Effective Screening and Texture Segmentation of Green Vegetation Cover Based on UAV Images

As an important part of the ecosystem, green vegetation coverage is crucial to people's sensory and mental health. Using reliable data sets to classify and identify the green vegetation cover on the land surface and explore its spatial distribution law can provide important reference for the work of regional ecosystem managers and urban planners. The optimization of effective screening methods for green vegetation coverage areas is an important requirement to measure the surface vegetation status. UAV aerial images feature high definition, large scale, small area and high up-to-dateness. However, at present, there are few studies based on the reliable UAV aerial image system to identify green vegetation cover and further explore its spatial changes. In this study, 701 residential neighborhoods in Beijing were taken as the research objects, and the green vegetation of 7,695 sample points was identified by UAV. The green vegetation coverage was measured, and the spatial distribution pattern of green vegetation in different land surface areas was quantitatively compared. The results show that the image processing method proposed in this paper can effectively detect the boundary of green vegetation cover area from UAV aerial images, the correlation of texture segmentation is good, and the segmentation performance is better than other methods. The distribution of green vegetation cover in the research target area is uneven, with 63.79% of the research area having relatively low (Level 2) and medium (Level 3) green vegetation coverage, which indicates that the green vegetation coverage area in the research area is insufficient to meet the needs of regional ecosystem development. The characteristics of green vegetation cover in 16 districts in the study area are different, showing different spatial distribution patterns;except Xicheng District, there are 211 points without landscape in the area covered by green vegetation in 15 districts. The results can provide support for urban land surface planning and management.

Two-stage hybrid network for segmentation of COVID-19 pneumonia lesions in CT images: a multicenter study.

COVID-19 has been spreading continuously since its outbreak, and the detection of its manifestations in the lung via chest computed tomography (CT) imaging is essential to investigate the diagnosis and prognosis of COVID-19 as an indispensable step. Automatic and accurate segmentation of infected lesions is highly required for fast and accurate diagnosis and further assessment of COVID-19 pneumonia. However, the two-dimensional methods generally neglect the intraslice context, while the three-dimensional methods usually have high GPU memory consumption and calculation cost. To address these limitations, we propose a two-stage hybrid UNet to automatically segment infected regions, which is evaluated on the multicenter data obtained from seven hospitals. Moreover, we train a 3D-ResNet for COVID-19 pneumonia screening. In segmentation tasks, the Dice coefficient reaches 97.23% for lung segmentation and 84.58% for lesion segmentation. In classification tasks, our model can identify COVID-19 pneumonia with an area under the receiver-operating characteristic curve value of 0.92, an accuracy of 92.44%, a sensitivity of 93.94%, and a specificity of 92.45%. In comparison with other state-of-the-art methods, the proposed approach could be implemented as an efficient assisting tool for radiologists in COVID-19 diagnosis from CT images.

Prognostic Value of Coronary CT Angiography-Derived Fractional Flow Reserve in Non-obstructive Coronary Artery Disease: A Prospective Multicenter Observational Study

Coronary artery disease (CAD) is a major contributor to morbidity and mortality worldwide. Myocardial ischemia may occur in patients with normal or non-obstructive CAD on invasive coronary angiography (ICA). The comprehensive evaluation of coronary CT angiography (CCTA) integrated with fractional flow reserve derived from CCTA (CT-FFR) to CAD may be essential to improve the outcomes of patients with non-obstructive CAD. China CT-FFR Study-2 (ChiCTR2000031410) is a large-scale prospective, observational study in 29 medical centers in China. The primary purpose is to uncover the relationship between the CCTA findings (including CT-FFR) and the outcome of patients with non-obstructive CAD. At least 10,000 patients with non-obstructive CAD but without previous revascularization will be enrolled. A 5-year follow-up will be performed. The primary endpoint is the occurrence of major adverse cardiovascular events (MACE), including all-cause mortality, non-fatal myocardial infarct, unplanned revascularization, and hospitalization for unstable angina. Clinical characteristics, laboratory and imaging examination results will be collected to analyze their prognostic value.

