| Title: | On time effectiveness of manual editing of motor unit spike trains |
|---|
| Authors: | ID Murks, Nina, Faculty of Electrical Engineering and Computer Science, University of Maribor, Maribor, Slovenia (Author)
ID Škarabot, Jakob, School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough, UK (Author)
ID Kramberger, Matej, Faculty of Electrical Engineering and Computer Science, University of Maribor, Maribor, Slovenia (Author)
ID Sedej, Gašper, Faculty of Electrical Engineering and Computer Science, University of Maribor, Maribor, Slovenia (Author)
ID Valenčič, Tamara, School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough, UK (Author)
ID Thomason, Haydn, School of Sport, Exercise and Health Sciences, Loughborough University, Loughborough, UK (Author)
ID Divjak, Matjaž, Faculty of Electrical Engineering and Computer Science, University of Maribor, Maribor, Slovenia (Author)
ID Holobar, Aleš, Faculty of Electrical Engineering and Computer Science, University of Maribor, Maribor, Slovenia (Author) |
|---|
| Files: |  ISEK_2024_Murks.pdf (265,22 KB)
MD5: 0C6884671D2DA253BE3E4D59911BE527
|
|---|
| Language: | English |
|---|
| Work type: | Scientific work |
|---|
| Typology: | 1.12 - Published Scientific Conference Contribution Abstract |
|---|
| Organization: | FERI - Faculty of Electrical Engineering and Computer Science
|
|---|
| Abstract: | BACKGROUND AND AIM: Automatic methods for motor unit (MU) spike train identification from HDEMG are extensively used, but segmentation of the MU spike trains into discharge patterns still requires manual editing. This represents one of the major bottlenecks in analysis. We explored how the time efficiency of manual editing depends on the quality of the identified spike trains (Pulse-to-Noise Ratio - PNR), the operator’s level of experience, and muscle contraction levels.
METHODS: Experimental signals were acquired from the First Dorsal Interosseous, Tibialis Anterior, Vastus Lateralis, and Biceps Brachii (BB). Two male subjects performed isometric contractions for every muscle at 10, 30, 50, and 70% of MVC. The contractions were ~25 seconds long and were measured with a 13×5 electrode array. The same contraction levels were simulated in Soleus and BB with a 9×10 electrode array and 20 dB noise. All signals were decomposed using the Convolution-Kernel-Compensation method and manually edited by 9 operators (2 beginners with < 50, 2 intermediates with < 200, 3 advanced with < 1000,
and 2 experts with > 1000 edited signals worth of experience). All operators underwent a tutorial to standardize the editing procedure. Results were analyzed with a linear mixed- effects model of editing_time ~ PNR * contraction_level * operator’s _experience * signal_type + (PNR | muscle:subject). We only examined MUs with a PNR of 25 dB or higher.
RESULTS: The editing time decreased with PNR (F=63.9, P<0.0001) with 55.9 ± 80.3 s and 8.1 ± 15 s spent for editing of MUs with PNR of 25-39 dB and 40-54 dB, respectively. Editing time increased with the contraction level (F=10.9, P<0.0001) as 42.6 ± 63.2 s, 54.2 ± 82 s, 57.8 ± 83.8 s, and 60 ± 82.9 s were spent editing the 10, 20, 30, 50 and 70% contraction level. The level of experience predicted the editing time (F=10.5, P<0.0001). Beginners required 65.9 ±
90.4 s, intermediates 55.1 ± 72.4 s, advanced operators 51.9 ± 90.9 s and experts 37.1 ± 40.6 s. Signal type (synthetic or experimental) did not influence editing time but was in significant interaction with contraction level (F=6.4, P=0.001), and operator’s level of experience (F=4.7, P=0.003). There was also a significant interaction between PNR and contraction level (F=10.4, P<0.0001), PNR and operator’s level of experience (F=7.1, P<0.0001), and contraction level
and operator’s level of experience (F=3.3, P=0.001).
CONCLUSION: The time required for MU editing reduces almost linearly with the operator’s level of experience and is dependent on the contraction level and PNR.
FUNDING: This research was funded by the European Union’s Horizon Europe Research and Innovation Program [HybridNeuro project, GA No. 101079392]. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or Research Executive Agency. Neither the European Union nor the granting authority can be held responsible for them. |
|---|
| Keywords: | high-density surface electromyography, motor unit, spike train, convolution kernel compensation |
|---|
| Publication status: | Not published |
|---|
| Publication version: | Preprint, working version of publication, not peer-reviewed |
|---|
| Year of publishing: | 2024 |
|---|
| Number of pages: | 2 |
|---|
| PID: | 20.500.12556/DKUM-97208  |
|---|
| UDC: | 004.8:61 |
|---|
| COBISS.SI-ID: | 204433667 |
|---|
| Publication date in DKUM: | 26.02.2026 |
|---|
| Views: | 147 |
|---|
| Downloads: | 6 |
|---|
| Metadata: |    |
|---|
| Categories: | Misc.
|
|---|
|
:
|
Copy citation |
|---|
| | | | Average score: | (0 votes) |
|---|
| Your score: | Voting is allowed only for logged in users. |
|---|
| Share: |  |
|---|
Hover the mouse pointer over a document title to show the abstract or click
on the title to get all document metadata. |
