Thoracic Chest Computed Tomography Research Articles

Skeletal muscle quantity and quality evaluation in heart failure: comparing thoracic versus abdominopelvic CT approaches

Given the critical role of skeletal muscle in healthy aging, low muscle mass (myopenia) and quality (myosteatosis) can be used as predictors of poor functional and cardiometabolic outcomes. Myopenia is also a part of sarcopenia and malnutrition diagnostic criteria. However, there is limited evidence for using chest computed tomography (CT) to evaluate muscle health. We aimed to compare chest CT landmarks to the widely used L3 vertebra for single-slice skeletal muscle evaluation in patients with heart failure (HF). Patients admitted for acute decompensated HF between January 2017 and December 2018 were retrospectively analyzed. Body composition measurements were made on CT of the chest and abdomen/pelvis with or without contrast one month before discharge. Skeletal muscle index (SMI) and intermuscular adipose tissue percentage (IMAT%) were calculated at several thoracic levels (above the aortic arch, T8, and T12) and correlated to the widely used L3 level. A total of 200 patients were included, 89 (44.5%) female. The strongest correlation of thoracic SMI (for muscle quantity) and IMAT% (for muscle quality) with L3 was at the T12 level (r = 0.834, p < 0.001 and r = 0.757, p < 0.001, respectively). Cutoffs to identify low muscle mass for T12 SMI (derived from the lowest sex-stratified L3 SMI tertile) were 31.1 cm²/m² in men and 26.3 cm²/m² in women. SMI and IMAT% at T12 had excellent correlations with the widely used L3 level for muscle quantity and quality evaluation in patients with HF.

Residual radiological opacities correlate with disease outcomes in ICU-treated COVID-19.

Few studies consider both radiological and functional outcomes in COVID-19 survivors treated in the intensive care unit (ICU). We investigated clinical findings and pulmonary abnormalities on chest computed tomography (CT) and compared outcomes of severe versus mild-moderate acute respiratory distress syndrome (ARDS) on long-term follow-up. This longitudinal cohort study included 118 COVID-19 patients (median age, 58 years; 79% men). Thoracic CT scans were performed 4, 10, and 22 months after hospital discharge. Two independent blinded radiologists analyzed the 10 months scans and scored the radiology findings semi-quantitatively, as no/minor versus widespread opacities [low-radiology opacity grade (ROG) versus high-ROG]. ARDS severity was based on the PaO2/FiO2 ratio. The 6 min walk test (6MWT) was performed after 3 and 9 months, and lung diffusion capacity for carbon monoxide (DLCO) and lung volume measurement after 9 and 15 months. Dynamic spirometry was done at all time points. Residual symptoms and health-related quality-of-life (HRQL) were evaluated using validated questionnaires. At 10 months, most patients (81/118; 69%) were classified as high-ROG, of which 70% had severe ARDS during hospitalisation; 69% of those with mild-moderate ARDS also had high-ROG. Patients with high-ROG had longer ICU stay and lower PaO2/FiO2 during hospitalisation (p < 0.01). At 9 months follow-up, patients with high-ROG had smaller lung volumes as % of predicted values [mean (±CI): 80 (77-84) vs. 93 (88-98) (p < 0.001)], lower DLCO as % of predicted values [74 (70-78) vs. 87 (82-92) (p < 0.001)], lower oxygen saturation during 6MWT (p = 0.02), and a tendency to more severe dyspnoea (p = 0.07), but no difference was found in HRQL compared with no/minor ROG (p = 0.92). A higher opacity score was related to lower DLCO at follow-up (r = -0.48, p < 0.001, Spearman rank test). Severe ARDS patients had slightly more severe fatigue at 9 months compared to mild-moderate, but no differences in dyspnoea or lung function at follow-up. Fibrotic-like changes were found in 93% of patients examined with CT scans at 2 years (55/118; 47%). Severe ARDS could predict widespread opacities (ROG > 25%) in most patients at follow-up at 10 months (AUC 0.74). Residual radiological abnormalities in ICU-treated COVID-19 patients, evaluated for up to 2 years, relate to persisting symptoms and impaired lung function, demanding careful follow-up regardless of ARDS severity at hospitalisation.

