Next Article in Journal
Investigation of the Frequency of Coronary Artery Anomalies in MDCT Coronary Angiography and Comparison of Atherosclerotic Involvement between Anomaly Types
Previous Article in Journal
Follow-Up Assessment of Intracranial Aneurysms Treated with Endovascular Coiling: Comparison of Compressed Sensing and Parallel Imaging Time-of-Flight Magnetic Resonance Angiography
Article

Novel COVID-19 Diagnosis Delivery App Using Computed Tomography Images Analyzed with Saliency-Preprocessing and Deep Learning

1
Postgraduate Department, Higher Technological Institute of Lerdo, National Technological Institute of Mexico Campus Lerdo, Lerdo 35150, Mexico
2
Medical Family Unit, Institute of Security and Social Services for State Workers, Torreon 27268, Mexico
*
Author to whom correspondence should be addressed.
Academic Editor: Emilio Quaia
Tomography 2022, 8(3), 1618-1630; https://doi.org/10.3390/tomography8030134
Received: 29 April 2022 / Revised: 1 June 2022 / Accepted: 9 June 2022 / Published: 20 June 2022
(This article belongs to the Section Artificial Intelligence in Medical Imaging)
This app project was aimed to remotely deliver diagnoses and disease-progression information to COVID-19 patients to help minimize risk during this and future pandemics. Data collected from chest computed tomography (CT) scans of COVID-19-infected patients were shared through the app. In this article, we focused on image preprocessing techniques to identify and highlight areas with ground glass opacity (GGO) and pulmonary infiltrates (PIs) in CT image sequences of COVID-19 cases. Convolutional neural networks (CNNs) were used to classify the disease progression of pneumonia. Each GGO and PI pattern was highlighted with saliency map fusion, and the resulting map was used to train and test a CNN classification scheme with three classes. In addition to patients, this information was shared between the respiratory triage/radiologist and the COVID-19 multidisciplinary teams with the application so that the severity of the disease could be understood through CT and medical diagnosis. The three-class, disease-level COVID-19 classification results exhibited a macro-precision of more than 94.89% in a two-fold cross-validation. Both the segmentation and classification results were comparable to those made by a medical specialist. View Full-Text
Keywords: artificial intelligence; computed tomography; COVID-19; SARS-CoV-2; medical diagnostic imaging artificial intelligence; computed tomography; COVID-19; SARS-CoV-2; medical diagnostic imaging
Show Figures

Figure 1

MDPI and ACS Style

Tello-Mijares, S.; Woo, F. Novel COVID-19 Diagnosis Delivery App Using Computed Tomography Images Analyzed with Saliency-Preprocessing and Deep Learning. Tomography 2022, 8, 1618-1630. https://doi.org/10.3390/tomography8030134

AMA Style

Tello-Mijares S, Woo F. Novel COVID-19 Diagnosis Delivery App Using Computed Tomography Images Analyzed with Saliency-Preprocessing and Deep Learning. Tomography. 2022; 8(3):1618-1630. https://doi.org/10.3390/tomography8030134

Chicago/Turabian Style

Tello-Mijares, Santiago, and Fomuy Woo. 2022. "Novel COVID-19 Diagnosis Delivery App Using Computed Tomography Images Analyzed with Saliency-Preprocessing and Deep Learning" Tomography 8, no. 3: 1618-1630. https://doi.org/10.3390/tomography8030134

Find Other Styles

Article Access Map by Country/Region

1
Back to TopTop