Elevated Extracellular Volume Fraction and Reduced Global Longitudinal Strains in Participants Recovered from COVID-19 without Clinical Cardiac Findings.

Background It is unknown if there are cardiac abnormalities in persons who have recovered from coronavirus disease 2019 (COVID-19) without cardiac symptoms or in those who have normal biomarkers and normal electrocardiograms. Purpose To evaluate cardiac involvement in participants who had recovered from COVID-19 without clinical evidence of cardiac involvement by using cardiac MRI. Materials and Methods This prospective observational cohort study included 40 participants who had recovered from COVID-19 with moderate (n = 24) or severe (n = 16) pneumonia and who had no cardiovascular medical history, were without cardiac symptoms, had normal electrocardiograms, had normal serologic cardiac enzyme levels, and had been discharged for more than 90 days between May and September 2020. Demographic characteristics were recorded, serum cardiac enzyme levels were measured, and cardiac MRI was performed. Cardiac function, native T1, extracellular volume fraction (ECV), and two-dimensional (2D) strain were quantitatively evaluated and compared with values in control subjects (n = 25). Comparisons among the three groups were performed by using one-way analysis of variance with Bonferroni-corrected post hoc comparisons (for normal distribution) or Kruskal-Wallis tests with post hoc pairwise comparisons (for nonnormal distribution). Results Forty participants (mean age, 54 years ± 12 [standard deviation]; 24 men) were enrolled; participants had a mean time between admission and cardiac MRI of 158 days ± 18 and between discharge and cardiac MRI examination of 124 days ± 17. There were no left or right ventricular size or functional differences between participants who had recovered from COVID-19 and healthy control subjects. Only one (3%) participant had positive late gadolinium enhancement located at the mid inferior wall. Global ECV values were elevated in participants who had recovered from COVID-19 with moderate or severe pneumonia compared with those in healthy control subjects (median ECV, 29.7% vs 31.4% vs 25.0%, respectively; interquartile range, 28.0%-32.9% vs 29.3%-34.0% vs 23.7%-26.0%, respectively; P < .001 for both). The 2D global left ventricular longitudinal strain was reduced in both groups of participants (moderate COVID-19 group, -12.5% [interquartile range, -15.5% to -10.7%]; severe COVID-19 group, -12.5% [interquartile range, -15.4% to -8.7%]) compared with the healthy control group (-15.4% [interquartile range, -17.6% to -14.6%]) (P = .002 and P = .001, respectively). Conclusion Cardiac MRI myocardial tissue and strain imaging parameters suggest that a proportion of participants who had recovered from COVID-19 had subclinical myocardial abnormalities detectable months after recovery. © RSNA, 2021 Online supplemental material is available for this article.

Classification of Severe and Critical Covid-19 Using Deep Learning and Radiomics.

OBJECTIVE: The coronavirus disease 2019 (COVID-19) is rapidly spreading inside China and internationally. We aimed to construct a model integrating information from radiomics and deep learning (DL) features to discriminate critical cases from severe cases of COVID-19 using computed tomography (CT) images. METHODS: We retrospectively enrolled 217 patients from three centers in China, including 82 patients with severe disease and 135 with critical disease. Patients were randomly divided into a training cohort (n = 174) and a test cohort (n = 43). We extracted 102 3-dimensional radiomic features from automatically segmented lung volume and selected the significant features. We also developed a 3-dimensional DL network based on center-cropped slices. Using multivariable logistic regression, we then created a merged model based on significant radiomic features and DL scores. We employed the area under the receiver operating characteristic curve (AUC) to evaluate the model's performance. We then conducted cross validation, stratified analysis, survival analysis, and decision curve analysis to evaluate the robustness of our method. RESULTS: The merged model can distinguish critical patients with AUCs of 0.909 (95% confidence interval [CI]: 0.859-0.952) and 0.861 (95% CI: 0.753-0.968) in the training and test cohorts, respectively. Stratified analysis indicated that our model was not affected by sex, age, or chronic disease. Moreover, the results of the merged model showed a strong correlation with patient outcomes. SIGNIFICANCE: A model combining radiomic and DL features of the lung could help distinguish critical cases from severe cases of COVID-19.