CLASSIFIED COVID-19 BY DENSENET121-BASED DEEP TRANSFER LEARNING FROM CT-SCAN IMAGES

The COVID-19 disease, which has recently emerged and has been considered a worldwide pandemic, has had a significant impact on the lives of millions of people and has forced a substantial load on healthcare organizations. Numerous deep-learning models have been utilized for diagnosing coronaviruses from chest computed tomography (CT) images. However, in light of the limited availability of datasets on COVID-19, the pre-trained deep learning networks were used. The main objective of this research is to construct and develop an automated approach for the early detection and diagnosis of COVID-19 in thoracic CT images. This paper proposes the DDTL-COV model, a deep transfer learning model based on DenseNet121, to classify patients on CT scans as either COVID or non-COVID, utilizing weights obtained from the ImageNet dataset. Two datasets were used to train the DDTL-COV model: the SARS-CoV-2 CT-scan dataset and the COVID19-CT dataset. In the SARS-CoV-2 CT dataset, the model achieved a good accuracy of 99.6%. However, on the second dataset (COVID19-CT dataset), its performance shows an accuracy rate of 89%. These results show that the model performed better than alternative methods.

Feasibility of thoracic CT in assessing anemia for aplastic anemia patients undergoing allogeneic hematopoietic stem cell transplantation.

Anemia is an important clinical symptom for aplastic anemia (AA) patients who are suffered with peripheral pancytopenia. To evaluate the accuracy of diagnosing anemia with non-invasive chest computed tomography (CT) for AA patients. The CT attenuation of left ventricular (LV) cavity and interventricular septum (IVS) on unenhanced thoracic CT images of AA patients are retrospectively analyzed, including 84 AA patients in pre-transplant and 1-month (n = 82), 2-month (n = 72), 3-month (n = 75), 6-month (n = 74) and 12-month (n = 70) followed patients in post-transplant. The difference (IVS-LV) and ratio (LV/IVS) of the CT attenuation between LV cavity and interventricular septum are calculated. Serum hemoglobin is estimated within 24 hours of CT imaging. The CT attenuations of IVS-LV and LV/IVS are correlated with hemoglobin, and their variation tendency is analyzed during the treatment of a-HSCT. A receiver operating characteristic (ROC) curve analysis is then performed for the diagnosis of anemia. The CT attenuations of IVS-LV and LV/IVS well correlate with hemoglobin (r = -0.618 and 0.628, respectively, P < 0.001). The variation tendency of IVS-LV and LV/IVS is similar to that of hemoglobin with opposite directions during one-year follow-up of a-HSCT. When a threshold of CT attenuation of IVS-LV and LV/IVS is set at 11.5HU and 0.77, respectively, both the sensitivity and specificity in diagnosing anemia are good (74.7% and 73.8% in CT attenuation of IVS-LV; 77.4% and 70.4% in LV/LVS, respectively). Both CT attenuation of LV/IVS and IVS-LV had similar accuracy in diagnosing anemia for AA patients. The non-invasive chest CT can offer a new possibility to complementarily evaluate anemia for AA patients in the diagnostic radiology reports.

Importance of incidental coronary artery calcification in early diagnosis of obstructive coronary artery disease.

The early diagnosis of coronary artery disease (CAD) enables early intervention for the modifiable risk factors of the disease. Coronary artery calcification (CAC) detected incidentally on standard noncontrast chest computed tomography (CT) provides an opportunity for the early diagnosis of CAD. The purpose of this study was to demonstrate that CAC should be routinely reported when evaluating thoracic CT examinations. Routine reporting of CAC will contribute to the early diagnosis of CAD. The present study included 279 patients who underwent conventional coronary angiography (CAG) and CT within one month before undergoing CAG. The CAG and CT images of the patients were evaluated retrospectively. The levels of coronary artery stenosis were determined in reference to the CAG images. The CAC scores of the patients were calculated using the Weston method based on their chest CT images. The mean age of the patients was 63.2 ± 11.5 (range, 41-93) years, and 172 (61.6%) of them were men. The Weston score (WS) was 0 in 18.9% of the patients with obstructive CAD (OCAD), whereas it was ≥ 7 in 27.9% of patients. All patients with a WS of ≥ 7 had OCAD. All patients without luminal stenosis or < 50% stenosis had a WS of < 7. The CAC score is useful for the diagnosis of CAD and OCAD. If CAC is identified on standard noncontrast chest CT, it should be scored and reported accordingly. The WS can be used for CAC scoring.

Imaging in the Intensive Care Unit.

