Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs
Abstract Medical image classification tasks frequently encounter challenges associated with class imbalance, resulting in biased model training and suboptimal classification performance. To address this issue, the combination of class decomposition and transfer learning has proven to be effective in...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-08-01
|
| Series: | Discover Artificial Intelligence |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44163-023-00078-0 |
| _version_ | 1797557799369048064 |
|---|---|
| author | Patryk Buczek Usama Zidan Mohamed Medhat Gaber Mohammed M. Abdelsamea |
| author_facet | Patryk Buczek Usama Zidan Mohamed Medhat Gaber Mohammed M. Abdelsamea |
| author_sort | Patryk Buczek |
| collection | DOAJ |
| description | Abstract Medical image classification tasks frequently encounter challenges associated with class imbalance, resulting in biased model training and suboptimal classification performance. To address this issue, the combination of class decomposition and transfer learning has proven to be effective in classifying imbalanced medical imaging datasets. Nevertheless, in order to further augment the performance gains achieved through the utilisation of class decomposition within deep learning frameworks, we propose a novel model coined imbalance-Aware Decomposition for Class-Decomposed Classification (iDeComp) model. By incorporating a conditional Generative Adversarial Network (GAN) model, iDeComp is capable of generating additional samples specifically tailored to underrepresented decomposed subclasses. This paper investigates the application of iDeComp using two different medical imaging datasets. iDeComp selects underrepresented samples from the training set of the sublevel classes within each dataset, which are then employed to train separate conditional Deep Convolutional GAN (DCGAN) models and verification models. The conditional DCGAN model is responsible for generating additional samples, while the verification model critically evaluates the appropriateness of the synthesised images. Subsequently, the resulting augmented samples are utilized to train the classification model. To assess the effectiveness of iDeComp, we employ various evaluation metrics including accuracy, precision, recall, and F1 score. The results obtained from our experiments clearly indicate that iDeComp outperforms existing approaches in terms of classifying both imbalanced datasets. |
| first_indexed | 2024-03-10T17:21:18Z |
| format | Article |
| id | doaj.art-5162eda6ee064e4d9a21e8702c0b364b |
| institution | Directory Open Access Journal |
| issn | 2731-0809 |
| language | English |
| last_indexed | 2024-03-10T17:21:18Z |
| publishDate | 2023-08-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Artificial Intelligence |
| spelling | doaj.art-5162eda6ee064e4d9a21e8702c0b364b2023-11-20T10:18:54ZengSpringerDiscover Artificial Intelligence2731-08092023-08-013111410.1007/s44163-023-00078-0Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANsPatryk Buczek0Usama Zidan1Mohamed Medhat Gaber2Mohammed M. Abdelsamea3School of Computing and Digital Technology, Birmingham City UniversitySchool of Computing and Digital Technology, Birmingham City UniversitySchool of Computing and Digital Technology, Birmingham City UniversitySchool of Computing and Digital Technology, Birmingham City UniversityAbstract Medical image classification tasks frequently encounter challenges associated with class imbalance, resulting in biased model training and suboptimal classification performance. To address this issue, the combination of class decomposition and transfer learning has proven to be effective in classifying imbalanced medical imaging datasets. Nevertheless, in order to further augment the performance gains achieved through the utilisation of class decomposition within deep learning frameworks, we propose a novel model coined imbalance-Aware Decomposition for Class-Decomposed Classification (iDeComp) model. By incorporating a conditional Generative Adversarial Network (GAN) model, iDeComp is capable of generating additional samples specifically tailored to underrepresented decomposed subclasses. This paper investigates the application of iDeComp using two different medical imaging datasets. iDeComp selects underrepresented samples from the training set of the sublevel classes within each dataset, which are then employed to train separate conditional Deep Convolutional GAN (DCGAN) models and verification models. The conditional DCGAN model is responsible for generating additional samples, while the verification model critically evaluates the appropriateness of the synthesised images. Subsequently, the resulting augmented samples are utilized to train the classification model. To assess the effectiveness of iDeComp, we employ various evaluation metrics including accuracy, precision, recall, and F1 score. The results obtained from our experiments clearly indicate that iDeComp outperforms existing approaches in terms of classifying both imbalanced datasets.https://doi.org/10.1007/s44163-023-00078-0Class imbalanceConditional GANTransfer learningMedical image classification |
| spellingShingle | Patryk Buczek Usama Zidan Mohamed Medhat Gaber Mohammed M. Abdelsamea Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs Discover Artificial Intelligence Class imbalance Conditional GAN Transfer learning Medical image classification |
| title | Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs |
| title_full | Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs |
| title_fullStr | Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs |
| title_full_unstemmed | Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs |
| title_short | Idecomp: imbalance-aware decomposition for class-decomposed classification using conditional GANs |
| title_sort | idecomp imbalance aware decomposition for class decomposed classification using conditional gans |
| topic | Class imbalance Conditional GAN Transfer learning Medical image classification |
| url | https://doi.org/10.1007/s44163-023-00078-0 |
| work_keys_str_mv | AT patrykbuczek idecompimbalanceawaredecompositionforclassdecomposedclassificationusingconditionalgans AT usamazidan idecompimbalanceawaredecompositionforclassdecomposedclassificationusingconditionalgans AT mohamedmedhatgaber idecompimbalanceawaredecompositionforclassdecomposedclassificationusingconditionalgans AT mohammedmabdelsamea idecompimbalanceawaredecompositionforclassdecomposedclassificationusingconditionalgans |