Correlation between quadriceps activation and proprioception in patients with knee osteoarthritis
Data files
Jan 24, 2026 version files 1.90 GB
-
full_indicators_summary.csv
12.65 KB
-
full_sEMG_data_in_correct_format.zip
1.90 GB
-
README.md
8.63 KB
Abstract
Objective: To investigate the correlation between the activation status of quadriceps muscle and the proprioceptive accuracy in patients with knee osteoarthritis (KOA).
Methods: We included 43 outpatients diagnosed with KOA and 45 health subjects from October 2024 to May 2025. The correlation between the absolute error in knee Joint Position Reproduction (JPR) and the root mean square (RMS) and median frequency (MDF) of quadriceps surface electromyography during walking and stair climbing was analyzed.
Results: The JPR of 45° in KOA patient is significantly higher than health subjects. There was no significant difference in 30° and 60° JRP between the two groups. While walking, there is no significant difference of the sEMG data in two group. While going up stair, the RMS of LVM and RVM in KOA group had increased activation compared with the health subjects. In KOA group, while going up stair, the JPR absolute error angle of 30° was negatively correlated with the MDF of RVL, the JPR absolute error angle of 45° was positively correlated with the RMS of RVM, and the JPR absolute error angle of 60° was positively correlated with RMS of LVM and RVM. There are no significantly correlation in JPR absolute error angle and the surface electromyography of the quadriceps muscle while walking and in health control group. Partial correlation analysis was performed on the aforementioned correlated indicators with gender included as a control variable, and the results indicated that in women group, the JPR absolute error angle of 60° was positively correlated with RMS of LVM and RVM. In KOA group, the RMS of LVL, RVL and RVM in walking was positively correlated with the RMS of LVL, RVL and RVM in going up and down stair, and the MDF of LVL, RVL and RVM in walking was positively correlated with the MDF of LVL, RVL and RVM in going up and down stair. In health control group, the MDF of LVL, RVL and RVM in walking was positively correlated with the MDF of LVL, RVL and RVM in going up and down stair.
Conclusion: This study indicated that during stair climb, KOA patients display a unique pattern of elevated RMS of the VM. Notable relationships were identified between the absolute error angles of JPR and sEMG parameters of VM. The pronounced positive correlations in muscle activation patterns between walking and stair upward and downward activities indicate a fixed, task-invariant neuromuscular approach in KOA patients, possibly resulting from compromised sensorimotor integration due to proprioceptive impairments. These findings offer new insights into the pathogenesis of KOA from a neuromuscular-sensory viewpoint, emphasizing the need for future rehabilitation therapies to concurrently target proprioceptive accuracy and the enhancement of muscle activation patterns.
Dataset DOI: 10.5061/dryad.fn2z34v8s
Description of the data and file structure
Surface Electromyography(sEMG) Data of the Quadriceps Muscle
The 8-channel distributed wireless electromyography acquisition system (Shenzhen Runyi Taiyi Technology Co., Ltd.) was utilized for data collection, with a sampling frequency of 2000 Hz. Subjects were prepared by cleaning the skin of both lower limbs with 75% alcohol, and any hair in the areas of electrode placement was shaved. Surface electrodes were affixed to the following muscles in both lower limbs: the vastus lateralis (VL), located at the highest point of the muscle belly in the lower third of the line connecting the lateral edge of the patella and the anterior superior iliac spine; and the VM, placed at the midpoint of the line connecting the medial edge of the patella and the anterior superior iliac spine. The electrodes were positioned on the most prominent part of each muscle belly, aligned with the direction of the muscle fibers. Subjects were instructed to remain barefoot, and the testing procedure was thoroughly explained to them prior to the commencement of the test to ensure their understanding and comfort.
Maximum Voluntary Contraction (MVC) Standardization: Participants were instructed to perform bodyweight squats to their maximum comfortable depth under standardized conditions. A three-dimensional motion capture gait system (Opti_Knee, Innomotion Inc., Shanghai, China) was used to determined the longitudinal axis of the femur and the longitudinal axis of the tibia create a 90-degree angle, the body inclines forward, and the center of gravity is situated at the base of the thumb on the anterior foot. The subjects maintained the squat position for 3 seconds, then stood up straight[15,16]. This process was repeated three times to determine the MVC sEMG.
Treadmill Walking: Participants stood still on the treadmill and adjusted the speed between 1 km/h and 3 km/h according to their habitual walking speeds. After 10 s of stable walking, the sEMG data of quadriceps and knee joint velocity data was collected for about 15s (15 walking gait cycles) for both knee joints. A synchronizer was employed to simultaneously collect sEMG data alongside motion capture gait system data.
Walking Up and Down Steps: Subjects stood naturally on the ground with their legs relaxed. Upon hearing the walking command, the subject began to ascend the steps, first lifting the right thigh, followed by the left thigh, and then the right thigh again, taking one step at a time for a total of three steps. After reaching the third step, the subject remained still for one second. Upon hearing the walking command again, the subject then descended the steps, first lowering the right thigh, followed by the left thigh, and finally standing on a flat surface. It is important to maintain an upright trunk throughout the process of ascending and descending the steps. Collect the data of going up and down the stairs three times.
Files and variables
File: full_indicators_summary.csv
Description: The comprehensive data summary of all the subjects collected.