Radiology plays an important role in the management of the most seriously ill patients in the hospital. Over the years, continued advances in imaging technology have contributed to an improvement in patient care. However, even with such advances, the portable chest radiograph (CXR) remains one of the most commonly requested radiographic examinations. While they provide valuable information, CXRs remain relatively insensitive at revealing abnormalities and are often nonspecific. Chest computed tomography (CT) can display findings that are occult on CXR and is particularly useful at identifying and characterizing pleural effusions, detecting barotrauma including small pneumothoraces, distinguishing pneumonia from atelectasis, and revealing unsuspected or additional abnormalities which could result in increased morbidity and mortality if left untreated. CT pulmonary angiography is the modality of choice in the evaluation of pulmonary emboli which can complicate the hospital course of the ICU patient. This article will provide guidance for interpretation of CXR and thoracic CT images, discuss some of the invasive devices routinely used, and review the radiologic manifestations of common pathologic disease states encountered in ICU patients. In addition, imaging findings and complications of more specific clinical scenarios in which the incidence has increased in the ICU setting, such as patients who are immunocompromised, have interstitial lung disease, or COVID-19, will also be discussed. Communication between the radiologist and intensivist, particularly on complicated cases, is important to help increase diagnostic accuracy and leads to an improvement in the management of the most critically ill patients.

Comparison of Chest Computed Tomography Differential Diagnosis of Ground Glass Opacities in the COVID-19 Patients

Aim: To compare chest computed tomography (CT) diagnosis of ground glass opacities in the COVID-19 patients. Study design: Retrospective study. Place and duration of study: Department of Radiology, Ghulam Muhammad Mahar Medical College Sukkur from 1st July 2020 to 31 August 2021. Methodology: Fifty patients on differential CT diagnosis of ground glass opacities seen in COVID 19 patients were enrolled. Thoracic CT images by applying auto exposure-control settings and ranges of scan were done. The noise-index was kept as 12.3. Using helical 16 slice Alexion CT-Toshiba. Keeping a comparison with viral infection CT images a list of 7 signs which were positive for Covid CT scan were recorded. Peripheral lesions meant any lesions which effects peripheral area up to 3 to 4 cm lung periphery with/without having central dispersal. A hazy-opacity was termed as ground glass. Results: Mean age of the patients was 49.1±10.2 years with 27 (54%) males and 23(46%) females. Mix ground glass opacitites and consolidation were also the features of the CT imaging in coronavirus posisitve cases. Man-Whitney test results showed that combined-CT scoring had a SE value as 0.044 with a confidence interval between 0.756-0.927. Comparing the differential CT values within COVID and non COVID patients based on RT PCR results it was observed that posterior region lower lobe involvement was a feature of COVID-19 patients while crazy paving pattern and peripheral distribution was also seen in corona patients. Conclusion: Present study highlights that chest CT helps in differentitaing corona virus from other causes of pneumonias and grond glass opacities. Key words: Chest CT; Differential diagnosis; COVID-19; Ground glass opacities (GGO)

AB0808 Costovertebral, Costotransverse, Sternoclavicular and Manubriosternal Joint Involvement in Spondyloarthritis: Preliminary Results of Single Centre Experience

BackgroundCostovertebral (CV), costotransverse (CTr), sternoclavicular (SC) and manubriosternal (MS) joint involvements may be in spondyloarthritis (SpA). However, the radiological and clinical characteristics of these involvements need further assessment.ObjectivesIn this study, we aimed to determine the characteristics of CV, CTr, SC, and MS joint involvements in SpA patients.MethodsSpA patients (searched from the hospital data-processing system according to ICD-10 system to determine the patients having M07 (enteropathic arthropathies), M45 (ankylosing spondylitis (AS)), M46 (other inflammatory spondylopathies) and/or M48 (other spondylopathies) codes) who have been requested a chest computed tomography (CT) for any reason between January 2010 and December 2020 included in this retrospective cross-sectional analysis. Chest CT reports were reviewed, and any CV, CTr, SC or MS joint involvement attributed to SpA accepted as positive CT for SpA involvement. Demographic/clinic data including smoking status, disease characteristics (duration, drug history, Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), Bath Ankylosing Spondylitis Metrology Index (BASMI) and other related parameters) at the time of CT scan were collected and analyzed.ResultsAfter the confirmation of the diagnoses among 1700 patients, total of 339 SpA patients were included in the study. 244 were diagnosed with ankylosing spondylitis (AS), 66 with psoriatic arthritis (PsA), 14 with enteropathic arthritis, 13 with peripheral SpA, and 2 with non-radiographic axial SpA. According to CT reports, the number of patients with inflammatory changes in any of the CV, CTr, SC or MS joints on thoracic CT were 44. Of these patients, 42 were diagnosed with AS (95,5%), 1 with PsA (2,3%), and 1 (2,3%) with peripheral SpA. 39 CV, 9 SC, 5 MS joints, and 3 CTr joints were affected. Joint involvements (17%) were higher in patients with AS than in other patient groups (p&lt;0.001). In the chest CT positive group, Schober’s test, and chest expansion were significantly lower than chest CT negative group. Smoking history, C-reactive protein (CRP) value at diagnosis and at CT evaluation, and disease duration were associated with joint involvement. Demographic and clinical characteristics of patients with and without chest CT involvement were shown in Table 1.Table 1.Demographic and clinical characteristics of patients with and without involvement on chest CTData are presented as median (25-75 percentile) or number (%)CT; computerized tomography, BMI; body mass index, ESR; erythrocyte sedimentation rate, CRP; C-reactive protein, BASDAI; Bath Ankylosing Spondylitis. Disease Activity Index, BASMI; Bath Ankylosing Spondylitis Metrology Index, BASFI; Bath Ankylosing Spondylitis Functional IndexConclusionIn this study, we have found that CV, CTr, SC, and MS joint involvement on thorax CT were more common in men and smokers and were associated with disease duration and appears to affect chest expansion and Schober’s tests. In previous studies, CV and CTr joint involvement in AS patients with chest pain were 80-82% and 60%, respectively [1,2]. Our study showed that 17% of AS patients had involvement regardless of pain symptoms. As this is the preliminary data and the CTs will be re-evaluated by our radiologist, we will be capable to report more unbiased results.

