11
views
0
recommends
+1 Recommend
1 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Lung Infection Quantification of COVID-19 in CT Images with Deep Learning

      Preprint

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          CT imaging is crucial for diagnosis, assessment and staging COVID-19 infection. Follow-up scans every 3-5 days are often recommended for disease progression. It has been reported that bilateral and peripheral ground glass opacification (GGO) with or without consolidation are predominant CT findings in COVID-19 patients. However, due to lack of computerized quantification tools, only qualitative impression and rough description of infected areas are currently used in radiological reports. In this paper, a deep learning (DL)-based segmentation system is developed to automatically quantify infection regions of interest (ROIs) and their volumetric ratios w.r.t. the lung. The performance of the system was evaluated by comparing the automatically segmented infection regions with the manually-delineated ones on 300 chest CT scans of 300 COVID-19 patients. For fast manual delineation of training samples and possible manual intervention of automatic results, a human-in-the-loop (HITL) strategy has been adopted to assist radiologists for infection region segmentation, which dramatically reduced the total segmentation time to 4 minutes after 3 iterations of model updating. The average Dice simiarility coefficient showed 91.6% agreement between automatic and manual infaction segmentations, and the mean estimation error of percentage of infection (POI) was 0.3% for the whole lung. Finally, possible applications, including but not limited to analysis of follow-up CT scans and infection distributions in the lobes and segments correlated with clinical findings, were discussed.

          Related collections

          Most cited references1

          • Record: found
          • Abstract: found
          • Article: not found

          Importation and Human-to-Human Transmission of a Novel Coronavirus in Vietnam

          To the Editor: The emergence and spread of a novel coronavirus (2019-nCoV) from Wuhan, China, has become a global health concern. 1 Since the detection of the coronavirus in late December 2019, several countries have reported sporadic imported cases among travelers returning from China. 2 We report one family cluster of 2019-nCoV originating from a Chinese man. On January 22, 2020, a 65-year-old man with a history of hypertension, type 2 diabetes, coronary heart disease for which a stent had been implanted, and lung cancer was admitted to the emergency department of Cho Ray Hospital, the referral hospital in Ho Chi Minh City, for low-grade fever and fatigue. He had become ill with fever on January 17, a total of 4 days after he and his wife had flown to Hanoi from the Wuchang district in Wuhan, where outbreaks of 2019-nCoV were occurring. He reported that he had not been exposed to a “wet market” (a market where dead and live animals are sold) in Wuhan. Throat swabs obtained from the patient tested positive for 2019-nCoV on real-time reverse-transcription–polymerase-chain-reaction (RT-PCR) assays. 3 On admission to the hospital, the man was isolated and treated empirically with antiviral agents, broad-spectrum antibiotics, and supportive therapies. Chest radiographs obtained on admission showed an infiltrate in the upper lobe of the left lung (Figure 1A). On January 25, he received supplemental oxygen through a nasal cannula at a rate of 5 liters per minute because of increasing dyspnea with hypoxemia. The partial pressure of oxygen was 57.2 mm Hg while he was breathing ambient air, and a progressive infiltrate and consolidation were observed on chest radiographs (Figure 1B through 1D). His fever disappeared on January 25, and his clinical condition has improved since January 26. His wife had no symptoms of illness while they were traveling. She was healthy as of January 28. The couple’s healthy 27-year-old son had lived in Long An, a province 40 km southwest of Ho Chi Minh City, since October 2019. He had not traveled to a region where 2019-nCoV was spreading, and he had not had any known contact with any person returning from such a region. On January 17, he met his father in Nha Trang in central Vietnam and shared a bedroom with his parents for 3 days in a hotel room that had an air conditioner. On January 20, a dry cough and fever developed in the son. He also reported having had vomiting and loose stools one time before the admission. This suggests that the incubation period for 2019-nCoV may have been 3 days or less in this case. When the son presented at Cho Ray Hospital with his father on January 22, his illness, characterized by a fever (39°C), was recognized and he was immediately isolated. Chest radiographs and other laboratory examinations in this patient showed no abnormalities except for an increased level of C-reactive protein (13.9 mg per liter). Real-time RT-PCR assays for influenza A and B viruses and nonstructural protein 1 antigen rapid tests for dengue viruses were negative in both the father and son. A throat swab in the son was positive for 2019-nCoV. His father was thought to be the source of infection. However, sequencing of strains from the two patients to ascertain the transmission of 2019-nCoV from the father to son has not been performed. The son’s condition was stable after January 23. This family had traveled to four cities across Vietnam using various forms of transportation, including planes, trains, and taxis. A total of 28 close contacts have been identified, and symptoms of an upper respiratory infection have not developed in any of them. This family cluster of 2019-nCoV infection that occurred outside China 4 arouses concern regarding human-to-human transmission.
            Bookmark

            Author and article information

            Journal
            10 March 2020
            Article
            2003.04655
            df970ca6-da24-4afb-89c9-c0e15a6437c4

            http://arxiv.org/licenses/nonexclusive-distrib/1.0/

            History
            Custom metadata
            23 pages, 6 figures
            cs.CV eess.IV q-bio.QM

            Computer vision & Pattern recognition,Quantitative & Systems biology,Electrical engineering

            Comments

            Comment on this article