ABSTRACT
Coronavirus disease (COVID-19) is a newly discovered viral sickness that can be fatal. The majority of patients will experience mild to severe respiratory problems and will improve without need for special treatment. Persons over 65, and for those who are underlying medical disorders such cardiovascular disease, asthma, respiratory illness, and cancer, are more prone for developing severe symptoms. In these conditions, 3D volumetric imaging has proven to be a useful technique for COVID-19 patient diagnosis and prognosis. We present a new approach for detecting and classifying COVID-19 infection using 3D volumetric lung imaging in this work. For the detection and classification process, we have used 3D volumetric image processing and deep learning techniques, respectively. Early recognition and finding are basic elements to stop COVID-19 spreading. Various profound learning-based approaches had been proposed for COVID-19 separating CT examines as an instrument to computerize and assist with finding. These methods suffer with at least one of the faults listed below: (i) They treat each CT scan individually (ii) These methods are trained and tested on the same dataset. To address these two challenges, we present an accurate deep learning technique for COVID-19 screening using a democratic framework in this paper. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
ABSTRACT
During the last years, deep learning has been used intensively in medical domain making considerable progress in the diagnosis of diseases from radiology images. This is mainly due to the availability of proven algorithms on several computer vision tasks and the publicly accessible medical datasets. However, most approaches that apply deep learning techniques to radiology images transform these images into a format that conforms with the inputs of conventional learning algorithms and deal with the dataset as a set of 2D independent slices instead of volumetric images. In this work we deal with the problem of preparing DICOM CT scans as 3D images for a machine learning/deep learning architecture. We propose a general preprocessing pipeline composed of four stages for volumetric images processing followed by a 3D CNN architecture for 3D images classification. The proposed pipeline is evaluated through a case study for COVID-19 detection from chest CT scans. Experiment results demonstrate the effectiveness of the proposed preprocessing operations. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
ABSTRACT
Coronavirus disease (COVID-19) is a newly discovered viral sickness that can be fatal. The majority of patients will experience mild to severe respiratory problems and will improve without need for special treatment. Persons over 65, and for those who are underlying medical disorders such cardiovascular disease, asthma, respiratory illness, and cancer, are more prone for developing severe symptoms. In these conditions, 3D volumetric imaging has proven to be a useful technique for COVID-19 patient diagnosis and prognosis. We present a new approach for detecting and classifying COVID-19 infection using 3D volumetric lung imaging in this work. For the detection and classification process, we have used 3D volumetric image processing and deep learning techniques, respectively. Early recognition and finding are basic elements to stop COVID-19 spreading. Various profound learning-based approaches had been proposed for COVID-19 separating CT examines as an instrument to computerize and assist with finding. These methods suffer with at least one of the faults listed below: (i) They treat each CT scan individually (ii) These methods are trained and tested on the same dataset. To address these two challenges, we present an accurate deep learning technique for COVID-19 screening using a democratic framework in this paper. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.