CT features of community-acquired pneumonia at the emergency department.

BACKGROUND: Chest computed tomography (CT) was reported to improve the diagnosis of community-acquired pneumonia (CAP) as compared to chest X-ray (CXR). The aim of this study is to describe the CT-patterns of CAP in a large population visiting the emergency department and to see if some of them are more frequently missed on CXR. MATERIALS AND METHODS: This is an ancillary analysis of the prospective multicenter ESCAPED study including 319 patients. We selected the 163 definite or probable CAP based on adjudication committee classification; 147 available chest CT scans were reinterpreted by 3 chest radiologists to identify CAP patterns. These CT-patterns were correlated to epidemiological, biological and microbiological data, and compared between false negative and true positive CXR CAP. RESULTS: Six patterns were identified: lobar pneumonia (51/147, 35%), including 35 with plurifocal involvement; lobular pneumonia (43/147, 29%); unilobar infra-segmental consolidation (24/147, 16%); bronchiolitis (16/147, 11%), including 4 unilobar bronchiolitis; atelectasis and bronchial abnormalities (8/147, 5.5%); interstitial pneumonia (5/147, 3.5%). Bacteria were isolated in 41% of patients with lobar pneumonia-pattern (mostly Streptococcus pneumoniae and Mycoplasma pneumonia) versus 19% in other patients (p = 0.01). Respiratory viruses were equally distributed within all patterns. CXR was falsely negative in 46/147 (31%) patients. Lobar pneumonia was significantly less missed on CXR than other patterns (p = 0.003), especially lobular pneumonia and unilobar infra-segmental consolidation, missed in 35% and 58% of cases, respectively. CONCLUSION: Lobar and lobular pneumonias are the most frequent CT-patterns. Lobar pneumonia is appropriately detected on CXR and mainly due to Streptococcus pneumoniae or Mycoplasma pneumoniae. Chest CT is very useful to identify CAP in other CT-patterns. Prior the COVID pandemic, CAP was rarely responsible for interstitial opacities on CT.

Quantitative CT comparison between COVID-19 and mycoplasma pneumonia suspected as COVID-19: a longitudinal study.

OBJECTIVE: The purpose of this study was to compare imaging features between COVID-19 and mycoplasma pneumonia (MP). MATERIALS AND METHODS: The data of patients with mild COVID-19 and MP who underwent chest computed tomography (CT) examination from February 1, 2020 to April 17, 2020 were retrospectively analyzed. The Pneumonia-CT-LKM-PP model based on a deep learning algorithm was used to automatically quantify the number, volume, and involved lobes of pulmonary lesions, and longitudinal changes in quantitative parameters were assessed in three CT follow-ups. RESULTS: A total of 10 patients with mild COVID-19 and 13 patients with MP were included in this study. There was no difference in lymphocyte counts at baseline between the two groups (1.43 ± 0.45 vs. 1.44 ± 0.50, p = 0.279). C-reactive protein levels were significantly higher in MP group than in COVID-19 group (p < 0.05). The number, volume, and involved lobes of pulmonary lesions reached a peak in 7-14 days in the COVID-19 group, but there was no peak or declining trend over time in the MP group (p < 0.05). CONCLUSION: Based on the longitudinal changes of quantitative CT, pulmonary lesions peaked at 7-14 days in patients with COVID-19, and this may be useful to distinguish COVID-19 from MP and evaluate curative effects and prognosis.

A CT radiomics analysis of COVID-19-related ground-glass opacities and consolidation: Is it valuable in a differential diagnosis with other atypical pneumonias?

PURPOSE: To evaluate the discrimination of parenchymal lesions between COVID-19 and other atypical pneumonia (AP) by using only radiomics features. METHODS: In this retrospective study, 301 pneumonic lesions (150 ground-glass opacity [GGO], 52 crazy paving [CP], 99 consolidation) obtained from nonenhanced thorax CT scans of 74 AP (46 male and 28 female; 48.25±13.67 years) and 60 COVID-19 (39 male and 21 female; 48.01±20.38 years) patients were segmented manually by two independent radiologists, and Location, Size, Shape, and First- and Second-order radiomics features were calculated. RESULTS: Multiple parameters showed significant differences between AP and COVID-19-related GGOs and consolidations, although only the Range parameter was significantly different for CPs. Models developed by using the Bayesian information criterion (BIC) for the whole group of GGO and consolidation lesions predicted COVID-19 consolidation and AP GGO lesions with low accuracy (46.1% and 60.8%, respectively). Thus, instead of subjective classification, lesions were reclassified according to their skewness into positive skewness group (PSG, 78 AP and 71 COVID-19 lesions) and negative skewness group (NSG, 56 AP and 44 COVID-19 lesions), and group-specific models were created. The best AUC, accuracy, sensitivity, and specificity were respectively 0.774, 75.8%, 74.6%, and 76.9% among the PSG models and 0.907, 83%, 79.5%, and 85.7% for the NSG models. The best PSG model was also better at predicting NSG lesions smaller than 3 mL. Using an algorithm, 80% of COVID-19 and 81.1% of AP patients were correctly predicted. CONCLUSION: During periods of increasing AP, radiomics parameters may provide valuable data for the differential diagnosis of COVID-19.

COVID-19 mimics on chest CT: a pictorial review and radiologic guide.

