1
|
Killekar A, Grodecki K, Lin A, Cadet S, McElhinney P, Razipour A, Chan C, Pressman BD, Julien P, Chen P, Simon J, Maurovich-Horvat P, Gaibazzi N, Thakur U, Mancini E, Agalbato C, Munechika J, Matsumoto H, Menè R, Parati G, Cernigliaro F, Nerlekar N, Torlasco C, Pontone G, Dey D, Slomka P. Rapid quantification of COVID-19 pneumonia burden from computed tomography with convolutional long short-term memory networks. J Med Imaging (Bellingham) 2022; 9:054001. [PMID: 36090960 PMCID: PMC9446878 DOI: 10.1117/1.jmi.9.5.054001] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/04/2022] [Accepted: 08/16/2022] [Indexed: 11/14/2022] Open
Abstract
Purpose: Quantitative lung measures derived from computed tomography (CT) have been demonstrated to improve prognostication in coronavirus disease 2019 (COVID-19) patients but are not part of clinical routine because the required manual segmentation of lung lesions is prohibitively time consuming. We aim to automatically segment ground-glass opacities and high opacities (comprising consolidation and pleural effusion). Approach: We propose a new fully automated deep-learning framework for fast multi-class segmentation of lung lesions in COVID-19 pneumonia from both contrast and non-contrast CT images using convolutional long short-term memory (ConvLSTM) networks. Utilizing the expert annotations, model training was performed using five-fold cross-validation to segment COVID-19 lesions. The performance of the method was evaluated on CT datasets from 197 patients with a positive reverse transcription polymerase chain reaction test result for SARS-CoV-2, 68 unseen test cases, and 695 independent controls. Results: Strong agreement between expert manual and automatic segmentation was obtained for lung lesions with a Dice score of 0.89 ± 0.07 ; excellent correlations of 0.93 and 0.98 for ground-glass opacity (GGO) and high opacity volumes, respectively, were obtained. In the external testing set of 68 patients, we observed a Dice score of 0.89 ± 0.06 as well as excellent correlations of 0.99 and 0.98 for GGO and high opacity volumes, respectively. Computations for a CT scan comprising 120 slices were performed under 3 s on a computer equipped with an NVIDIA TITAN RTX GPU. Diagnostically, the automated quantification of the lung burden % discriminate COVID-19 patients from controls with an area under the receiver operating curve of 0.96 (0.95-0.98). Conclusions: Our method allows for the rapid fully automated quantitative measurement of the pneumonia burden from CT, which can be used to rapidly assess the severity of COVID-19 pneumonia on chest CT.
Collapse
Affiliation(s)
- Aditya Killekar
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | | | - Andrew Lin
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Sebastien Cadet
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Priscilla McElhinney
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Aryabod Razipour
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Cato Chan
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Barry D. Pressman
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Peter Julien
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Peter Chen
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | | | | | | | - Udit Thakur
- Monash Health, Melbourne, Victoria, Australia
| | | | - Cecilia Agalbato
- University of Milan, Centro Cardiologico Monzino IRCCS, Milan, Italy
| | | | | | - Roberto Menè
- IRCCS Istituto Auxologico Italiano, Department of Cardiovascular, Neural and Metabolic Sciences, Milan, Italy
- University of Milano-Bicocca, Department of Medicine and Surgery, Milan, Italy
| | - Gianfranco Parati
- IRCCS Istituto Auxologico Italiano, Department of Cardiovascular, Neural and Metabolic Sciences, Milan, Italy
- University of Milano-Bicocca, Department of Medicine and Surgery, Milan, Italy
| | - Franco Cernigliaro
- IRCCS Istituto Auxologico Italiano, Department of Cardiovascular, Neural and Metabolic Sciences, Milan, Italy
- University of Milano-Bicocca, Department of Medicine and Surgery, Milan, Italy
| | | | - Camilla Torlasco
- IRCCS Istituto Auxologico Italiano, Department of Cardiovascular, Neural and Metabolic Sciences, Milan, Italy
- University of Milano-Bicocca, Department of Medicine and Surgery, Milan, Italy
| | - Gianluca Pontone
- University of Milan, Centro Cardiologico Monzino IRCCS, Milan, Italy
| | - Damini Dey
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| | - Piotr Slomka
- Cedars-Sinai Medical Center, Department of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences and Imaging, Los Angeles, California, United States
| |
Collapse
|
2
|
Farinhas J, Lee JS. Imaging of the head and neck during the COVID19 pandemic. OPERATIVE TECHNIQUES IN OTOLARYNGOLOGY-HEAD AND NECK SURGERY 2022; 33:147-157. [PMID: 35505951 PMCID: PMC9047486 DOI: 10.1016/j.otot.2022.04.011] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
Abstract
There is a wide spectrum of clinical manifestation of COVID-19 in the head and neck, but often these do not have an imaging correlate. This review will highlight the most common imaging features of COVID-19 in the head and neck that can be seen on routine head and neck CT and MRI. In addition, situations where a more dedicated imaging protocol is required will be highlighted. Finally, as mass vaccination efforts are underway worldwide, post vaccination imaging can often complicate cancer surveillance imaging. Post vaccination imaging features and recommendations will be discussed.
