Detection Of Ground-glass Opacity Research Articles

Detection of Ground Glass Opacities in COVID-19 Lung CT Images using Frangi Multiscale Vesselness Measure

In this work, an automated technique for the quick and accurate detection of Ground Glass Opacities (GGO) in chest CT images of COVID-19 pneumonia patients is presented. The method uses mathematical morphology-based methods and Otsu's thresholding during the segmentation of GGO regions in order to extract the lung fields. The program uses the Frangi Multiscale Vesselness Measure to identify and remove these structures based on the anatomical features of bronchioles. Using a dataset of 155 lung CT images, the study outperformed previous algorithms in terms of sensitivity, specificity, and accuracy, all exceeding 97.22%. Radiologists can be helped by this automated screening method to quickly find GGOs in lung CT scans.

Optimization Ground Glass Opacities (GGO) Detection Using Multipixel Interpolation Techniques

Ground Glass Opacities (GGO) are a picture of abnormal lung conditions characterized by white or gray areas. This picture of GGO in the lungs could previously be detected based on the results of medical examinations such as Computerized Tomography (CT scan) and Magnetic Resonance Imaging (MRI) images of patients suffering from Covid-19. However, from the results of the examination, it can be seen that the CT scan and MRI images still have a noise level that is too high, causing difficulties in describing the distribution pattern of the GGO itself. The purpose of this study was to optimize the detection of GGO on MRI images using the Multipixel Interpolation technique. The detection process adopts several stages including image preprocessing, edge detection process, and gradient morphological segmentation. Image preprocessing is done to remove noise and improve the MRI input image. The edge detection process is carried out to detect lung organs automatically using the Canny method which is optimized with the multipixel interpolation technique. The final stage of the research is the segmentation process using a gradient morphology technique to see the spread of GGO in patients with Covid-19 contained in the MRI image. The results of this study present an overview of the GGO pattern with fairly good results. The results of the GGO pattern description will also measure the level of spread to see the severity of pneumonia. Based on the results presented, this research is useful as an alternative solution in the process of diagnosis and treatment of Covid-19 patients.

Ground glass opacity detection and segmentation using CT images: an image statistics framework

Lung cancer is one of the most profound causes of cancer-related deaths in the world. Early detection is known to significantly improve the chances of survival. Several detection and diagnostic methods are used for this purpose. CT is one of the most widely used non-invasive medical imaging modalities in this domain. The biggest challenge faced by radiologists in this case is detection and diagnosis of cancerous lung nodules. The growth of ground glass opacity (GGO) lesions is an indication of malignancy. However, GGO is difficult to capture for physicians as it manifests in the form of tiny, faint shadows. This research paper proposes an approach for aiding GGO identification in CT lung images for improved lung cancer prognosis. In the proposed approach, morphological reconstruction is used for segmentation. Once the region of interest (ROI) is extracted, statistical analysis using mean, standard deviation, variance, entropy, skewness, kurtosis, minimum grey scale value, maximum grey scale value and range is performed. The same statistical measures are determined for normal lung and distribution plot is drawn for comparison. It is observed that maximum grey-scale value demonstrates minimum overlap of approximately 7.4%. To reduce this, a joint feature by summing values of feature mean, skewness, and maximum grey-scale value was used. This approach reduced the overlap to approximately 1.32%. Lastly, ANN was used for classification of GGO and non-GGO lung tissue with an achieved accuracy of 99.5%.

Semiquantitative CT imaging as a tool in improving detection of ground glass patches in patients with COVID-19 pneumonia and for better follow-up

