A Grad-CAM-based knowledge distillation method for the detection of tuberculosis
Automatic screening for tuberculosis (TB) from X-ray images using artificial intelligence techniques has attracted the attention of researchers in the fields of computing and medicine. However, existing models are computationally intensive and require high computer hardware, which limits the use of...
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
IEEE
2023
|
| _version_ | 1825949068923240448 |
|---|---|
| author | Ding, Zeyu Yaakob, Razali Azman, Azreen Mohd Rum, Siti Nurulain Zakaria, Norfadhlina Ahmad Nazri, Azree Shahril |
| author_facet | Ding, Zeyu Yaakob, Razali Azman, Azreen Mohd Rum, Siti Nurulain Zakaria, Norfadhlina Ahmad Nazri, Azree Shahril |
| author_sort | Ding, Zeyu |
| collection | UPM |
| description | Automatic screening for tuberculosis (TB) from X-ray images using artificial intelligence techniques has attracted the attention of researchers in the fields of computing and medicine. However, existing models are computationally intensive and require high computer hardware, which limits the use of people in areas where medical resources are scarce. Another problem with the existing model is poor interpretability. The model only provides the final result and lacks intuitive information about the location of the lesion. To solve these problems, this paper proposes a Grad-CAM-based knowledge distillation method for the detection of TB. Firstly, this study used Unet to extract the lung region, avoiding the influence of regions outside the lung on the detection results. Subsequently, five models (Densenet121, Inception V3, Resnet18, Mobilenet V3, VGG16) are applied to TB detection, and the attention maps of each model are visualized using Grad-CAM. These attention maps are applied to knowledge distillation to finally obtain a lightweight interpretable TB detection model. This model achieves 91.2% and 85.7% accuracy on Shenzhen and Montgomery datasets, which verifies the effectiveness of the model. |
| first_indexed | 2024-03-06T08:38:53Z |
| format | Conference or Workshop Item |
| id | upm.eprints-37619 |
| institution | Universiti Putra Malaysia |
| last_indexed | 2024-03-06T08:38:53Z |
| publishDate | 2023 |
| publisher | IEEE |
| record_format | dspace |
| spelling | upm.eprints-376192023-09-28T05:02:53Z http://psasir.upm.edu.my/id/eprint/37619/ A Grad-CAM-based knowledge distillation method for the detection of tuberculosis Ding, Zeyu Yaakob, Razali Azman, Azreen Mohd Rum, Siti Nurulain Zakaria, Norfadhlina Ahmad Nazri, Azree Shahril Automatic screening for tuberculosis (TB) from X-ray images using artificial intelligence techniques has attracted the attention of researchers in the fields of computing and medicine. However, existing models are computationally intensive and require high computer hardware, which limits the use of people in areas where medical resources are scarce. Another problem with the existing model is poor interpretability. The model only provides the final result and lacks intuitive information about the location of the lesion. To solve these problems, this paper proposes a Grad-CAM-based knowledge distillation method for the detection of TB. Firstly, this study used Unet to extract the lung region, avoiding the influence of regions outside the lung on the detection results. Subsequently, five models (Densenet121, Inception V3, Resnet18, Mobilenet V3, VGG16) are applied to TB detection, and the attention maps of each model are visualized using Grad-CAM. These attention maps are applied to knowledge distillation to finally obtain a lightweight interpretable TB detection model. This model achieves 91.2% and 85.7% accuracy on Shenzhen and Montgomery datasets, which verifies the effectiveness of the model. IEEE 2023 Conference or Workshop Item PeerReviewed Ding, Zeyu and Yaakob, Razali and Azman, Azreen and Mohd Rum, Siti Nurulain and Zakaria, Norfadhlina and Ahmad Nazri, Azree Shahril (2023) A Grad-CAM-based knowledge distillation method for the detection of tuberculosis. In: 2023 9th International Conference on Information Management (ICIM 2023), 17-19 Mar. 2023, Oxford, United Kingdom. (pp. 72-77). https://ieeexplore.ieee.org/document/10145170 10.1109/ICIM58774.2023.00019 |
| spellingShingle | Ding, Zeyu Yaakob, Razali Azman, Azreen Mohd Rum, Siti Nurulain Zakaria, Norfadhlina Ahmad Nazri, Azree Shahril A Grad-CAM-based knowledge distillation method for the detection of tuberculosis |
| title | A Grad-CAM-based knowledge distillation method for the detection of tuberculosis |
| title_full | A Grad-CAM-based knowledge distillation method for the detection of tuberculosis |
| title_fullStr | A Grad-CAM-based knowledge distillation method for the detection of tuberculosis |
| title_full_unstemmed | A Grad-CAM-based knowledge distillation method for the detection of tuberculosis |
| title_short | A Grad-CAM-based knowledge distillation method for the detection of tuberculosis |
| title_sort | grad cam based knowledge distillation method for the detection of tuberculosis |
| work_keys_str_mv | AT dingzeyu agradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT yaakobrazali agradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT azmanazreen agradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT mohdrumsitinurulain agradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT zakarianorfadhlina agradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT ahmadnazriazreeshahril agradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT dingzeyu gradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT yaakobrazali gradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT azmanazreen gradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT mohdrumsitinurulain gradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT zakarianorfadhlina gradcambasedknowledgedistillationmethodforthedetectionoftuberculosis AT ahmadnazriazreeshahril gradcambasedknowledgedistillationmethodforthedetectionoftuberculosis |