Timely Diagnosis and Treatment Shortens the Time to Resolution of Coronavirus Disease (COVID-19) Pneumonia and Lowers the Highest and Last CT Scores From Sequential Chest CT.

OBJECTIVE. This study aims to assess correlations of the time from symptom onset to diagnosis and treatment with the time to disease resolution and CT scores as based on findings from sequential chest CT examinations. MATERIALS AND METHODS. Thirty patients with coronavirus disease (COVID-19) confirmed by reverse transcription-polymerase chain reaction analysis underwent chest CT examinations. Five patients who did not have positive CT findings or who had not yet fulfilled criteria for discharge from the hospital were excluded. CT scores were determined according to CT findings and lung involvement. The time from symptom onset to diagnosis and treatment was recorded for each patient, and on the basis of this information, patients with COVID-19 were divided into group 1 (patients for whom this interval was ≤ 3 days) and group 2 (those for whom this interval was > 3 days). The CT scores for each group were fitted using a Lorentzian line-shape curve to show the variation tendency during treatment. The differences in age, sex, and last CT scores determined before discharge between the two groups were analyzed, and correlations of the time from symptom onset to diagnosis and treatment with the time to disease resolution as well as with the highest CT score also underwent statistical analysis. RESULTS. A total of 25 subjects were enrolled in the study. The fitted tendency curves for group 1 and group 2 were significantly different, with peak points showing that the estimated highest CT score was 10 and 16 for each group, respectively, and the time to disease resolution was 6 and 13 days, respectively. The Mann-Whitney test showed that the last CT scores were lower for group 1 than for group 2 (p = 0.025), although the chi-square test found no difference in age and sex between the groups. The time from symptom onset to diagnosis and treatment had a positive correlation with the time to disease resolution (r = 0.93; p = 0.000) as well as with the highest CT score (r = 0.83; p = 0.006). CONCLUSION. Timely diagnosis and treatment are key to providing a better prognosis for patients with COVID-19.

Rapid identification of COVID-19 severity in CT scans through classification of deep features.

BACKGROUND: Chest CT is used for the assessment of the severity of patients infected with novel coronavirus 2019 (COVID-19). We collected chest CT scans of 202 patients diagnosed with the COVID-19, and try to develop a rapid, accurate and automatic tool for severity screening follow-up therapeutic treatment. METHODS: A total of 729 2D axial plan slices with 246 severe cases and 483 non-severe cases were employed in this study. By taking the advantages of the pre-trained deep neural network, four pre-trained off-the-shelf deep models (Inception-V3, ResNet-50, ResNet-101, DenseNet-201) were exploited to extract the features from these CT scans. These features are then fed to multiple classifiers (linear discriminant, linear SVM, cubic SVM, KNN and Adaboost decision tree) to identify the severe and non-severe COVID-19 cases. Three validation strategies (holdout validation, tenfold cross-validation and leave-one-out) are employed to validate the feasibility of proposed pipelines. RESULTS AND CONCLUSION: The experimental results demonstrate that classification of the features from pre-trained deep models shows the promising application in COVID-19 severity screening, whereas the DenseNet-201 with cubic SVM model achieved the best performance. Specifically, it achieved the highest severity classification accuracy of 95.20% and 95.34% for tenfold cross-validation and leave-one-out, respectively. The established pipeline was able to achieve a rapid and accurate identification of the severity of COVID-19. This may assist the physicians to make more efficient and reliable decisions.

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.