BackgroundCoronavirus disease 2019 (COVID-19) is a global pandemic caused by severe acute respiratory syndrome coronavirus 2 influencing millions of people worldwide. It has clinical symptoms going from mild symptoms in about 80% of patients to a case mortality rate of about 2% in hospitalized patients associated with radiologic findings at chest CT which is showing multifocal bilateral ground glass opacities (GGO) and consolidative patches with subpleural and peri-bronchovascular predominant distribution. The role of chest CT in COVID-19 is very crucial, so this study hypothesized that increasing the accuracy and rapidity of CT in the detection of COVID-19-related pneumonia will offer rapid management and intervention of affected cases and gain better outcomes. The aim of this study is to offer and assess the ability of a software computer program in helping the radiologists in rapid detection of COVID-19 pneumonic criteria.ResultsThis cross-sectional study involved 73 patients with clinical symptoms and real-time polymerase chain reaction test positive results diagnosed as COVID-19. They were referred to perform chest CT; their CT images were sent to a separate workstation to be automated and processed through the COVID-19 detector, and compared the finding of the radiologist and the COVID-19 detector. The median number of lesions was 2 among the studied participants ranging from 1 to 12 lesions. The most common affected site of the lesions was the lower lobes. There was a significant strong agreement (P value < 0.001, kappa = 0.923) between the radiologist and the semiquantitative CT assessment in the detection of GGO among patients with COVID-19 pneumonia. Also, there were 6 patients who underwent follow-up by semiquantitative CT and radiologist; the median number of lesions was 1 among the studied participants ranging from 1 to 8 lesions. There was a significant strong agreement (P value = 0.001, Kappa = 0.856) between the radiologist and the semiquantitative CT assessment in the detection of GGO during follow-up among patients with COVID-19 pneumonia.ConclusionsThe tested computer program can accurately detect COVID-19 pneumonia as it has better visualization in detecting GGO for diagnosing and following up on COVID-19 pneumonia.

A localization-independent approach for invisible and impalpable ground-glass opacity nodules detection in an in vitro lung specimen: two case reports.

A growing number of ground-glass opacity (GGO) nodules are screened out in lungs. Small GGOs are frequently neither visible nor palpable, thus undetectable during operation. Various nodule localization techniques have been developed to facilitate the intraoperative detection of GGO nodules; however, general localization techniques are infeasible or inappropriate in some cases. The detection of small GGO is a great challenge, even within a surgical specimen in the absence of preoperative localization. A localization-independent approach for GGO detection is urgently needed. Herein, we report two cases with invisible and impalpable small GGO which were not appropriate for preoperative localization. The lesions were anatomically resected under the guidance of three-dimensional (3D) reconstruction and got an adequate margin distance. A vessel (artery, vein, or bronchus) which had advanced into or immediately adjacent to the nodule was assigned as a reference vessel. By dissecting and tracing the reference vessel from proximal to distal, the GGO lesions were successfully detected in the surgical specimens, to the eventual obtainment of an accurate pathological diagnosis. Via the two case reports, we introduced an easily handled approach, namely dissecting and tracing a reference vessel, for GGO detection. The novel approach was first described. Combined with precise anatomical segmentectomy guided by 3D reconstruction, it provides an alternative scheme for GGO resection with no need for preoperative localization.

Efficacy of chest X-ray in the diagnosis of COVID-19 pneumonia: comparison with computed tomography through a simplified scoring system designed for triage

BackgroundThere is no consensus on the imaging modality to be used in the diagnosis and management of Coronavirus disease 2019 (COVID-19) pneumonia. The purpose of this study was to make a comparison between computed tomography (CT) and chest X-ray (CXR) through a scoring system that can be beneficial to the clinicians in making the triage of patients diagnosed with COVID-19 pneumonia at their initial presentation to the hospital.ResultsPatients with a negative CXR (30.1%) had significantly lower computed tomography score (CTS) (p < 0.001). Among the lung zones where the only infiltration pattern was ground glass opacity (GGO) on CT images, the ratio of abnormality seen on CXRs was 21.6%. The cut-off value of X-ray score (XRS) to distinguish the patients who needed intensive care at follow-up (n = 12) was 6 (AUC = 0.933, 95% CI = 0.886–0.979, 100% sensitivity, 81% specificity).ConclusionsComputed tomography is more effective in the diagnosis of COVID-19 pneumonia at the initial presentation due to the ease detection of GGOs. However, a baseline CXR taken after admission to the hospital can be valuable in predicting patients to be monitored in the intensive care units.