Collapse
|
3
|
Mousseaux E, Fayol A, Danchin N, Soulat G, Charpentier E, Livrozet M, Carves JB, Tea V, Salem FB, Chamandi C, Hulot JS, Puymirat E. Association between coronary artery calcifications and 6-month mortality in hospitalized patients with COVID-19. Diagn Interv Imaging 2021; 102:717-725. [PMID: 34312110 PMCID: PMC8275480 DOI: 10.1016/j.diii.2021.06.007] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/25/2021] [Revised: 06/28/2021] [Accepted: 06/29/2021] [Indexed: 12/14/2022]
Abstract
Purpose The purpose of this study was to evaluate the association between coronary artery calcium (CAC) visual score and 6-month mortality in patients with coronavirus disease 2019 (COVID-19). Material and methods A single-center prospective observational cohort was conducted in 169 COVID-19 consecutive hospitalized patients between March 13 and April 1, 2020, and follow-up for 6-months. A four-level visual CAC scoring was assessed by analyzing images obtained after the first routine non-ECG-gated CT performed to detect COVID-19 pneumonia. Results Among 169 confirmed COVID-19 patients (118 men, 51 women; mean age, 65.6 ± 18.8 [SD] years; age range: 30–95 years) 63 (37%) presented with either moderate (n = 26, 15.3%) or heavy (n = 37, 21.8%) CAC detected by CT and 20 (11.8%) had history of cardiovascular disease requiring specific preventive treatment. At six months, mortality rate (45/169; 26.6%) increased with magnitude of CAC and was 7/64 (10.9%), 11/42 (26.2%), 10/26 (38.5%), 17/37 (45.9%) for no-CAC, mild-CAC, moderate-CAC and heavy-CAC groups, respectively (P = 0.001). Compared to the no CAC group, risk of death increased after adjustment with magnitude of CAC (HR: 2.23, 95% CI: 0.73–6.87, P = 0.16; HR: 2.78, 95% CI: 0.85–9.07, P0.09; HR: 5.38, 95% CI: 1.57–18.40, P = 0.007; in mild CAC, moderate and heavy CAC groups, respectively). In patients without previous coronary artery disease (154/169; 91%), mortality increased from 10.9% to 45.8% (P = 0.001) according to the magnitude of CAC categories. After adjustment, presence of moderate or heavy CAC was associated with higher mortality (HR: 2.26, 95% CI: 1.09–4.69, P = 0.03). Conclusion By using non-ECG-gated CT during the initial pulmonary assessment of COVID-19, heavy CAC is independently associated with 6-month mortality in patients hospitalized for severe COVID-19 pneumonia.
Collapse
Affiliation(s)
- Elie Mousseaux
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Assistance Publique-Hôpitaux des Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France.
| | - Antoine Fayol
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France; CIC1418 and DMU CARTE, Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Nicolas Danchin
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Cardiology, Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Gilles Soulat
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Assistance Publique-Hôpitaux des Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France
| | - Etienne Charpentier
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Assistance Publique-Hôpitaux des Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Marine Livrozet
- Université de Paris, Faculté de Médecine, 75006 Paris, France; CIC1418 and DMU CARTE, Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Jean-Baptiste Carves
- Université de Paris, Faculté de Médecine, 75006 Paris, France; CIC1418 and DMU CARTE, Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Victoria Tea
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France
| | - Fares Ben Salem
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Assistance Publique-Hôpitaux des Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Chekrallah Chamandi
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France
| | - Jean-Sébastien Hulot
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France; CIC1418 and DMU CARTE, Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| | - Etienne Puymirat
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Institut National de la Santé et de la Recherche Médicale, PARCC, UMR970, 75015 Paris, France; Department of Cardiology, Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges-Pompidou, 75015 Paris, France
| |
Collapse
|
4
|
Chassagnon G, Regard L, Soyer P, Revel MP. COVID-19 after 18 months: Where do we stand? Diagn Interv Imaging 2021; 102:491-492. [PMID: 34183299 PMCID: PMC8222566 DOI: 10.1016/j.diii.2021.06.003] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/15/2021] [Accepted: 06/15/2021] [Indexed: 12/30/2022]
Affiliation(s)
- Guillaume Chassagnon
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Hôpital Cochin, AP-HP, 27, rue du Faubourg St Jacques, 75014 Paris, France.