Chest imaging is often used as a complementary tool in the evaluation of coronavirus disease 2019 (COVID-19) patients, helping physicians to augment their clinical suspicion. Despite not being diagnostic for COVID-19, chest CT may help clinicians to isolate high suspicion patients with suggestive imaging findings. However, COVID-19 findings on CT are also common to other pulmonary infections and non-infectious diseases, and radiologists and point-of-care physicians should be aware of possible mimickers. This state-of-the-art review goal is to summarize and illustrate possible etiologies that may have a similar pattern on chest CT as COVID-19. The review encompasses both infectious etiologies, such as non-COVID viral pneumonia, Mycoplasma pneumoniae, Pneumocystis jiroveci, and pulmonary granulomatous infectious, and non-infectious disorders, such as pulmonary embolism, fat embolism, cryptogenic organizing pneumonia, non-specific interstitial pneumonia, desquamative interstitial pneumonia, and acute and chronic eosinophilic pneumonia.

Presenting characteristics, comorbidities, and outcomes of patients coinfected with COVID-19 and Mycoplasma pneumoniae in the USA.

Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) is spreading at a rapid pace, and the World Health Organization declared it as pandemic on 11 March 2020. Mycoplasma pneumoniae is an "atypical" bacterial pathogen commonly known to cause respiratory illness in humans. The coinfection from SARS-CoV-2 and mycoplasma pneumonia is rarely reported in the literature to the best of our knowledge. We present a study in which 6 of 350 patients confirmed with COVID-19 were also diagnosed with M. pneumoniae infection. In this study, we described the clinical characteristics of patients with coinfection. Common symptoms at the onset of illness included fever (six [100%] patients); five (83.3%) patients had a cough, shortness of breath, and fatigue. The other symptoms were myalgia (66.6%), gastrointestinal symptoms (33.3%-50%), and altered mental status (16.7%). The laboratory parameters include lymphopenia, elevated erythrocyte sedimentation rate, C-reactive protein, lactate dehydrogenase, interleukin-6, serum ferritin, and D-dimer in all six (100%) patients. The chest X-ray at presentation showed bilateral infiltrates in all the patients (100%). We also described electrocardiogram findings, complications, and treatment during hospitalization in detail. One patient died during the hospital course.

[Early differential diagnosis between COVID-19 and mycoplasma pneumonia with chest CT scan].

OBJECTIVE: To early differentiate between coronavirus disease 2019 (COVID-19) and adult mycoplasma pneumonia with chest CT scan. METHODS: Twenty-six patients with COVID-19 and 21 patients with adult mycoplasma pneumonia confirmed with RT-PCR test were enrolled from Zibo First Hospital and Lanshan People's Hospital during December 1st 2019 and March 14th 2020. The early chest CT manifestations were analyzed and compared between the two groups. RESULTS: The interstitial changes with ground glass density shadow (GGO) were similar in two groups during first chest CT examination (P>0.05). There were more lung lobes involved on the first chest CT in COVID-19 patients, which were mostly distributed in the dorsal outer zone (23/26, 88.5%), and nearly half of them (12/26, 46.2%) were accompanied by crazy-paving sign; while the lesions in adult mycoplasma pneumonia patients were mostly distributed along the bronchi, and the bronchial wall was thickened (19/21, 90.5%), accompanied with tree buds / fog signs (19/21, 90.5%). The above CT signs were significantly different between the two kinds of pneumonia (all P<0.01). COVID-19 had a longer course compared with mycoplasma pneumonia, the disease peaks of COVID-19 patients was on day (10.5±3.8), while the disease on CT was almost absorbed on day (7.9±2.2) in adult mycoplasma pneumonia. The length of hospital stay in COVID-19 patients was significantly longer than that of mycoplasma pneumonia patients [(19.5±4.3) d vs (7.9±2.2) d, P<0.01]. CONCLUSIONS: The lesions of adult mycoplasma pneumonia are mostly distributed along the bronchi with tree buds/fog signs, while the lesions of COVID-19 are mainly distributed in the dorsal outer zone accompanied by crazy-paving sign, which can early distinguish two diseases.

A fully automatic deep learning system for COVID-19 diagnostic and prognostic analysis.

Coronavirus disease 2019 (COVID-19) has spread globally, and medical resources become insufficient in many regions. Fast diagnosis of COVID-19 and finding high-risk patients with worse prognosis for early prevention and medical resource optimisation is important. Here, we proposed a fully automatic deep learning system for COVID-19 diagnostic and prognostic analysis by routinely used computed tomography.We retrospectively collected 5372 patients with computed tomography images from seven cities or provinces. Firstly, 4106 patients with computed tomography images were used to pre-train the deep learning system, making it learn lung features. Following this, 1266 patients (924 with COVID-19 (471 had follow-up for >5 days) and 342 with other pneumonia) from six cities or provinces were enrolled to train and externally validate the performance of the deep learning system.In the four external validation sets, the deep learning system achieved good performance in identifying COVID-19 from other pneumonia (AUC 0.87 and 0.88, respectively) and viral pneumonia (AUC 0.86). Moreover, the deep learning system succeeded to stratify patients into high- and low-risk groups whose hospital-stay time had significant difference (p=0.013 and p=0.014, respectively). Without human assistance, the deep learning system automatically focused on abnormal areas that showed consistent characteristics with reported radiological findings.Deep learning provides a convenient tool for fast screening of COVID-19 and identifying potential high-risk patients, which may be helpful for medical resource optimisation and early prevention before patients show severe symptoms.