ES25.04 Multiple Pulmonary Lesions: Metastasis vs Multiple Primaries, From Martini's Criteria to Molecular Profiling

ES25.04 Multiple Pulmonary Lesions: Metastasis vs Multiple Primaries, From Martini's Criteria to Molecular Profiling

Bronchiectasis is one of the indicators of severe coronavirus disease 2019 pneumonia.

Bronchiectasis is one of the indicators of severe coronavirus disease 2019 pneumonia.

Accuracy of Conventional and Machine Learning Enhanced Chest Radiography for the Assessment of COVID-19 Pneumonia: Intra-Individual Comparison with CT.

Purpose: To evaluate diagnostic accuracy of conventional radiography (CXR) and machine learning enhanced CXR (mlCXR) for the detection and quantification of disease-extent in COVID-19 patients compared to chest-CT. Methods: Real-time polymerase chain reaction (rt-PCR)-confirmed COVID-19-patients undergoing CXR from March to April 2020 together with COVID-19 negative patients as control group were retrospectively included. Two independent readers assessed CXR and mlCXR images for presence, disease extent and type (consolidation vs. ground-glass opacities (GGOs) of COVID-19-pneumonia. Further, readers had to assign confidence levels to their diagnosis. CT obtained ≤ 36 h from acquisition of CXR served as standard of reference. Inter-reader agreement, sensitivity for detection and disease extent of COVID-19-pneumonia compared to CT was calculated. McNemar test was used to test for significant differences. Results: Sixty patients (21 females; median age 61 years, range 38–81 years) were included. Inter-reader agreement improved from good to excellent when mlCXR instead of CXR was used (k = 0.831 vs. k = 0.742). Sensitivity for pneumonia detection improved from 79.5% to 92.3%, however, on the cost of specificity 100% vs. 71.4% (p = 0.031). Overall, sensitivity for the detection of consolidation was higher than for GGO (37.5% vs. 70.4%; respectively). No differences could be found in disease extent estimation between mlCXR and CXR, even though the detection of GGO could be improved. Diagnostic confidence was better on mlCXR compared to CXR (p = 0.013). Conclusion: In line with the current literature, the sensitivity for detection and quantification of COVID-19-pneumonia was moderate with CXR and could be improved when mlCXR was used for image interpretation.

URDNet: A Unified Regression Network for GGO Detection in Lung CT Images

We present a 3D deep neural network known as URDNet for detecting ground-glass opacity (GGO) nodules in 3D CT images. Prior work on GGO detection repurposes classifiers on a large number of windows to perform detection or fine-tuning by box regression based on a previous window classification step. Instead, we consider GGO detection as a multitarget regression problem to focus on the location of GGO. Furthermore, to capture multiscale information, we introduce a backbone network which is a contracting-expanding structure similar to 2D U-net, but we inject the source CT inputs into each layer in the contracting pathway to prevent source information loss at different scales. At last, we propose a two-stage training method for URDNet. In the first stage, the backbone of the network for feature extraction is trained, and in the second, the overall URDNet is fine-tuned based on the previous pretrained weights. By using this training method in conjunction with data augmentation and hard negative mining techniques, our URDNet can be effectively trained even on a small amount of annotated CT images. We evaluate the proposed method on the LIDC-IDRI dataset. It achieves the sensitivity of 90.8% with only 1 false positive per scan. Experimental results show that our detection method achieves the superior detection performance over the state-of-the-art methods. Due to its simplicity and effective, URDNet can be easier to apply to medical IoT systems for improving the efficiency of overall health systems.

Computed Tomography-Guided Hookwire Localization for Ground-glass Opacity Dominant Pulmonary Nodules Before Video-Assisted Thoracoscopic Surgery: Analysis of 472 Lesions

