8
views
0
recommends
+1 Recommend
1 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: not found

      Artificial Intelligence-Driven Assessment of Radiological Images for COVID-19

      review-article

      Read this article at

      ScienceOpenPublisherPMC
      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

          Artificial Intelligence (AI) methods have significant potential for diagnosis and prognosis of COVID-19 infections. Rapid identification of COVID-19 and its severity in individual patients is expected to enable better control of the disease individually and at-large. There has been remarkable interest by the scientific community in using imaging biomarkers to improve detection and management of COVID-19. Exploratory tools such as AI-based models may help explain the complex biological mechanisms and provide better understanding of the underlying pathophysiological processes. The present review focuses on AI-based COVID-19 studies as applies to chest x-ray (CXR) and computed tomography (CT) imaging modalities, and the associated challenges. Explicit radiomics, deep learning methods, and hybrid methods that combine both deep learning and explicit radiomics have the potential to enhance the ability and usefulness of radiological images to assist clinicians in the current COVID-19 pandemic. The aims of this review are: first, to outline COVID-19 AI-analysis workflows, including acquisition of data, feature selection, segmentation methods, feature extraction, and multi-variate model development and validation as appropriate for AI-based COVID-19 studies. Secondly, existing limitations of AI-based COVID-19 analyses are discussed, highlighting potential improvements that can be made. Finally, the impact of AI and radiomics methods and the associated clinical outcomes are summarized. In this review, pipelines that include the key steps for AI-based COVID-19 signatures identification are elaborated. Sample size, non-standard imaging protocols, segmentation, availability of public COVID-19 databases, combination of imaging and clinical information and full clinical validation remain major limitations and challenges. We conclude that AI-based assessment of CXR and CT images has significant potential as a viable pathway for the diagnosis, follow-up and prognosis of COVID-19.

