CLASSIFICATION OF CHEST X-RAY IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORKS
Апстракт
In this paper a comparison between three different
types of trained VGG convolutional neural networks (CNNs) is
proposed for the classification of a pediatric chest X-ray image
data set. A deep convolutional neural network with an
architecture resembling the VGGNet is presented using dropout,
decay and data scaling. Since the dataset had a class imbalance,
this was solved using a simple method called data scaling. The
training of the neural network was done using small batches with
a binary cross entropy loss function. The same neural network
was then implemented adding batch normalization layers, and
comparisons were made. Furthermore, the chest X-ray dataset
was also trained using transfer learning with a pre-trained
neural network VGG16 on the ImageNet dataset. Later on
juxtapositions were made on using both techniques. Additionally,
in applying these methods we were able to achieve a classification
with the accuracy higher than 0.95 and 0.97 for the training and
val...idation datasets, whilst incorporating only 30 epochs.
Кључне речи:
convolutional neural networks / deep learning / transfer learning / batch normalization / chest X-ray dataset / classification / dropoutИзвор:
Зборник радова ‐ 64. Конференција за електронику, телекомуникације, рачунарство, аутоматику и нуклеарну технику / Proceedings of Papers – 7th International Conference on Electrical, Electronic and Computing Engineering IcETRAN 2020, 2020, SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 18-SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 23Издавач:
- Belgrade : Društvo za ETRAN
- Beograd : Akademska misao
Финансирање / пројекти:
- MISSION4.0 - Deep Machine Learning and Swarm Intelligence-Based Optimization Algorithms for Control and Scheduling of Cyber-Physical Systems in Industry 4.0 (RS-ScienceFundRS-AI-6523109)
- Министарство науке, технолошког развоја и иновација Републике Србије, институционално финансирање - 200105 (Универзитет у Београду, Машински факултет) (RS-MESTD-inst-2020-200105)
- Развој методологија за повећање радне способности, поузданости и енергетске ефикасности машинских система у енергетици (RS-MESTD-Technological Development (TD or TR)-35029)
Колекције
Институција/група
Mašinski fakultetTY - CONF AU - Laban, Lara AU - Jovanović, Radiša AU - Vesović, Mitra AU - Zarić, Vladimir PY - 2020 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4496 AB - In this paper a comparison between three different types of trained VGG convolutional neural networks (CNNs) is proposed for the classification of a pediatric chest X-ray image data set. A deep convolutional neural network with an architecture resembling the VGGNet is presented using dropout, decay and data scaling. Since the dataset had a class imbalance, this was solved using a simple method called data scaling. The training of the neural network was done using small batches with a binary cross entropy loss function. The same neural network was then implemented adding batch normalization layers, and comparisons were made. Furthermore, the chest X-ray dataset was also trained using transfer learning with a pre-trained neural network VGG16 on the ImageNet dataset. Later on juxtapositions were made on using both techniques. Additionally, in applying these methods we were able to achieve a classification with the accuracy higher than 0.95 and 0.97 for the training and validation datasets, whilst incorporating only 30 epochs. PB - Belgrade : Društvo za ETRAN PB - Beograd : Akademska misao C3 - Зборник радова ‐ 64. Конференција за електронику, телекомуникације, рачунарство, аутоматику и нуклеарну технику / Proceedings of Papers – 7th International Conference on Electrical, Electronic and Computing Engineering IcETRAN 2020 T1 - CLASSIFICATION OF CHEST X-RAY IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORKS EP - SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 23 SP - SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 18 UR - https://hdl.handle.net/21.15107/rcub_machinery_4496 ER -
@conference{ author = "Laban, Lara and Jovanović, Radiša and Vesović, Mitra and Zarić, Vladimir", year = "2020", abstract = "In this paper a comparison between three different types of trained VGG convolutional neural networks (CNNs) is proposed for the classification of a pediatric chest X-ray image data set. A deep convolutional neural network with an architecture resembling the VGGNet is presented using dropout, decay and data scaling. Since the dataset had a class imbalance, this was solved using a simple method called data scaling. The training of the neural network was done using small batches with a binary cross entropy loss function. The same neural network was then implemented adding batch normalization layers, and comparisons were made. Furthermore, the chest X-ray dataset was also trained using transfer learning with a pre-trained neural network VGG16 on the ImageNet dataset. Later on juxtapositions were made on using both techniques. Additionally, in applying these methods we were able to achieve a classification with the accuracy higher than 0.95 and 0.97 for the training and validation datasets, whilst incorporating only 30 epochs.", publisher = "Belgrade : Društvo za ETRAN, Beograd : Akademska misao", journal = "Зборник радова ‐ 64. Конференција за електронику, телекомуникације, рачунарство, аутоматику и нуклеарну технику / Proceedings of Papers – 7th International Conference on Electrical, Electronic and Computing Engineering IcETRAN 2020", title = "CLASSIFICATION OF CHEST X-RAY IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORKS", pages = "SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 23-SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 18", url = "https://hdl.handle.net/21.15107/rcub_machinery_4496" }
Laban, L., Jovanović, R., Vesović, M.,& Zarić, V.. (2020). CLASSIFICATION OF CHEST X-RAY IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORKS. in Зборник радова ‐ 64. Конференција за електронику, телекомуникације, рачунарство, аутоматику и нуклеарну технику / Proceedings of Papers – 7th International Conference on Electrical, Electronic and Computing Engineering IcETRAN 2020 Belgrade : Društvo za ETRAN., SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 18-SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 23. https://hdl.handle.net/21.15107/rcub_machinery_4496
Laban L, Jovanović R, Vesović M, Zarić V. CLASSIFICATION OF CHEST X-RAY IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORKS. in Зборник радова ‐ 64. Конференција за електронику, телекомуникације, рачунарство, аутоматику и нуклеарну технику / Proceedings of Papers – 7th International Conference on Electrical, Electronic and Computing Engineering IcETRAN 2020. 2020;:SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 18-SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 23. https://hdl.handle.net/21.15107/rcub_machinery_4496 .
Laban, Lara, Jovanović, Radiša, Vesović, Mitra, Zarić, Vladimir, "CLASSIFICATION OF CHEST X-RAY IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORKS" in Зборник радова ‐ 64. Конференција за електронику, телекомуникације, рачунарство, аутоматику и нуклеарну технику / Proceedings of Papers – 7th International Conference on Electrical, Electronic and Computing Engineering IcETRAN 2020 (2020):SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 18-SESSION AII2: ARTIFICIAL INTELIGENCE: DEEP NEURAL NETWORKS AND APPLICATION pp. 23, https://hdl.handle.net/21.15107/rcub_machinery_4496 .