Recent advances in imaging technique and the deterioration of environment have promoted increased detection of ground-glass opacities (GGOs). The aims of the present study were to evaluate the clinical efficiency and safety of CT-guided hookwire localization of GGO-dominant pulmonary nodules before video-assisted thoracoscopic surgery (VATS) and to summarize the key points for differential diagnosis of GGOs. From April 2008 to December 2015, 472 solitary GGO-dominant (GGO component > 50%) pulmonary nodules localized by hookwire were retrospectively analyzed. They ranged from 2.9 to 29.8 mm and were successfully localized with mean duration of 8.1 ± 3.6 min. Postoperative complications were mild and included asymptomatic needle track hemorrhage (28.8%), minimal pneumothorax (8.9%), and haemoptysis (0.4%). All patients accepted VATS resection but one patient (0.2%) was converted to thoracotomy because of strong pleural adhesion. Histological results revealed 102 benign nodules and 370 primary pulmonary adenocarcinomas. The malignant rate of lesions with 50–90% GGO component common in T1bN0M0 and T1cN0M0 was higher than that of over 90% occurred more in T1aN0M0 (83.5% and 87.0% vs. 52.3%, P < 0.001). Besides, there were significant statistical differences between malignant and benign GGOs in terms of age, nodular size, percentage of GGO component, margin, spiculation, lobulation, and vascular convergence (P < 0.05). CT-guided hookwire location for GGO-dominant pulmonary nodules before VATS showed to have a great clinical value for diagnosis and treatment without serious complications. If a GGO-dominant nodule with larger size, 50–90% GGO component, irregular margin, spiculation, lobulation, and vascular convergence was detected in an elder patient, it might have a higher malignant tendency.

P1.05-07 Clinical Significance of Pleural Indentation of Lung Adenocarcinoma Presented as Ground Glass Opacities

After the emphasis on lung cancer screening, incidental detection of ground glass opacities (GGOs) has increased. Such GGOs commonly have peripheral location and pleural indentation, which may increase the risk of visceral pleural invasion (VPI). Although the impact of pleural indentation among solid lung cancers is well established, such impact among GGO lung cancers is not well known. Patients who underwent lung resection with curative intention of GGO lung cancer from April 2007 to January 2016 were retrospectively evaluated. Demographics, radiographic findings, pathological findings, and clinical outcomes including recurrence were analyzed. A total 404 of patients were analyzed, and pleural indentation was observed in 258 patients (63.9%). All cancers turned out to be adenocarcinoma. Older age, worse lung function, longer diameter, and tumor shadow disappearance rate were associated with indentation. VPI was observed in 36 (8.9%) patients, and indentation was independently associated with VPI (odds ratio 9.34, p=0.032) after adjustment of pleural attachment, diameter, and tumor shadow disappearance rate. During a median follow-up duration of 48.3 months, 34 (8.4%) patients recurred. Although indentation was not associated with earlier recurrence (adjusted hazard ratio 2.16, p=0.168), concurrent pleural attachment with indentation was associated with recurrence among those with solid portion≥8mm (adjusted hazard ratio 3.05, p=0.019). Pleural indentation was associated with increased risk of VPI in GGO lung adenocarcinoma. Although indentation itself did not increase the risk of recurrence, concurrent pleural attachment with indentation may increase the risk of recurrence among selected patients.

The use of ultrasound in detecting and defining ground-glass opacities: results of an ex vivo evaluation.

To evaluate the role of ultrasound in detecting and defining ground-glass opacities (GGOs) in surgical specimens of patients undergoing thoracoscopic diagnostic resection. We performed an observational single-centre study of all consecutive patients undergoing thoracoscopic diagnostic resection of GGOs. In each patient, the specimen was scanned with ultrasound; then, a needle was inserted into the lesion to facilitate its detection by the pathologist. We evaluated the rate of detection with ultrasound, compared the size and depth measurements of the lesions as determined from ultrasound scans with those from the histological specimens and correlated the ultrasound findings with the histological subtypes of adenocarcinomas. We reviewed 17 tissue samples. The final diagnoses were 1 (6%) atypical adenomatous hyperplasia, 5 (29%) adenocarcinomas in situ, 4 (24%) minimally invasive adenocarcinomas and 7 (41%) invasive adenocarcinomas. All tumours were successfully identified using ultrasound. The size (P = 0.87) and depth (P = 0.25) of the lesions measured with ultrasound did not significantly differ from the measurements obtained from the histological specimens. In addition, ultrasound size (r = 0.945; P < 0.0001) and depth (r = 0.588; P = 0.013) were significantly correlated with the pathological measurements. All lesions with hyperechoic findings (n = 6) were pure GGOs, whereas lesions with mixed echoic (n = 11) patterns were mixed GGOs. We were unable to differentiate the histological subtypes of adenocarcinomas using the ultrasound scans. Detection of GGOs on ultrasound scans is feasible, but differentiation of the histological subtypes of adenocarcinomas is not possible. The next step is to evaluate the intraoperative reproducibility of our results.