          Related collections

          Most cited references101

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

          Sensitivity of Chest CT for COVID-19: Comparison to RT-PCR

          Summary In a series of 51 patients with chest CT and RT-PCR assay performed within 3 days, the sensitivity of CT for COVID-19 infection was 98% compared to RT-PCR sensitivity of 71% (p<.001). Introduction In December 2019, an outbreak of unexplained pneumonia in Wuhan [1] was caused by a new coronavirus infection named COVID-19 (Corona Virus Disease 2019). Noncontrast chest CT may be considered for early diagnosis of viral disease, although viral nucleic acid detection using real-time polymerase chain reaction (RT-PCR) remains the standard of reference. Chung et al. reported that chest CT may be negative for viral pneumonia of COVID-19 [2] at initial presentation (3/21 patients). Recently, Xie reported 5/167 (3%) patients who had negative RT-PCR for COVID-19 at initial presentation despite chest CT findings typical of viral pneumonia [3]. The purpose of this study was to compare the sensitivity of chest CT and viral nucleic acid assay at initial patient presentation. Materials and Methods The retrospective analysis was approved by institutional review board and patient consent was waived. Patients at Taizhou Enze Medical Center (Group) Enze Hospital were evaluated from January 19, 2020 to February 4, 2020. During this period, chest CT and RT-PCR (Shanghai ZJ Bio-Tech Co, Ltd, Shanghai, China) was performed for consecutive patients who presented with a history of 1) travel or residential history in Wuhan or local endemic areas or contact with individuals with individuals with fever or respiratory symptoms from these areas within 14 days and 2) had fever or acute respiratory symptoms of unknown cause. In the case of an initial negative RT-PCR test, repeat testing was performed at intervals of 1 day or more. Of these patients, we included all patients who had both noncontrast chest CT scan (slice thickness, 5mm) and RT-PCR testing within an interval of 3 days or less and who had an eventual confirmed diagnosis of COVID-19 infection by RT-PCR testing (Figure 1). Typical and atypical chest CT findings were recorded according to CT features previously described for COVD-19 (4,5). The detection rate of COVID-19 infection based on the initial chest CT and RT-PCR was compared. Statistical analysis was performed using McNemar Chi-squared test with significance at the p <.05 level. Figure 1: Flowchart for patient inclusion. Results 51 patients (29 men and 22 women) were included with median age of 45 (interquartile range, 39- 55) years. All patients had throat swab (45 patients) or sputum samples (6 patients) followed by one or more RT-PCR assays. The average time from initial disease onset to CT was 3 +/- 3 days; the average time from initial disease onset to RT-PCR testing was 3 +/- 3 days. 36/51 patients had initial positive RT-PCR for COVID-19. 12/51 patients had COVID-19 confirmed by two RT-PCR nucleic acid tests (1 to 2 days), 2 patients by three tests (2-5 days) and 1 patient by four tests (7 days) after initial onset. 50/51 (98%) patients had evidence of abnormal CT compatible with viral pneumonia at baseline while one patient had a normal CT. Of 50 patients with abnormal CT, 36 (72%) had typical CT manifestations (e.g. peripheral, subpleural ground glass opacities, often in the lower lobes (Figure 2) and 14 (28%) had atypical CT manifestations (Figure 3) [2]. In this patient sample, difference in detection rate for initial CT (50/51 [98%, 95% CI 90-100%]) patients was greater than first RT-PCR (36/51 [71%, 95%CI 56-83%]) patients (p<.001). Figure 2a: Examples of typical chest CT findings compatible with COVID-19 pneumonia in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 74 years old with fever and cough for 5 days. Axial chest CT shows bilateral subpleural ground glass opacities (GGO). B, female, 55 years old, with fever and cough for 7 days. Axial chest CT shows extensive bilateral ground glass opacities and consolidation; C, male, 43 years old, presenting with fever and cough for 1 week. Axial chest CT shows small bilateral areas of peripheral GGO with minimal consolidation; D, female, 43 years old presenting with fever with cough for 5 days. Axial chest CT shows a right lung region of peripheral consolidation. Figure 2b: Examples of typical chest CT findings compatible with COVID-19 pneumonia in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 74 years old with fever and cough for 5 days. Axial chest CT shows bilateral subpleural ground glass opacities (GGO). B, female, 55 years old, with fever and cough for 7 days. Axial chest CT shows extensive bilateral ground glass opacities and consolidation; C, male, 43 years old, presenting with fever and cough for 1 week. Axial chest CT shows small bilateral areas of peripheral GGO with minimal consolidation; D, female, 43 years old presenting with fever with cough for 5 days. Axial chest CT shows a right lung region of peripheral consolidation. Figure 2c: Examples of typical chest CT findings compatible with COVID-19 pneumonia in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 74 years old with fever and cough for 5 days. Axial chest CT shows bilateral subpleural ground glass opacities (GGO). B, female, 55 years old, with fever and cough for 7 days. Axial chest CT shows extensive bilateral ground glass opacities and consolidation; C, male, 43 years old, presenting with fever and cough for 1 week. Axial chest CT shows small bilateral areas of peripheral GGO with minimal consolidation; D, female, 43 years old presenting with fever with cough for 5 days. Axial chest CT shows a right lung region of peripheral consolidation. Figure 2d: Examples of typical chest CT findings compatible with COVID-19 pneumonia in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 74 years old with fever and cough for 5 days. Axial chest CT shows bilateral subpleural ground glass opacities (GGO). B, female, 55 years old, with fever and cough for 7 days. Axial chest CT shows extensive bilateral ground glass opacities and consolidation; C, male, 43 years old, presenting with fever and cough for 1 week. Axial chest CT shows small bilateral areas of peripheral GGO with minimal consolidation; D, female, 43 years old presenting with fever with cough for 5 days. Axial chest CT shows a right lung region of peripheral consolidation. Figure 3a: Examples of chest CT findings less commonly reported in COVID-19 infection (atypical) in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 36 years old with cough for 3 days. Axial chest CT shows a small focal and central ground glass opacity (GGO) in the right upper lobe; B, female, 40 years old. Axial chest CT shows small peripheral linear opacities bilaterally. C, male, 38 years old. Axial chest CT shows a GGO in the central left lower lobe; D, male, 31 years old with fever for 1 day. Axial chest CT shows a linear opacity in the left lower lateral mid lung. Figure 3b: Examples of chest CT findings less commonly reported in COVID-19 infection (atypical) in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 36 years old with cough for 3 days. Axial chest CT shows a small focal and central ground glass opacity (GGO) in the right upper lobe; B, female, 40 years old. Axial chest CT shows small peripheral linear opacities bilaterally. C, male, 38 years old. Axial chest CT shows a GGO in the central left lower lobe; D, male, 31 years old with fever for 1 day. Axial chest CT shows a linear opacity in the left lower lateral mid lung. Figure 3c: Examples of chest CT findings less commonly reported in COVID-19 infection (atypical) in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 36 years old with cough for 3 days. Axial chest CT shows a small focal and central ground glass opacity (GGO) in the right upper lobe; B, female, 40 years old. Axial chest CT shows small peripheral linear opacities bilaterally. C, male, 38 years old. Axial chest CT shows a GGO in the central left lower lobe; D, male, 31 years old with fever for 1 day. Axial chest CT shows a linear opacity in the left lower lateral mid lung. Figure 3d: Examples of chest CT findings less commonly reported in COVID-19 infection (atypical) in patients with epidemiological and clinical presentation suspicious for COVID-19 infection. A, male, 36 years old with cough for 3 days. Axial chest CT shows a small focal and central ground glass opacity (GGO) in the right upper lobe; B, female, 40 years old. Axial chest CT shows small peripheral linear opacities bilaterally. C, male, 38 years old. Axial chest CT shows a GGO in the central left lower lobe; D, male, 31 years old with fever for 1 day. Axial chest CT shows a linear opacity in the left lower lateral mid lung. Discussion In our series, the sensitivity of chest CT was greater than that of RT-PCR (98% vs 71%, respectively, p<.001). The reasons for the low efficiency of viral nucleic acid detection may include: 1) immature development of nucleic acid detection technology; 2) variation in detection rate from different manufacturers; 3) low patient viral load; or 4) improper clinical sampling. The reasons for the relatively lower RT-PCR detection rate in our sample compared to a prior report are unknown (3). Our results support the use of chest CT for screening for COVD-19 for patients with clinical and epidemiologic features compatible with COVID-19 infection particularly when RT-PCR testing is negative.
            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping

            Background Radiomic features may quantify characteristics present in medical imaging. However, the lack of standardized definitions and validated reference values have hampered clinical use. Purpose To standardize a set of 174 radiomic features. Materials and Methods Radiomic features were assessed in three phases. In phase I, 487 features were derived from the basic set of 174 features. Twenty-five research teams with unique radiomics software implementations computed feature values directly from a digital phantom, without any additional image processing. In phase II, 15 teams computed values for 1347 derived features using a CT image of a patient with lung cancer and predefined image processing configurations. In both phases, consensus among the teams on the validity of tentative reference values was measured through the frequency of the modal value and classified as follows: less than three matches, weak; three to five matches, moderate; six to nine matches, strong; 10 or more matches, very strong. In the final phase (phase III), a public data set of multimodality images (CT, fluorine 18 fluorodeoxyglucose PET, and T1-weighted MRI) from 51 patients with soft-tissue sarcoma was used to prospectively assess reproducibility of standardized features. Results Consensus on reference values was initially weak for 232 of 302 features (76.8%) at phase I and 703 of 1075 features (65.4%) at phase II. At the final iteration, weak consensus remained for only two of 487 features (0.4%) at phase I and 19 of 1347 features (1.4%) at phase II. Strong or better consensus was achieved for 463 of 487 features (95.1%) at phase I and 1220 of 1347 features (90.6%) at phase II. Overall, 169 of 174 features were standardized in the first two phases. In the final validation phase (phase III), most of the 169 standardized features could be excellently reproduced (166 with CT; 164 with PET; and 164 with MRI). Conclusion A set of 169 radiomics features was standardized, which enabled verification and calibration of different radiomics software. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kuhl and Truhn in this issue.
              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Artificial Intelligence Distinguishes COVID-19 from Community Acquired Pneumonia on Chest CT

              Background Coronavirus disease has widely spread all over the world since the beginning of 2020. It is desirable to develop automatic and accurate detection of COVID-19 using chest CT. Purpose To develop a fully automatic framework to detect COVID-19 using chest CT and evaluate its performances. Materials and Methods In this retrospective and multi-center study, a deep learning model, COVID-19 detection neural network (COVNet), was developed to extract visual features from volumetric chest CT exams for the detection of COVID-19. Community acquired pneumonia (CAP) and other non-pneumonia CT exams were included to test the robustness of the model. The datasets were collected from 6 hospitals between August 2016 and February 2020. Diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC), sensitivity and specificity. Results The collected dataset consisted of 4356 chest CT exams from 3,322 patients. The average age is 49±15 years and there were slightly more male patients than female (1838 vs 1484; p-value=0.29). The per-exam sensitivity and specificity for detecting COVID-19 in the independent test set was 114 of 127 (90% [95% CI: 83%, 94%]) and 294 of 307 (96% [95% CI: 93%, 98%]), respectively, with an AUC of 0.96 (p-value<0.001). The per-exam sensitivity and specificity for detecting CAP in the independent test set was 87% (152 of 175) and 92% (239 of 259), respectively, with an AUC of 0.95 (95% CI: 0.93, 0.97). Conclusions A deep learning model can accurately detect COVID-19 and differentiate it from community acquired pneumonia and other lung diseases.
                Bookmark

                Author and article information

                Journal
                Comput Biol Med
                Comput Biol Med
                Computers in Biology and Medicine
                Published by Elsevier Ltd.
                0010-4825
                1879-0534
                21 July 2021
                21 July 2021
                : 104665
                Affiliations
                [a ]Department of Radiology and Molecular Imaging, College of Medicine and Health Science, Sultan Qaboos University, PO. Box 35, Al Khod, Muscat 123, Oman
                [b ]Department of Radiology, Royal Hospital, Muscat, Oman
                [c ]Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211 Geneva 4, Switzerland
                [d ]Geneva University Neurocenter, Geneva University, Geneva, Switzerland
                [e ]Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, Groningen, Netherlands
                [f ]Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
                [g ]Departments of Radiology and Physics, University of British Columbia, Vancouver BC, Canada
                [h ]Department of Integrative Oncology, BC Cancer Research Institute, Vancouver BC, Canada
                Author notes
                []Corresponding author: , Department of Radiology and Molecular Imaging, College of Medicine and Health Science, Sultan Qaboos University, PO. Box 35, Al Khod, Muscat 123, Oman, Tel: 0096897006246
                [∗∗ ]Corresponding author: , Tel: 0096879199889
                Article
                S0010-4825(21)00459-5 104665
                10.1016/j.compbiomed.2021.104665
                8291996
                34343890
                985c6e8b-93b5-4b4f-8a18-b2c091a3c927
                © 2021 Published by Elsevier Ltd.

                Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.

                History
                : 26 May 2021
                : 11 July 2021
                : 17 July 2021
                Categories
                Article

                covid-19,computed tomography,chest x-ray,artificial intelligence,radiomics,deep learning,deep radiomics

                Comments

                Comment on this article