Performance of a qualitative rapid chromatographic immunoassay to diagnose COVID-19 in patients in a middle-income country

We evaluated a rapid chromatographic immunoassay (IgG/IgM antibodies) and an ELISA assay to diagnose COVID-19 in patient sat two Brazilian hospitals. A total of 122 subjects with COVID-19 were included: 106 SARS-COV-2 RT-PCR-positive patients and 16 RT-PCR-negative patients with symptoms and chest computed tomography (CT) consistent with COVID-19. Ninety-six historical blood donation samples were used as controls. Demographic and clinical characteristics were retrieved from electronic records. Sensitivity and specificity were calculated, as were their 95% binomial confidence intervals using the Clopper-Pearson method. All analyses were performed in R version 3.6.3. The sensitivity of the chromatographic immunoassay in all RT-PCR-positive patients, irrespective of the timing of symptom onset, was 85.8% (95% binomial CI 77.7% to 91.9%). This increased with time after symptom onset, and at >14 days was 94.9% (85.9% to 98.9%). The specificity was 100% (96.4% to 100%). 15/16 (94%) RT- PCR-negative cases tested positive. The most frequent comorbidities were hypertension and diabetes mellitus and the most frequent symptoms were fever, cough, and dyspnea. All RT-PCR-negative patients had pneumonia. The most frequent thoracic CT findings were ground glass changes (n = 11, 68%), which were bilateral in 9 (56%) patients, and diffuse reticulonodular infiltrates (n = 5, 31%). The COVID-19 rapid chromatographic immunoassay evaluated in this study had a high sensitivity and specificity using plasma, particularly after 14 days from symptom onset. ELISA and qualitative rapid chromatographic immunoassays can be used for the diagnosis of RT-PCR-negative patients.

Hepatic steatosis as an independent risk factor for severe disease in patients with COVID-19: A computed tomography study.

Background and AimHepatic steatosis (HS) is associated with diabetes, hypertension, and obesity, comorbidities recently related to COVID‐19 severity. Here, we assessed if tomographic HS is also a risk factor for severe COVID‐19 pneumonia.MethodsWe included 213 patients with a positive real time polymerase chain reaction (RT‐PCR) test and chest computed tomography (CT) from an out‐hospital facility and a hospital. We obtained information on demographics; weight; height; smoking history; diabetes; hypertension; and cardiovascular, lung, and renal disease. Two radiologists scored the CO‐RADs system (COVID‐19 Reporting and Data System) (1 = normal, 2 = inconsistent, 3–4 = indeterminate, and 5 = typical findings) and the chest CT severity index (≥20 of 40 was considered severe disease). They evaluated the liver‐to‐spleen ratio (CTL/S) and defined tomographic steatosis as a CTL/S index ≤0.9. We used descriptive statistics, χ2 and t student tests, logistic regression, and reported odds ratio (OR) with 95% confidence interval (CI).ResultsOf the patients, 61% were men, with a mean age of 51.2 years, 48.3% were CO‐RADs 1 and 51.7% CO‐RADs 2–5. Severe tomographic disease was present in 103 patients (48.4%), all CO‐RADs 5. This group was older; mostly men; and with a higher prevalence of obesity, hypertension, diabetes, and HS (69.9 vs 29%). On multivariate analysis, age (OR 1.058, 95% CI 1.03–1.086, P < 0.0001), male gender (OR 1.9, 95% CI 1.03–3.8, P = 0.04), and HS (OR 4.9, 95% CI 2.4–9.7, P < 0.0001) remained associated.ConclusionHS was independently associated with severe COVID pneumonia. The physiopathological explanation of this finding remains to be elucidated. CTL/S should be routinely measured in thoracic CT scans in patients with COVID‐19 pneumonia.