A Review of Ground Glass Opacity Detection Methods in Lung CT Images

A Review of Ground Glass Opacity Detection Methods in Lung CT Images

Detection of ground-glass opacities by use of hybrid iterative reconstruction (iDose) and low-dose 256-section computed tomography: a phantom study

The detection of ground-glass opacities (GGOs) is an important issue in lung cancer screening with low-dose CT. The iterative reconstruction (IR) technique has the ability to improve the image quality relative to the filtered back projection (FBP) technique with low-dose CT. Our purpose was to investigate the ability to detect GGO in a chest phantom using a low-dose CT and hybrid IR, named iDose. Simulated GGOs in a chest phantom were scanned with 256-section CT at tube current second products of 20, 50, 100, and 200 mAs. Five radiologists visually assessed the detectability of GGOs in the phantom. The contrast-to-noise ratio (CNR) for GGOs was used as an estimate of image quality. Comparison of the detectability and CNR between standard images with 200 mAs-FBP and low-dose images with 20, 50, and 100-mAs FBP/iDose were performed by ANOVA with Dunnett's and Tukey's test. The detectability was significantly lower at 20-mAs FBP/iDose and 50-mAs FBP than that at 200-mAs FBP (p < 0.05). There was no significant difference between 50-mAs iDose and 200-mAs FBP and between 100-mAs iDose/FBP and 200-mAs FBP. The CNR was significantly higher on iDose images than that on FBP images at each mAs value. The CNR at 200-mAs FBP was the same as that at 50-mAs iDose (CNR:1.8). The hybrid IR technique and low-dose CT imaging with 50 mAs enabled noise and to maintain the detectability for GGOs in a chest phantom that is equivalent to the reference acquisitions of 200 mAs with FBP.

Characteristics of the ground-glass opacity nodules showing growth during follow-up.

7060 Background: The National Lung Screening Trial implemented in the United States demonstrated a significant reduction in lung cancer mortality by screening with computed tomography (CT). These findings will lead to a sharp increase in the use of CT, and consequently an increase in the detection of ground-glass opacity (GGO) nodules. Non-invasive methods, including follow-up with CT, in order to determine which nodules require invasive diagnostic strategies, should rapidly be defined. The aim of this study is to elucidate the characteristics of GGO nodules that show growth during follow-up with CT. Methods: Between 2000 and 2008, from our database of over 60,000 CT examinations, we identified 178 GGO nodules, which showed a GGO area greater than 20% and measured less than 2cm in diameter from 172 patients. We investigated the clinical characteristics and CT images of the GGO nodules in relation to changes of the GGO nodules during follow-up period. Results: The nodule sizes ranged from 4 to 20 mm, with a median of 11 mm. One hundred and fifteen nodules were observed to be pure GGO. Follow-up periods ranged from 1 to 111 months, with a mean of 26 months. During the follow up period, 20 nodules showed fading or shrinkage, 121 showed no change in size, and 37 showed growth of at least 2 mm in diameter. Growth ranged from 2 to 14 mm, with a means of 4 mm. Time to 2mm nodule growth curves were caliculated with Kaplan-Meier methods using nodules excluding those showed fading or shrinkage. The 2- and 5-year cumulative percentages of growth nodule were calculated to be 23% and 31%. According to the nodule size and proportion of GGO, these were 4% and 14% for nodules less than 10 mm, 25% and 34% for 10 to 14 mm, both 60% for greater than 14 mm, 13% and 18% for pure GGO, and 40% and 51% for mixed GGO. No pure GGO nodule smaller or equal to 7mm showed growth. Univariate analysis revealed that nodule size, proportion of GGO, age, gender, history of lung cancer, and smoking history were factors predicting growth. Multivariate analysis disclosed that nodule size is the only predictive factor. Conclusions: Follow-up with CT is effective in the selection of patients who require further examination. Size of GGO nodule was strongly associated with the growth of GGO nodule during the follow-up.