| | - Lucile Regard
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Pulmonology, Hôpital Cochin, AP-HP, 75014 Paris, France
| | - Philippe Soyer
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Hôpital Cochin, AP-HP, 27, rue du Faubourg St Jacques, 75014 Paris, France
| | - Marie-Pierre Revel
- Université de Paris, Faculté de Médecine, 75006 Paris, France; Department of Radiology, Hôpital Cochin, AP-HP, 27, rue du Faubourg St Jacques, 75014 Paris, France
| |
Collapse
|
5
|
Kato S, Ishiwata Y, Aoki R, Iwasawa T, Hagiwara E, Ogura T, Utsunomiya D. Imaging of COVID-19: An update of current evidences. Diagn Interv Imaging 2021; 102:493-500. [PMID: 34088635 PMCID: PMC8148573 DOI: 10.1016/j.diii.2021.05.006] [Citation(s) in RCA: 14] [Impact Index Per Article: 4.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/11/2021] [Revised: 05/16/2021] [Accepted: 05/17/2021] [Indexed: 12/15/2022]
Abstract
Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been reported as a global emergency. As respiratory dysfunction is a major clinical presentation of COVID-19, chest computed tomography (CT) plays a central role in the diagnosis and management of patients with COVID-19. Recent advances in imaging approaches using artificial intelligence have been essential as a quantification and diagnostic tool to differentiate COVID-19 from other respiratory infectious diseases. Furthermore, cardiovascular involvement in patients with COVID-19 is not negligible and may result in rapid worsening of the disease and sudden death. Cardiac magnetic resonance imaging can accurately depict myocardial involvement in SARS-CoV-2 infection. This review summarizes the role of the radiology department in the management and the diagnosis of COVID-19, with a special emphasis on ultra-high-resolution CT findings, cardiovascular complications and the potential of artificial intelligence.
Collapse
Affiliation(s)
- Shingo Kato
- Department of Diagnostic Radiology, Yokohama City University Graduate School of Medicine, 236-0004 Yokohama, Kanagawa, Japan.
| | - Yoshinobu Ishiwata
- Department of Diagnostic Radiology, Yokohama City University Graduate School of Medicine, 236-0004 Yokohama, Kanagawa, Japan
| | - Ryo Aoki
- Department of Diagnostic Radiology, Yokohama City University Medical Center, 232-0024 Yokohama, Kanagawa, Japan
| | - Tae Iwasawa
- Department of Diagnostic Radiology, Kanagawa Cardiovascular and Respiratory Center, 236-0051 Yokohama, Kanagawa, Japan
| | - Eri Hagiwara
- Department of Respiratory Medicine, Kanagawa Cardiovascular and Respiratory Center, 236-0051 Yokohama, Kanagawa, Japan
| | - Takashi Ogura
- Department of Respiratory Medicine, Kanagawa Cardiovascular and Respiratory Center, 236-0051 Yokohama, Kanagawa, Japan
| | - Daisuke Utsunomiya
- Department of Diagnostic Radiology, Yokohama City University Graduate School of Medicine, 236-0004 Yokohama, Kanagawa, Japan
| |
Collapse
|
6
|
Feasibility of lung imaging with a large field-of-view spectral photon-counting CT system. Diagn Interv Imaging 2021; 102:305-312. [PMID: 33610503 DOI: 10.1016/j.diii.2021.01.001] [Citation(s) in RCA: 44] [Impact Index Per Article: 14.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/09/2020] [Revised: 01/04/2021] [Accepted: 01/05/2021] [Indexed: 12/12/2022]
Abstract
PURPOSE The purpose of this study was to characterize the technical capabilities and feasibility of a large field-of-view clinical spectral photon-counting computed tomography (SPCCT) prototype for high-resolution (HR) lung imaging. MATERIALS AND METHODS Measurement of modulation transfer function (MTF) and acquisition of a line pairs phantom were performed. An anthropomorphic lung nodule phantom was scanned with standard (120kVp, 62mAs), low (120kVp, 11mAs), and ultra-low (80kVp, 3mAs) radiation doses. A human volunteer underwent standard (120kVp, 63mAs) and low (120kVp, 11mAs) dose scans after approval by the ethics committee. HR images were reconstructed with 1024 matrix, 300mm field of view and 0.25mm slice thickness using a filtered-back projection (FBP) and two levels of iterative reconstruction (iDose 5 and 9). The conspicuity and sharpness of various lung structures (distal airways, vessels, fissures and proximal bronchial wall), image noise, and overall image quality were independently analyzed by three radiologists and compared to a previous HR lung CT examination of the same volunteer performed with a conventional CT equipped with energy