COVID 19 is an epidemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). On the frontlines of fighting the Coronavirus disease 2019 (COVID-19), Imaging Technology has played a significant role in the detection and the management of COVID 19.
Although the reverse-transcription polymerase chain reaction (RT-PCR) testing is considered as the standard method for screening suspected cases, the sensitivity of RT-PCR screening is relatively poor in some situations. Medical imaging is a useful supplement to RT-PCR testing for the diagnosis of COVID-19. Therefore, medical imaging, in particular, chest computed tomography (CT), is often used as a complementary examination in the diagnosis and management of COVID-19.
CT (Chest Computed Tomography) is the primary tool for screening and evaluating disease severity in patients with COVID-19. It is the most effective method than any other imaging techniques.
Chest CT scan of COVID-19 patients could be evaluated using the following characteristics:
Consolidation and Pulmonary Fibrosis are also typical CT signs in the late stage of COVID-19.
In a study involving 51 COVID-19 patients, concluded that lesions with consolidation could serve as either a marker of disease progression or of more severe disease. Consolidation was found to be more common in pregnant women. COVID-19 infection could increase the risk of pulmonary fibrosis, and the studies suggested that pulmonary fibrosis could become a critical complication in patients with COVID-19. Thus, it is important for clinicians to be alert to the occurrence of pulmonary fibrosis in COVID-19 patients.
Multiple lesions and crazy-paving patterns are also common in COVID-19 patients. Studies found that even on the initial CT scan, COVID-19 is more likely to manifest as multiple lesions rather than a single lesion. Moreover, in the early stage of the disease, the virus is more likely to invade the branches of the right inferior lobar bronchus and cause infection.
Screening of patients with an asymptomatic or atypical presentation is an important aspect of controlling the spread of COVID-19. As a diagnosis tool, CT is also useful to explore the imaging characteristics of asymptomatic and atypical patients.Research found that when the initial RT-PCR results of some patients were negative, chest CT showed typical radiographic findings, including GGO and/or mixed consolidation. Similarly, the CT signs of asymptomatic COVID-19 pneumonia included bifocal extra-zonal, bilateral and multifocal distribution.
However, not all asymptomatic patients have typical radiographic signs (e.g., striped shadows in the lungs). The studies have also confirmed that asymptomatic patients were contagious as well. Not all asymptomatic patients have typical CT signs. The combination of chest CT imaging, RT-PCR, and close follow-up should be used to detect these asymptomatic COVID-19 patients.
The use of CT findings for discriminating COVID-19 pneumonia from other pneumonia has attracted considerable attention. Upon reviewing coronavirus cases associated with severe acute respiratory syndrome (SARS) and the Middle East respiratory syndrome (MERS), GGO and/or lung consolidation were typically characterized on chest CT, which is different from the characteristics of other viruses.
Compared to other pneumonia, patients with COVID-19 are more likely to exhibit CT abnormalities characterized by peripheral distribution, fine reticular opacity, vascular thickening, and reverse halo sign, but are less likely to have significant levels of central-peripheral distribution, pleural effusion, and lymphadenopathy.
As the disease progresses, there is rapid infiltration in lobes of COVID-19 patients as suggested by sequential CT scans. A part of late infection stage COVID-19 patients also exhibits spider web and crazy-paving patterns.
These imaging interpretations played a key role not only in the diagnosis of COVID-19 but also in the monitoring of disease progression and the evaluation of therapeutic efficacy.
In addition to chest CT, other imaging modalities are also used as complementary of chest CT in the diagnosis of COVID-19, including positron emission tomography - CT (PET/CT), lung ultrasound, and magnetic resonance imaging (MRI). In the following subsections, a brief introduction of these imaging techniques is provided.
