels, with changes in synergy composition and activation timing correlated with increased muscular demands. As power level increased, the CI showed reduced muscle coactivation at the knee and greater muscle coactivation at the ankle. The SI revealed a greater contribution of the synergy weights of the extensor muscles than those of the flexor muscles at the knee. In contrast, the relative EMG contribution of hip extensor and flexor muscles remained consistent with increasing power levels. The SCI increased significantly with increasing power level, suggesting a reduction in the size of the synergy space and improved neuromuscular coordination. These findings provide insight into how the central nervous system modulates its response to increasing mechanical demands. Combining synergy and coactivation indices offers a promising approach to assess motor control, inform rehabilitation, and optimize performance in cycling tasks.


  • Open Access

    Comparison of kinematics and kinetics between OpenCap and a marker-based motion capture system in cycling

    by Reza Kakavand, Reza Ahmadi, Atousa Parsaei, W. Brent Edwards, Amin Komeili


  • Open Access

    A Machine Learning Approach for Predicting Pedaling Force Profile in Cycling

    by Reza Ahmadi, Shahram Rasoulian, Samira Fazeli Veisari, Atousa Parsaei, Hamidreza Heidary, Walter Herzog, Amin Komeili

    Accurate measurement of pedaling kinetics and kinematics is vital for optimizing rehabilitation, exercise training, and understanding musculoskeletal biomechanics. Pedal reaction force, the main external force in cycling, is essential for musculoskeletal modeling and closely correlates with lower-limb muscle activity and joint reaction forces. However, sensor instrumentation like 3-axis pedal force sensors is costly and requires extensive postprocessing. Recent advancements in machine learning (ML), particularly neural network (NN) models, provide promising solutions for kinetic analyses. In this study, an NN model was developed to predict radial and mediolateral forces, providing a low-cost solution to study pedaling biomechanics with stationary cycling ergometers. Fifteen healthy individuals performed a 2 min pedaling task at two different self-selected (58 ± 5 RPM) and higher (72 ± 7 RPM) cadences. Pedal forces were recorded using a 3-axis force system. The dataset included pedal force, crank angle, cadence, power, and participants’ weight and height. The NN model achieved an inter-subject normalized root mean square error (nRMSE) of 0.15 ± 0.02 and 0.26 ± 0.05 for radial and mediolateral forces at high cadence, respectively, and 0.20 ± 0.04 and 0.22 ± 0.04 at self-selected cadence. The NN model’s low computational time suits real-time pedal force predictions, matching the accuracy of previous ML algorithms for estimating ground reaction forces in gait.


  • Open Access

    A Machine Learning Approach for Predicting Pedaling Force Profile in Cycling

    by Reza Ahmadi, Shahram Rasoulian, Samira Fazeli Veisari, Atousa Parsaei, Hamidreza Heidary, Walter Herzog, Amin Komeili

    Accurate measurement of pedaling kinetics and kinematics is vital for optimizing rehabilitation, exercise training, and understanding musculoskeletal biomechanics. Pedal reaction force, the main external force in cycling, is essential for musculoskeletal modeling and closely correlates with lower limb muscle activity and joint reaction forces. However, sensor instrumentation like 3-axis pedal force sensors is costly and requires extensive post-processing. Recent advancements in machine learning (ML), particularly neural network (NN) models, provide promising solutions for kinetic analyses. In this study, an NN model was developed to predict radial and mediolateral forces, providing a low-cost solution to study pedaling biomechanics with stationary cycling ergometers. Fifteen healthy individuals performed a 2-minute pedaling task at two different self-selected (58±5 rpm) and higher (72±7 rpm) cadences. Pedal forces were recorded using a 3-axis force system. The dataset included pedal force, crank angle, cadence, power, and participants' weight and height. The NN model achieved an inter-subject normalized root mean square error (nRMSE) of 0.15±0.02 and 0.26±0.05 for radial and mediolateral forces at high cadence, respectively, and 0.20±0.04 and 0.22±0.04 at self-selected cadence. The NN model's low computational time suits real-time pedal force predictions, matching the accuracy of previous ML algorithms for estimating ground reaction forces in gait.


  • Effect of anterior cruciate ligament injury on acceleration response of knee joint

    by Reza Ahmadi, Seyed-Hossein Hosseini Lorgan, Arash Sherafat Vaziri, Mohammad Naghi Tahmasebi, Ramin Shayan Moghadam, Farzam Farahmand

    This study investigated the effect of anterior cruciate ligament (ACL) injury on relative acceleration of the tibia and femur during a number of tests/activities, in order to assess the feasibility of acceleration-based diagnosis of ACL injury using inertial sensors. First, a detailed finite element model of the knee joint was developed to simulate the target tests/activities, and identify those in which a large difference between the maximum acceleration peaks (MAPs) of the healthy and ACL injured knees is likely to be observed. The promising tests/activities were entered in an experimental study, where the relative accelerations of the tibiae and femurs of 20 individuals with unilateral ACL injury, allocated randomly to two groups of conscious and unconscious test conditions, were recorded. Model predictions indicated MAP ratios>1.5 for the ACL-injured to healthy knees, during the anterior drawer, Lachman, and pivot-shift tests, as well as the lunge activity. The experimental MAP results indicated acceptable test–retest reliabilities for all tests (coefficient of variation2 m/s2 for all subjects during unconscious pivot shift tests, and >0.5 m/s2 for eight cases out of ten during conscious anterior drawer tests. It was concluded that the pivot shift test had a great repeatability and discriminative ability for acceleration-based diagnosis of ACL injury in unconscious condition. For the conscious condition, however, the anterior drawer test was appeared to be most promising.