Thoracic computerised tomography scans in one hundred eighteen orthopaedic patients during the COVID-19 pandemic: identification of chest lesions; added values; help in managing patients; burden on the computerised tomography scan department.

PurposeBased on the recent literature, chest computed tomography (CT) examination could aid for management of patients during COVID-19 pandemic. However, the role of chest CT in management of COVID-19 patients is not exactly the same for medical or surgical specialties. In orthopaedic or trauma emergency, abdomen, pelvis, cervical, dorsal, and lumbar spine CT are performed to investigate patients; the result is a thoracic CT scan incorporating usually the thorax; however, information about lung parenchyma can be obtained on this thorax CT, and manifestations of COVID-19 can be diagnosed. The objective of our study was to evaluate this role in orthopedic patients to familiarize orthopaedists with the value and limits of thoracic CT in orthopaedic surgery.Materials and methodsAmong the 1397 chest CT scans performed during the pandemic period from 1 March 2020 to 10 May 2020, in two centres with orthopaedic surgery, we selected all the 118 thoracic or chest CT performed for patients who presented to the Emergency Department of the hospital with a diagnosis of trauma for orthopaedic surgical treatment. Thirty-nine of these 118 patients were tested with PCR for the diagnosis of COVID-19 infection. Depending on clinical status (symptomatic or non-symptomatic), the information useful for the orthopaedist surgeon and obtained from the Chest CT scan according to the result of the PCR (gold standard) was graded from 0 (no or low value) to 3 (high value). The potential risks of chest CT as exposure to radiation, and specific pathway were analyzed and discussed. A group of patients treated during a previous similar period (1 March 2018 to 15 April 2018) was used as control for evaluation of the increase of CT scanning during the COVID-19 pandemic.ResultsAmong the 118 patients with chest CT, there were 16 patients with positive COVID-19 chest CT findings, and 102 patients with negative chest CT scan. With PCR results as reference, the sensitivity, specificity, positive predictive value of chest CT in indicating COVID-19 infection were 81%, 93%, and 86%, respectively (p = 0.001). A useful information for the orthopaedic surgeon (graded as 1 for 71 cases, as 2 for 5 cases, and as 3 for 11 cases) was obtained from 118 chest CT scans for 87 (74%) patients, while the CT was no value in 30 (25%) cases, and negative value in one (1%) case. Roughly 20% of the total number of CT scanner performed over the pandemic period was dedicated to COVID-19, but only 2% were for orthopaedic or trauma patients. However, this was ten times higher than during the previous control period of comparison.ConclusionAlthough extremely valuable for surgery management, these results should not be overstated. The CT findings studied are not specific for COVID-19, and the positive predictive value of CT will be low unless disease prevalence is high, which was the case during this period.

CT features of SARS-CoV-2 pneumonia according to clinical presentation: a retrospective analysis of 120 consecutive patients from Wuhan city.

ObjectivesTo characterize the chest computed tomography (CT) findings of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) according to clinical severity. We compared the CT features of common cases and severe cases, symptomatic patients and asymptomatic patients, and febrile and afebrile patients.MethodsThis was a retrospective analysis of the clinical and thoracic CT features of 120 consecutive patients with confirmed SARS-CoV-2 pneumonia admitted to a tertiary university hospital between January 10 and February 10, 2020, in Wuhan city, China.ResultsOn admission, the patients generally complained of fever, cough, shortness of breath, and myalgia or fatigue, with diarrhea often present in severe cases. Severe patients were 20 years older on average and had comorbidities and an elevated lactate dehydrogenase (LDH) level. There were no differences in the CT findings between asymptomatic and symptomatic common type patients or between afebrile and febrile patients, defined according to Chinese National Health Commission guidelines.ConclusionsThe clinical and CT features at admission may enable clinicians to promptly evaluate the prognosis of patients with SARS-CoV-2 pneumonia. Clinicians should be aware that clinically silent cases may present with CT features similar to those of symptomatic common patients.Key Points • The clinical features and predominant patterns of abnormalities on CT for asymptomatic, typic common, and severe cases were summarized. These findings may help clinicians to identify severe patients quickly at admission. • Clinicians should be cautious that CT findings of afebrile/asymptomatic patients are not better than the findings of other types of patients. These patients should also be quarantined. • The use of chest CT as the main screening method in epidemic areas is recommended. Electronic supplementary materialThe online version of this article (10.1007/s00330-020-06854-1) contains supplementary material, which is available to authorized users.