Évaluation d’un système de détection assisté par ordinateur des nodules parenchymateux pulmonaires avec verre dépoli au scanner multidétecteur

To determine the performance of a CAD system for lung nodules with ground glass opacity component on multidetector-row CT. Materials and methods. The CT examinations of 17 patients with at least one persistent subsolid nodule were reviewed. A first non-blinded consensus review by two expert radiologists resulted in the detection of 104 subsolid nodules larger than 3 mm (74 nodules of ground glass attenuation and 30 mixed nodules with solid and ground glass components). The results from this review were used as a gold standard to determine the performances of the CAD system and 3 independent clinical radiologists involved with the primary interpretations. The sensitivity of the CAD system for the detection of ground glass opacities and mixed nodules was 53% and 73% respectively. These values were not statistically different from the values for the 3 independent observers (42-66% for ground glass opacities and 63-80% for mixed nodules). The sensitivity of each observer significantly increased when the nodules detected by the CAD system were added to those detected by each observer (p<0.0001). A CAD system has a potential impact on the detection rate of subsolid nodules by radiologists.

Ground-glass opacity: High-resolution computed tomography and 64-multi-slice computed tomography findings comparison

Objective Comparative evaluation of ground-glass opacity using conventional high-resolution computed tomography technique and volumetric computed tomography by 64-row multi-slice scanner, verifying advantage of volumetric acquisition and post-processing technique allowed by 64-row CT scanner. Methods Thirty-four patients, in which was assessed ground-glass opacity pattern by previous high-resolution computed tomography during a clinical-radiological follow-up for their lung disease, were studied by means of 64-row multi-slice computed tomography. Comparative evaluation of image quality was done by both CT modalities. Results It was reported good inter-observer agreement ( k value 0.78–0.90) in detection of ground-glass opacity with high-resolution computed tomography technique and volumetric Computed Tomography acquisition with moderate increasing of intra-observer agreement ( k value 0.46) using volumetric computed tomography than high-resolution computed tomography. Conclusions In our experience, volumetric computed tomography with 64-row scanner shows good accuracy in detection of ground-glass opacity, providing a better spatial and temporal resolution and advanced post-processing technique than high-resolution computed tomography.

Commercially Available Computer-Aided Detection System for Pulmonary Nodules on Thin-Section Images Using 64 Detectors-Row CT: Preliminary Study of 48 Cases

Most studies of computer-aided detection (CAD) for pulmonary nodules have focused on solid nodule detection. The aim of this study was to evaluate the performance of a commercially available CAD system in the detection of pulmonary nodules with or without ground-glass opacity (GGO) using 64-detector-row computed tomography compared to visual interpretation. Computed tomographic examinations were performed on 48 patients with existing or suspicious pulmonary nodules on chest radiography. Three radiologists independently reported the location and pattern (GGO, solid, or part solid) of each nodule candidate on computed tomographic scans, assigned each a confidence score, and then analyzed all scans using the CAD system. A reference standard was established by a consensus panel of different radiologists, who found 229 noncalcified nodules with diameters > or = 4 mm. True-positive and false-positive results and confidence levels were used to generate jackknife alternative free-response receiver-operating characteristic plots. The sensitivity of GGO for 3 radiologists (60%-80%) was significantly higher than that for the CAD system (21%) (McNemar's test, P < .0001). For overall and solid nodules, the figure-of-merit values without and with the CAD system were significantly different (P = .005-.04) on jackknife alternative free-response receiver-operating characteristic analysis. For GGO and part-solid nodules, the figure-of-merit values with the CAD system were greater than those without the CAD system, indicating no significant differences. Radiologists are significantly superior to this CAD system in the detection of GGO, but the CAD system can still play a complementary role in detecting nodules with or without GGO.

Automatic detection of ground glass opacity by use of the density features on multi detector-row CT images

Automatic detection of ground glass opacity by use of the density features on multi detector-row CT images