PET is a sensitive but invasive imaging method that plays an important role in evaluating inflammatory and infectious pulmonary diseases, monitoring disease progression and treatment effect, and improving patient management. It is reported that lung lesions of patients with COVID-19 pneumonia were characterized by a high 18F-FDG uptake, the lymph nodes were involved, and the disseminated disease was absent on 18F-FDG PET/CT imaging. They suggest that 18F-FDG PET/CT can play an auxiliary diagnostic role in COVID-19, especially in the early stage, when the differential diagnosis is difficult. Therefore, PET/CT can play an important role in identifying changes in uptake patterns and locations during viral exposure, and patients with higher FDG uptake in lesions may have a longer recovery period.
In contrast, another different view states that 18F-FDG PET/CT is a more complex test than chest CT, and longer testing period for 18F-FDG PET/CT examinations may increase risk of disease transmission. However, further studies are needed to determine whether 18F-FDG PET/CT is an appropriate testing modality for COVID-19.
As a non-invasive, radiation-free, and portable imaging method, lung ultrasound (LUS) allows for the initial bedside screening of low-risk patients, diagnosis of suspected cases in the emergency room setting, prognostic stratification, and monitoring of the changes in pneumonia. Studies reported that lung ultrasonography could provide results comparable with chest CT for evaluation of COVID-19 pneumonia.
For severe or critical patients, especially those admitted to the ICU and requiring ventilation, LUS is necessary for patient management and monitoring the effectiveness of treatments. More importantly, the use of LUS can reduce the exposure risk between infected patients and health care workers, and discriminate between low-risk and higher-risk cases. For pregnant women with suspected COVID-19, chest CT examination should be avoided as much as possible due to the high radiation. As an alternative, it is recommended that obstetricians and gynecologists could perform lung examinations using LUS.
MRI is a powerful radiation-free imaging technique for visualizing soft tissues. However, it is not commonly applied to COVID-19 diagnosis due to a relatively long scanning time and high cost compared to CT and LUS. Nonetheless, the non-invasive MRI may help evaluate COVID-19 in children and pregnant women.
The infection of SARS-CoV-2 is mainly distributed in the lung, but three minimally invasive autopsies showed that the infection also involves the damages of heart, vessels, liver, kidney, and other organs.
Due to excellent performance for visualizing structural and functional information of various soft organs, MRI could be used to study the vulnerability of different organs for a better understanding of the pathogenesis and mechanisms of COVID-19 infection.
Due to the rapid spread of COVID-19, medical resources became insufficient in many regions. Manual diagnosis in CT scanning requires a lot of labor and consumes a lot of time. In order to reduce the workloads of radiologists, computer-aided diagnostic tools have been developed based on deep learning or machine learning technology. Using AI to assist the disease management of COVID-19 is important.
Despite promising results, many AI models were tested in small datasets. The studies using the small dataset often showed high AUC or accuracy. However, in several studies with the larger dataset the performance drops. In the future, a large external validation dataset should be used. Due to the limited training data of COVID-19, transfer learning using chest CT images could be a good strategy. However, the prognostic analysis of COVID-19 is essential to the management of COVID-19, and combined AI and CT imaging can play an important role in treatment and management. For example, AI and CT can be used to identify patients with a high risk of progressing to severe disease or even of death; it can also be used to evaluate or predict the response of specific therapies.
The COVID-19 is extremely contagious. Therefore, early disease recognition can prompt early patient isolation and early diagnosis and treatment. Various imaging technologies can play important roles in the management of COVID-19. In the current situation, imaging of COVID-19, particularly with chest CT, has a very high value because it shows characteristic manifestations and has enabled frontline clinicians to have primary diagnosis in their first contact with suspected patients, even in the presence of initially false negative lab results. Although CT is the primary tool for screening and evaluating disease severity in patients with COVID-19, other clinical imaging methods, such as PET, MRI, and ultrasound, have also added value by identifying specific imaging features that can be used for the diagnosis of pneumonia. Earlier diagnosis with the aid of imaging allows for early containment and response to this communicable disease as well as overcoming the outbreak as soon as possible through a joint effort.