Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques
Stress is a major part of our everyday life, associated with most activities we perform on a daily basis and if we are not careful about managing stress, it can have a detrimental impact on our health. Despite recent advances in this domain, HRV analysis is still the most common method to detect str...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-12-01
|
| Series: | Machine Learning with Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666827022000755 |
| _version_ | 1811293668369235968 |
|---|---|
| author | Syem Ishaque Naimul Khan Sri Krishnan |
| author_facet | Syem Ishaque Naimul Khan Sri Krishnan |
| author_sort | Syem Ishaque |
| collection | DOAJ |
| description | Stress is a major part of our everyday life, associated with most activities we perform on a daily basis and if we are not careful about managing stress, it can have a detrimental impact on our health. Despite recent advances in this domain, HRV analysis is still the most common method to detect stress, and although the results that have been produced are admirable, feature extraction is complicated and time consuming. We propose an algorithm to convert 1D (dimensional) ECG data from WESAD (wearable stress and affect detection dataset) into 2D ECG images, which are representative of stress/not stress. It does not require time consuming processes such as feature extraction and filtering. We utilize transfer learning to obtain competitive results. We also demonstrate that model compression techniques can significantly reduce the computational size of the algorithms, without sacrificing much of the performance, as evident from a classification accuracy of 90.62% using the quantization technique. Results substantiate the effectiveness of our proposed method and empirically demonstrates the potential of deep learning algorithms for edge computing and mobile applications, which utilizes low performing hardware. |
| first_indexed | 2024-04-13T05:04:54Z |
| format | Article |
| id | doaj.art-dcc6fa7edc7b4a4bbb8e378518a3d2f2 |
| institution | Directory Open Access Journal |
| issn | 2666-8270 |
| language | English |
| last_indexed | 2024-04-13T05:04:54Z |
| publishDate | 2022-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Machine Learning with Applications |
| spelling | doaj.art-dcc6fa7edc7b4a4bbb8e378518a3d2f22022-12-22T03:01:12ZengElsevierMachine Learning with Applications2666-82702022-12-0110100395Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniquesSyem Ishaque0Naimul Khan1Sri Krishnan2Corresponding author.; Department of Electrical, Computer and Biomedical Engineering, Ryerson University, Toronto, CanadaDepartment of Electrical, Computer and Biomedical Engineering, Ryerson University, Toronto, CanadaDepartment of Electrical, Computer and Biomedical Engineering, Ryerson University, Toronto, CanadaStress is a major part of our everyday life, associated with most activities we perform on a daily basis and if we are not careful about managing stress, it can have a detrimental impact on our health. Despite recent advances in this domain, HRV analysis is still the most common method to detect stress, and although the results that have been produced are admirable, feature extraction is complicated and time consuming. We propose an algorithm to convert 1D (dimensional) ECG data from WESAD (wearable stress and affect detection dataset) into 2D ECG images, which are representative of stress/not stress. It does not require time consuming processes such as feature extraction and filtering. We utilize transfer learning to obtain competitive results. We also demonstrate that model compression techniques can significantly reduce the computational size of the algorithms, without sacrificing much of the performance, as evident from a classification accuracy of 90.62% using the quantization technique. Results substantiate the effectiveness of our proposed method and empirically demonstrates the potential of deep learning algorithms for edge computing and mobile applications, which utilizes low performing hardware.http://www.sciencedirect.com/science/article/pii/S26668270220007552D stress imagesTransfer learning2D ECGComputer vision |
| spellingShingle | Syem Ishaque Naimul Khan Sri Krishnan Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques Machine Learning with Applications 2D stress images Transfer learning 2D ECG Computer vision |
| title | Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques |
| title_full | Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques |
| title_fullStr | Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques |
| title_full_unstemmed | Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques |
| title_short | Detecting stress through 2D ECG images using pretrained models, transfer learning and model compression techniques |
| title_sort | detecting stress through 2d ecg images using pretrained models transfer learning and model compression techniques |
| topic | 2D stress images Transfer learning 2D ECG Computer vision |
| url | http://www.sciencedirect.com/science/article/pii/S2666827022000755 |
| work_keys_str_mv | AT syemishaque detectingstressthrough2decgimagesusingpretrainedmodelstransferlearningandmodelcompressiontechniques AT naimulkhan detectingstressthrough2decgimagesusingpretrainedmodelstransferlearningandmodelcompressiontechniques AT srikrishnan detectingstressthrough2decgimagesusingpretrainedmodelstransferlearningandmodelcompressiontechniques |