Lung Nodule Detection from Feature Engineering to Deep Learning in Thoracic CT Images: a Comprehensive Review.

This paper presents a systematic review of the literature focused on the lung nodule detection in chest computed tomography (CT) images. Manual detection of lung nodules by the radiologist is a sequential and time-consuming process. The detection is subjective and depends on the radiologist's experiences. Owing to the variation in shapes and appearances of a lung nodule, it is very difficult to identify the proper location of the nodule from a huge number of slices generated by the CT scanner. Small nodules (< 10mm in diameter) may be missed by this manual detection process. Therefore, computer-aided diagnosis (CAD) system acts as a "second opinion" for the radiologists, by making final decision quickly with higher accuracy and greater confidence. The goal of this survey work is to present the current state of the artworks and their progress towards lung nodule detection to the researchers and readers in this domain. This review paper has covered the published works from 2009 to April 2018. Different nodule detection approaches are described elaborately in this work. Recently, it is observed that deep learning (DL)-based approaches are applied extensively for nodule detection and characterization. Therefore, emphasis has been given to convolutional neural network (CNN)-based DL approaches by describing different CNN-based networks.

Comparison of chest CT images of patients with acute respiratory distress syndrome and acute cardiogenic pulmonary edema

Objective To compare the chest computed tomography (CT) images of patients with acute respiratory distress syndrome (ARDS) and acute cardiogenic pulmonary edema (ACPE). Methods 30 cases of patients with ARDS and 30 cases of patients with ACPE from August 2016 to August 2018 were randomly selected. All patients underwent thoracic CT examination. The distributions of grinded glass density shadow (GGO) in the lungs and changes of extrapulmonary structure were statistically analyzed. Results The differences of distributions of GGO in the upper lobe, middle lobe, lower lobe and diffuse distribution between patients with ARDS and ACPE were not significant [16.7%(5/30), 0, 16.7%(5/30), 66.7%(20/30) vs 20.0%(6/30), 0, 20.0%(6/30), 60.0%(18/30), P>0.05]. The distributions of GGO in center lung [13.3%(4/30)] was significantly lower than patients with ACPE [46.7%(14/30)] (P 0.05]. The small boundary opacity of patients with ARDS [76.7%(23/30)] was significantly higher than patients with ACPE [26.7%(8/30)] (P<0.05), while the bronchial blood bundle thickening, pulmonary vascular shadow thickening, cardiac shadow enlargement, pericardial effusion were significantly lower than patients with ACPE [33.3% (10/30), 16.7%(5/30), 30.0%(9/30), 23.3%(7/30) vs 23.3%(7/30), 100.0%(30/30), 70.0%(21/30), 73.3%(22/30), 53.3%(16/30), P<0.05]. Conclusions The chest CT images of patients with ARDS and ACPE are different, mainly manifestated in small boundary opacity, bronchial blood bundle thickening, pulmonary vascular opacity, cardiac opacity, pericardial effusion. Key words: Respiratory distress syndrome, adult; Pulmonary edema; Edema, cardiac; Radiography, thoracic

Lesion detection on a combined “All-in-One” window compared to conventional window settings in thoracic oncology chest CT examinations