integrating detectors (120kVp, 10mAs, FBP). RESULTS Ten percent MTF was measured at 22.3lp/cm with a cut-off at 31lp/cm. Up to 28lp/cm were depicted. While mixed and solid nodules were easily depicted on standard and low-dose phantom images, higher iDose levels and slice thicknesses (1mm) were needed to visualize ground-glass components on ultra-low-dose images. Standard dose SPCCT images of in vivo lung structures were of greater conspicuity and sharpness, with greater overall image quality, and similar image noise (despite a flux reduction of 23%) to conventional CT images. Low-dose SPCCT images were of greater or similar conspicuity and sharpness, similar overall image quality, and lower but acceptable image noise (despite a flux reduction of 89%). CONCLUSIONS A large field-of-view SPCCT prototype demonstrates HR technical capabilities and high image quality for high resolution lung CT in human.
Collapse
|
7
|
Affiliation(s)
- Marie-Pierre Revel
- Department of Radiology, Hôpital Cochin, AP-HP, 75014 Paris, France; Université de Paris, Descartes-Paris 5, 75006 Paris, France.
| |
Collapse
|
8
|
Li J, Long X, Wang X, Fang F, Lv X, Zhang D, Sun Y, Hu S, Lin Z, Xiong N. Radiology indispensable for tracking COVID-19. Diagn Interv Imaging 2020; 102:69-75. [PMID: 33281082 PMCID: PMC7685040 DOI: 10.1016/j.diii.2020.11.008] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/16/2020] [Revised: 11/17/2020] [Accepted: 11/18/2020] [Indexed: 12/20/2022]
Abstract
Currently, chest computed tomography is recommended as the first-line imaging test for detecting COVID-19 pneumonia. The most typical CT imaging finding of COVID-19 patients is ground-glass opacity, combined with reticular and/or interlobular septal thickening and consolidation. CT is useful for monitoring patients with COVID-19, identifying associated vascular abnormalities and making differential diagnosis.
With the rapid spread of COVID-19 worldwide, early detection and efficient isolation of suspected patients are especially important to prevent the transmission. Although nucleic acid testing of SARS-CoV-2 is still the gold standard for diagnosis, there are well-recognized early-detection problems including time-consuming in the diagnosis process, noticeable false-negative rate in the early stage and lacking nucleic acid testing kits in some areas. Therefore, effective and rational applications of imaging technologies are critical in aiding the screen and helping the diagnosis of suspected patients. Currently, chest computed tomography is recommended as the first-line imaging test for detecting COVID-19 pneumonia, which could allow not only early detection of the typical chest manifestations, but also timely estimation of the disease severity and therapeutic effects. In addition, other radiological methods including chest X-ray, magnetic resonance imaging, and positron emission computed tomography also show significant advantages in the detection of COVID-19 pneumonia. This review summarizes the applications of radiology and nuclear medicine in detecting and diagnosing COVID-19. It highlights the importance for these technologies to curb the rapid transmission during the pandemic, considering findings from special groups such as children and pregnant women.
Collapse
Affiliation(s)
- Jingwen Li
- Department of Neurology, Tongji Medical College, Huazhong University of Science and Technology, Union Hospital, Wuhan, Hubei, China
| | - Xi Long
- Department of Radiology, Tongji Medical College, Huazhong University of Science and Technology, Union Hospital, Wuhan, Hubei, China
| | - Xinyi Wang
- Department of Neurology, Tongji Medical College, Huazhong University of Science and Technology, Union Hospital, Wuhan, Hubei, China
| | - Fang Fang
- Department of Radiology, Wuhan Red Cross Hospital, Wuhan, Hubei, China
| | - Xuefei Lv
- Department of Radiology, Wuhan Red Cross Hospital, Wuhan, Hubei, China
| | - Dandan Zhang
- Department of Radiology, Wuhan Red Cross Hospital, Wuhan, Hubei, China
| | - Yu Sun
- Department of Radiology, Wuhan Red Cross Hospital, Wuhan, Hubei, China
| | - Shaoping Hu
- Department of Radiology, Wuhan Red Cross Hospital, Wuhan, Hubei, China
| | - Zhicheng Lin
- Harvard Medical School, Mclean Hospital, 02478 Belmont, MA, USA
| | - Nian Xiong
- Department of Neurology, Tongji Medical College, Huazhong University of Science and Technology, Union Hospital, Wuhan, Hubei, China; Wuhan Red Cross Hospital, Wuhan, Hubei, China.
