Skip to content

josefigarola/EMG_Classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

54 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Enhancing EMG-Based Recognition of Fine Hand Movements with Incremental Learning and Warm Start

NeuroTechnology Laboratory Tecnologico de Monterrey

Research Paper This project retrieved Electromyographic signals or EMG signals from 6 different hand movements. In order to obtain the best results, 9 different classifier were used to train the model. To increase the accuracy of the model, Warm Start and Partial Fit are used to readjust the model's performance.

Each classifier uses 5 K-Folds, Cross validation and 36 time-domain features. These classifiers are: LDA, SVM, MLP and NNET. Some classifiers have different hyperparameters to obtain the 9 mentioned.

The dataset consists of 30 subjects, and 240 trials.

Studies

  • Single Subject Independent (SSI)
    • This studie trains the model using just one subject from the 30.
  • Leave One Out Participant (LOOP)
    • This studies trains the model using 29 subjects from the dataset and tests with the remaining subject.
  • Warm Start and Partial Fit
    • Readjusting the model's performance by using WS and PF
  • ANOVA and Tukey's HSD
    • Obtain models performances' statistics.

Scripts

  • main.ipynb
    • This script is the source of the results, it obtains the models of the training phase and tests.
  • ANOVA_TUKEY.ipynb
    • This script computes the statistical tests from the classifiers
  • PFvsWSInstro.ipynb
    • To comprehend how partial fit and warm start is computed using classifiers
  • LOOP_OL_Tables.ipynb, SSI_OL_Tables.ipynb and WSvsPF_OL_Tables.ipynb
    • Create overleaf LATEX tables of the results.
  • SSInLOOP.ipynb
    • Test the studies SSI and LOOP
  • Matlab/src/main2.mlx
    • First approach to EMG data analysis and classifiers.

Team

Jose Miguel Figarola Prado - [email protected] - [email protected]

Yelile Iga Valdes - [email protected]

Victor Andrés Gonzalez Saldaña - [email protected]

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published