PurposeThe purpose of this study was to investigate if lesion detection using a single “All-in-One” (AIO) window was non-inferior to lesion detection on conventional window settings in thoracic oncology chest computed tomography (CT) examinations. Materials and methodsIn a retrospective study, 50 consecutive chest CT examinations of 50 patients (31 men, 19 women; mean age 64±10 [SD] years, range: 35–82 years) containing 417 lesions, were reviewed by 6 radiologists, subdivided into 2 groups of 3 radiologists each, with similar levels of expertise in each group (senior staff member, junior staff member and radiology resident). All examinations were reviewed in conventional or AIO window settings by one of the groups. A ‘lesion’ was defined as any abnormality seen on the chest CT examination, including both benign and malignant lesions, findings in chest and upper abdomen, and measurable and non-measurable disease. Lesions were listed as ‘missed’ when they were not seen by at least two out of three observers. F-tests were used to evaluate the significance of the variables of interest within a mixed model framework and kappa statistics to report interobserver agreement. ResultsOn a reader level, 54/417 lesions (12.9%) were not detected by the senior staff member reading the studies in conventional window settings and 45/417 (10.8%) by the senior staff member reading the AIO images. For the junior staff member and radiology resident this was respectively 55/417 (13.2%) and 67/417 (16.1%) for the conventional window settings and 43/417 (10.3%) and 61/417 (14.6%) for the AIO window. On a lesion level, 68/417 (16.3%) were defined as ‘missed’ lesions (lesions not detected by at least 2 readers): 21/68 (30.9%) on the AIO-window, 30/68 (44.1%) on conventional views and 17/68 (25.0%) on both views. The use of the AIO window did not result in an increase of missed lesions (P>0.99). Interobserver agreement in both groups was similar (P=0.46). Regarding lesions that were categorized as ‘missed’ on the AIO window or on conventional window settings, there was no effect of location (chest or upper abdomen) (P=0.35), window (P=0.97) and organ (P=0.98). ConclusionsA single AIO-window is non-inferior to multiple conventional window settings for lesion detection on chest CT examinations in thoracic oncology patients.

Improving Quality of Chest Computed Tomography for Evaluation of Pediatric Malignancies.

Introduction:Atelectasis is a problem in sedated pediatric patients undergoing cross-sectional imaging, impairing the ability to accurately interpret chest computed tomography (CT) imaging for the presence of malignancy, often leading to additional maneuvers and/or repeat imaging with additional radiation exposure.Methods:A quality improvement team established a best-practice protocol to improve the quality of thoracic CT imaging in young patients with suspected primary or metastatic pulmonary malignancy. The specific aim was to increase the percentage of chest CT scans obtained for the evaluation of pulmonary nodules with acceptable atelectasis scores (0–1) in patients aged 0–5 years with malignancy, from a baseline of 45% to a goal of 75%.Results:A retrospective cohort consisted of 94 patients undergoing chest CT between February 2014 and January 2015 before protocol implementation. The prospective cohort included 195 patients imaged between February 2015 and April 2018. The baseline percentage of CT scans that were scored 0 or 1 on the atelectasis scale was 44.7%, which improved to 75% with protocol implementation. The mean atelectasis score improved from 1.79 (±0.14) to 0.7 (±0.09). Sedation incidence decreased substantially from 73.2% to 26.5% during the study period.Conclusions:Using quality improvement methodology including standardization of care, the percentage of children with atelectasis scores of 0–1 undergoing cross-sectional thoracic imaging improved from 45% to 75%. Also, eliminating the need for sedation in these patients has further improved image quality, potentially allowing for optimal detection of smaller nodules, and minimizing morbidity.

Automatic multiorgan segmentation in thorax CT images using U‐net‐GAN

Accurate and timely organs-at-risk (OARs) segmentation is key to efficient and high-quality radiation therapy planning. The purpose of this work is to develop a deep learning-based method to automatically segment multiple thoracic OARs on chest computed tomography (CT) for radiotherapy treatment planning. We propose an adversarial training strategy to train deep neural networks for the segmentation of multiple organs on thoracic CT images. The proposed design of adversarial networks, called U-Net-generative adversarial network (U-Net-GAN), jointly trains a set of U-Nets as generators and fully convolutional networks (FCNs) as discriminators. Specifically, the generator, composed of U-Net, produces an image segmentation map of multiple organs by an end-to-end mapping learned from CT image to multiorgan-segmented OARs. The discriminator, structured as an FCN, discriminates between the ground truth and segmented OARs produced by the generator. The generator and discriminator compete against each other in an adversarial learning process to produce the optimal segmentation map of multiple organs. Our segmentation results were compared with manually segmented OARs (ground truth) for quantitative evaluations in geometric difference, as well as dosimetric performance by investigating the dose-volume histogram in 20 stereotactic body radiation therapy (SBRT) lung plans. This segmentation technique was applied to delineate the left and right lungs, spinal cord, esophagus, and heart using 35 patients' chest CTs. The averaged dice similarity coefficient for the above five OARs are 0.97, 0.97, 0.90, 0.75, and 0.87, respectively. The mean surface distance of the five OARs obtained with proposed method ranges between 0.4 and 1.5mm on average among all 35 patients. The mean dose differences on the 20 SBRT lung plans are -0.001 to 0.155Gy for the five OARs. We have investigated a novel deep learning-based approach with a GAN strategy to segment multiple OARs in the thorax using chest CT images and demonstrated its feasibility and reliability. This is a potentially valuable method for improving the efficiency of chest radiotherapy treatment planning.