| |
Collapse
|
9
|
Görgülü Ö, Duyan M. rRT-PCR Results of a Covid-19 Diagnosed Geriatric Patient. ACTA ACUST UNITED AC 2020; 2:2423-2426. [PMID: 33103060 PMCID: PMC7567648 DOI: 10.1007/s42399-020-00590-9] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 10/13/2020] [Indexed: 12/12/2022]
Abstract
In this study, we aimed to present a geriatric patient with the diagnosis of COVID-19 and with contradictory results in rRT-PCR examinations in short time intervals. A 69-year-old male patient was admitted to the emergency room on the 18th day of May 2020, with the complaints of fever, sweating, myalgia, dry cough that continued for 5 days, and the lack of taste that started on the day he applied to the emergency room. Comorbidity factors include diabetes mellitus, bronchial asthma, and hypertension. The patient has a history of 36 years of smoking 1.5 packs per day. High laboratory findings during hospitalization: monocytes, creatinine, CRP (C-reactive protein). In the thorax CT, in the parenchyma areas of both lungs, there are increases in attenuation with multilobe distributions (more visible at the level of the upper lobes) in the form of ground-glass opacities. May 19, 2020, was subjected to the rRT-PCR test, repeated twice on the 19th of May which also resulted in positive. Despite rRT-PCR tests, which were negative on 27th of May and positive on 28th of May, the patient, whose symptoms disappeared, and general condition improved, was discharged on June 1, 2020, with the recommendation for home isolation. In our case, unlike the incubation period only, we encountered a negative rRT-PCR result on the 8th day after diagnosis. Therefore, the COVID-19 pandemic control and filiation evaluation with the rRT-PCR test may produce false negative results.
Collapse
Affiliation(s)
- Özkan Görgülü
- Department of Anesthesiology and Reanimation, Antalya Training and Research Hospital, Varlık Mh. Kazım Karabekir Cd, 07100 Antalya, Turkey
| | - Murat Duyan
- Department of Emergency Medicine, Antalya Training and Research Hospital, Antalya, Turkey
| |
Collapse
|
10
|
COVID-19 Screening Using a Lightweight Convolutional Neural Network with Generative Adversarial Network Data Augmentation. Symmetry (Basel) 2020. [DOI: 10.3390/sym12091530] [Citation(s) in RCA: 18] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/23/2022] Open
Abstract
COVID-19 is a disease that can be spread easily with minimal physical contact. Currently, the World Health Organization (WHO) has endorsed the reverse transcription-polymerase chain reaction swab test as a diagnostic tool to confirm COVID-19 cases. This test requires at least a day for the results to come out depending on the available facilities. Many countries have adopted a targeted approach in screening potential patients due to the cost. However, there is a need for a fast and accurate screening test to complement this targeted approach, so that the potential virus carriers can be quarantined as early as possible. The X-ray is a good screening modality; it is quick at capturing, cheap, and widely available, even in third world countries. Therefore, a deep learning approach has been proposed to automate the screening process by introducing LightCovidNet, a lightweight deep learning model that is suitable for the mobile platform. It is important to have a lightweight model so that it can be used all over the world even on a standard mobile phone. The model has been trained with additional synthetic data that were generated from the conditional deep convolutional generative adversarial network. LightCovidNet consists of three components, which are entry, middle, and exit flows. The middle flow comprises five units of feed-forward convolutional neural networks that are built using separable convolution operators. The exit flow is designed to improve the multi-scale capability of the network through a simplified spatial pyramid pooling module. It is a symmetrical architecture with three parallel pooling branches that enable the network to learn multi-scale features, which is suitable for cases wherein the X-ray images were captured from all over the world independently. Besides, the usage of separable convolution has managed to reduce the memory usage without affecting the classification accuracy. The proposed method managed to get the best mean accuracy of 0.9697 with a low memory requirement of just 841,771 parameters. Moreover, the symmetrical spatial pyramid pooling module is the most crucial component; the absence of this module will reduce the screening accuracy to just 0.9237. Hence, the developed model is suitable to be implemented for mass COVID-19 screening.
Collapse
|