2020/11/01 by Parham Yazdekhasty, Yazdekhasty, Parham, Ali Zindar +11
Computer Science · Medicine · #AI in cancer detection #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Radiomics and Machine Learning in Medical Imaging #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2011.00631
openalex publication_date 2020/11/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The new coronavirus infection has shocked the world since early 2020 with its\naggressive outbreak. Rapid detection of the disease saves lives, and relying on\nmedical imaging (Computed Tomography and X-ray) to detect infected lungs has\nshown to be effective. Deep learning and convolutional neural networks have\nbeen used for image analysis in this context. However, accurate identification\nof infected regions has proven challenging for two main reasons. Firstly, the\ncharacteristics of infected areas differ in different images. Secondly,\ninsufficient training data makes it challenging to train various machine\nlearning algorithms, including deep-learning models. This paper proposes an\napproach to segment lung regions infected by COVID-19 to help cardiologists\ndiagnose the disease more accurately, faster, and more manageable. We propose a\nbifurcated 2-D model for two types of segmentation. This model uses a shared\nencoder and a bifurcated connection to two separate decoders. One decoder is\nfor segmentation of the healthy region of the lungs, while the other is for the\nsegmentation of the infected regions. Experiments on publically available\nimages show that the bifurcated structure segments infected regions of the\nlungs better than state of the art.\n