Improvement of fully automated airway segmentation on volumetric computed tomographic images using a 2.5 dimensional convolutional neural net.

We propose a novel airway segmentation method in volumetric chest computed tomography (CT) and evaluate its performance on multiple datasets. The segmentation is performed voxel-by-voxel by a 2.5D convolutional neural net (2.5D CNN) trained in a supervised manner. To enhance the accuracy of the segmented airway tree, we simultaneously took three adjacent slices in each of the orthogonal directions including axial, sagittal, and coronal and fine-tuned the parameters that influence the tree length and the number of leakage. The gold standard of airway segmentation was generated by a semi-automated method using AVIEW™. The 2.5D CNN was trained and evaluated on a subset of inspiratory thoracic CT scans taken from the Korean obstructive lung disease study, which includes normal subjects and chronic obstructive pulmonary disease patients. The reliability and further practicality of our proposed method was demonstrated in multiple datasets. In eight test datasets collected by the same imaging protocol, the percentage detected tree length, false positive rate, and Dice similarity coefficient of our method were 92.16%, 7.74%, and 0.8997 ± 0.0892, respectively. In 20 test datasets of the EXACT'09 challenge, the percentage detected tree length was 60.1% and the false positive rate was 4.56%. Our fully automated (end-to-end) segmentation method could be applied in radiologic practice.

Automatic airway-artery analysis on lung CT to quantify airway wall thickening and bronchiectasis.

Bronchiectasis and airway wall thickening are commonly assessed in computed tomography (CT) by comparing the airway size with the size of the accompanying artery. Thus, in order to automate the quantification of bronchiectasis and wall thickening following a similar principle, there is a need for methods that automatically segment the airway and vascular trees, measure their size, and pair each airway branch with its accompanying artery. This paper combines and extends existing techniques to present a fully automated pipeline that, given a thoracic chest CT, segments, measures, and pairs airway branches with the accompanying artery, then quantifies airway wall thickening and bronchiectasis by measuring the wall-artery ratio (WAR) and lumen and outer wall airway-artery ratio (AAR). Measurements that do not use the artery size for normalization are also extracted, including wall area percentage (WAP), wall thickness ratio (WTR), and airway diameters. The method was thoroughly evaluated using 8000 manual annotations of airway-artery pairs from 24 full-inspiration pediatric CT scans (12 diseased and 12 controls). Limits of agreement between the automatically and manually measured diameters were comparable to interobserver limits of agreement. Differences in automatically obtained WAR, AAR, WAP, and WTR between bronchiectatic subjects and controls were similar as when manual annotations were used: WAR and outer AAR were significantly higher in the bronchiectatic subjects (p < 0.05), but lumen AAR, WAP, and WTR were not. Only measurements that use artery size for normalization led to significant differences between groups, highlighting the importance of airway-artery pairing. The fully automatic method presented in this paper could replace time-consuming manual annotations and visual scoring methods to quantify abnormal widening and thickening of airways.

Automated Lung Volumetry from Routine Thoracic CT Scans: How Reliable is the Result?

Today, lung volumes can be easily calculated from chest computed tomography (CT) scans. Modern postprocessing workstations allow automated volume measurement of data sets acquired. However, there are challenges in the use of lung volume as an indicator of pulmonary disease when it is obtained from routine CT. Intra-individual variation and methodologic aspects have to be considered. Our goal was to assess the reliability of volumetric measurements in routine CT lung scans. Forty adult cancer patients whose lungs were unaffected by the disease underwent routine chest CT scans in 3-month intervals, resulting in a total number of 302 chest CT scans. Lung volume was calculated by automatic volumetry software. On average of 7.2 CT scans were successfully evaluable per patient (range 2-15). Intra-individual changes were assessed. In the set of patients investigated, lung volume was approximately normally distributed, with a mean of 5283 cm(3) (standard deviation = 947 cm(3), skewness = -0.34, and curtosis = 0.16). Between different scans in one and the same patient the median intra-individual standard deviation in lung volume was 853 cm(3) (16% of the mean lung volume). Automatic lung segmentation of routine chest CT scans allows a technically stable estimation of lung volume. However, substantial intra-individual variations have to be considered. A median intra-individual deviation of 16% in lung volume between different routine scans was found.