A Modified Deep Convolutional Network for COVID-19 detection based on chest X-ray images

Fian Yulio Santoso, Hindriyanto Dwi Purnomo

20202020 3rd International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)18 citationsDOI

Abstract

COVID-19 pandemic caused vast impact worldwide. Many efforts have been made to tackle the pandemic, including in the deep learning community. In this research, a modification of deep neural network based on Xception model is proposed. The model is used for COVID-19 detection based on the chest X-ray images. The proposed model implements two stacks of two dense layers and batch normalization. The layers addition is used to avoid overfitting of the proposed model. The performance of the proposed model is compared to Resnet50, InceptionV3 and Xception. The experiment result shows that the proposed model has better performance than the other models used in the research. However, its computational time is higher than the other models used in the research.

Topics & Concepts

OverfittingConvolutional neural networkDeep learningComputer scienceNormalization (sociology)Coronavirus disease 2019 (COVID-19)Artificial intelligenceArtificial neural networkPattern recognition (psychology)PathologyInfectious disease (medical specialty)AnthropologyMedicineDiseaseSociologyCOVID-19 diagnosis using AICOVID-19 epidemiological studiesSmart Systems and Machine Learning