Variables
number: the id of subject
group:0 represents health subject,1 represents KOA subject
age:age ranges,40 represents 40-49 years old;50 represents 50-59 years old;60 represents 60-69 years old;70 represents 70-79 years old;80 represents 80-89 years old
gender: 0 represents female,1 represents male
BMI: Weight (kg) / [Height (m)²]
K-L grade: Kellgren-Lawrence grade
KOA period: Clinical stages of knee osteoarthritis
VAS score: Visual Analogue Scale
WOMAC score: Western Ontario and McMaster Universities Osteoarthritis Index
30°JPR:Absolute error angle of knee joint position reproduction in 30°(average value of the three measurements)
45°JPR:Absolute error angle of knee joint position reproduction in 45°(average value of the three measurements)
60°JPR:Absolute error angle of knee joint position reproduction in 60°(average value of the three measurements)
LW_LVM_RMS%:the RMS of left vastus medialis in walk test(Normalized with MVC data)
LS_LVM_RMS%:the RMS of left vastus medialis in up-down stair(Normalized with MVC data)
LW_LVL_RMS%:the RMS of left vastus lateralis in walk test(Normalized with MVC data)
LS_LVL_RMS%:the RMS of left vastus lateralis in up-down stair(Normalized with MVC data)
RW_RVM_RMS%:the RMS of right vastus medialis in walk test(Normalized with MVC data)
RS_RVM_RMS%:the RMS of right vastus medialis in up-down stair(Normalized with MVC data)
RW_RVL_RMS%:the RMS of right vastus medialis in walk test(Normalized with MVC data)
RS_RVL_RMS%:the RMS of right vastus medialis in up-down stair(Normalized with MVC data)
LW-LVM-MDF: the MDF of left vastus medialis in walk test
LW-LVL-MDF: the MDF of left vastus lateralis in walk test
LS-LVM-MDF: the MDF of left vastus medialis in up-down stair
LS-LVL-MDF: the MDF of left vastus lateralis in up-down stair
RW-RVM-MDF: the MDF of right vastus medialis in walk test
RW-RVL-MDF: the MDF of right vastus medialis in walk test
RS-RVM-MDF: the MDF of right vastus medialis in up-down stair
RS-RVL-MDF: the MDF of right vastus medialis in up-down stair
File: full_sEMG_data_in_correct_format.zip
Description: The compressed file contains all surface electromyography data, organized into three folders corresponding to the maximum voluntary contraction (MVC) test, walking test, and up-down stair test. Each folder includes 88 Excel files, with each file representing the data collected from an individual subject. The files are labeled according to the respective subject ID.
Variables
Variables definition for the excel file in the compressed file, take the Excel file "walk data_ (1).csv" as an example
EMG_IMU_1_EMG_CH1:the sEMG data of left vastus medialis
EMG_IMU_1_EMG_CH2:the sEMG data of left vastus medialis (this chennel was used to calculated and analysised)
EMG_IMU_2_EMG_CH1:the sEMG data of left rectus femoris
EMG_IMU_2_EMG_CH2:the sEMG data of left rectus femoris (this chennel was used to calculated and analysised)
EMG_IMU_3_EMG_CH1:the sEMG data of left vastus lateralis
EMG_IMU_3_EMG_CH2:the sEMG data of left vastus lateralis (this chennel was used to calculated and analysised)
EMG_IMU_4_EMG_CH1:the sEMG data of left biceps femoris
EMG_IMU_4_EMG_CH2:the sEMG data of left biceps femoris
EMG_IMU_5_EMG_CH1:the sEMG data of left biceps femoris
EMG_IMU_5_EMG_CH2:the sEMG data of left biceps femoris
EMG_IMU_9_EMG_CH1:the sEMG data of right vastus medialis
EMG_IMU_9_EMG_CH2:the sEMG data of right vastus medialis (this chennel was used to calculated and analysised)
EMG_IMU_10_EMG_CH1:the sEMG data of right rectus femoris
EMG_IMU_10_EMG_CH2:the sEMG data of right rectus femoris
EMG_IMU_11_EMG_CH1:the sEMG data of right vastus lateralis
EMG_IMU_11_EMG_CH2:the sEMG data of right vastus lateralis (this chennel was used to calculated and analysised)
EMG_IMU_12_EMG_CH1:the sEMG data of right biceps femoris
EMG_IMU_12_EMG_CH2:the sEMG data of right biceps femoris
EMG_IMU_13_EMG_CH1:the sEMG data of right biceps femoris
EMG_IMU_13_EMG_CH2:the sEMG data of right biceps femoris
Code/software
MATLAB(R2024a) analysis software was employed to process the original sEMG signals from the medial femoris and lateral femoris. The original electromyography signal is subjected to denoising via a 20-450 Hz bandpass Butterworth filter, followed by rectification and smoothing for preprocessing. In the task of ascending and descending stairs, the limb that initiates the movement on the stairs is defined as the leading phase, while the limb that follows subsequently is referred to as the trailing phase. MATLAB analysis software was then employed to derive the amplitude values for all exercise cycles, determine the time range of each signal cycle based on the raw signals, and identify the Root Mean Square (RMS) and the Median Frequency (MDF) for each muscle within the Support phase of the gait cycle (0–60%) and leading phase. Additionally, the RMS were normalized [RMS(walking or up stair)/RMS(MVC)] using data from the RVC test.
Access information
Other publicly accessible locations of the data:
- None
Data was derived from the following sources:
- None
Human subjects data
All the participants included have signed the informed consent form.
