A new algorithm for local surface smoothing with application to chest wall nodule segmentation in lung CT data

3D segmentation of lung CT data with graph-cuts: analysis of parameter sensitivities

10.1117/12.2219897 ◽

2016 ◽

Author(s):

Jung won Cha ◽

Neal Dunlap ◽

Brian Wang ◽

Amir Amini

Keyword(s):

Graph Cuts ◽

3D Segmentation ◽

Ct Data ◽

Lung Ct

Download Full-text

Preliminary Analysis of Thorax CT Image Pre-Processing and 3D Modeling of Patients with Pectus Excavatum

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.459.465 ◽

2012 ◽

Vol 459 ◽

pp. 465-468 ◽

Cited By ~ 1

Author(s):

Bo Tang ◽

Li Jiang

Keyword(s):

Chest Wall ◽

3D Modeling ◽

Pectus Excavatum ◽

Reference Value ◽

Spiral Ct ◽

Ct Image ◽

Operative Correction ◽

Ct Data ◽

Set Up ◽

Thorax Ct

Pectus excavatum is the most common congenital chest wall deformity. Extensive corrective surgery prior to age 3 may disturb chest wall growth and result in a constricted thorax. We paid particular attention to the role of thoracic spiral computed tomography (CT) image and 3D 1modeling of patient with pectus excavatum. Thoracic spiral CT was always performed on patient who had developed restrictive chest walls following pectus excavatum surgery. These patients then underwent a Nuss operation to elevate the sternum and attempt to correct their restrictive chest wall defects. The method used the CT image to set up 3D modeling reconstruction defined the orientation of the ribs and costal cartilages and their relationship to the sternum, allowing exact preoperative measurement of the bony rib cage and guiding individualized operative correction. It also has laid foundation for simulation of Nuss operative correction of pectus excavatum and has important reference value in selecting and positioning of support frame. As was shown in the result, the accuracy of costal cartilage’s model was promoted in certain extent. Manual work of model modification was significantly reduced and the period of 3D modeling was shortened by approximately 40%. The 3D modeling of spiral CT data is useful in both preoperative and postoperative evaluation.

Download Full-text

Experience with the use of glucocorticoids and netakimab in the treatment of patients with COVID-19 in outpatient settings

Clinical Medicine (Russian Journal) ◽

10.30629/0023-2149-2020-98-6-449-455 ◽

2020 ◽

Vol 98 (6) ◽

pp. 449-455

Author(s):

R. Е. Pavlov ◽

S. V. Tsarenko ◽

A. V. Sekinaeva ◽

P. N. Rybin ◽

M. V. Nagibinа ◽

...

Keyword(s):

Repeated Administration ◽

Betamethasone Dipropionate ◽

Laboratory Methods ◽

Comprehensive Examination ◽

Coronavirus Infection ◽

Ct Data ◽

Outpatient Settings ◽

Lung Ct ◽

Complex Drug

Goal. To study the eﬃcacy and safety of therapy with netakimab (NTK) and the complex drug — betamethasone dipropionate + betamesone phosphate (B) in patients with COVID-19 in outpatient settings. Material and methods. A retrospective analysis of the treatment of 12 patients with severe new coronavirus infection who received therapy was performed simultaneously with netakimab (NTK) (the first administration at a dose of 60–120 mg, if indicated-the second administration at a dose of 60 mg) and a combined drug-betamethasone dipropionate + betamesone phosphate (B), at a dose of 2 ml oﬃcial solution. Treatment was performed at 7th (95% CI [6; 10]) day from the beginning of the disease. Repeated administration of NTK was performed in older patients due to insuﬃcient eﬀect of stopping hyperthermia and/or hypoxemia. A comprehensive examination of patients before and after NTK + B therapy was performed, which included clinical (assessment of the severity and duration of fever, analysis of gas exchange indicators), instrumental (computed tomography of the lungs) and laboratory methods, including determination of ferritin, IL-6. Results. Simultaneous use of NTK + B therapy resulted in compensation of hyperthermia and/or SpO2 > 93% 2.5 (95% CI [1; 3]) days after the first administration, a significant decrease in the level of inflammatory markers, and positive dynamics according to lung CT data. In all cases, increased respiratory support or additional antibiotic therapy, as well as hospitalization, were not required. The search for articles on the topic was carried out in the PubМed database in English and Russian, published in the period from 2001 to 2020.

Download Full-text

Haptic Landmark Positioning and Automatic Landmark Transfer in 4D Lung CT Data

Informatik aktuell - Bildverarbeitung für die Medizin 2008 ◽

10.1007/978-3-540-78640-5_63 ◽

2008 ◽

pp. 313-317

Author(s):

Matthias Färber ◽

Björn Gawenda ◽

Christian-Arved Bohn ◽

Heinz Handels

Keyword(s):

Ct Data ◽

Lung Ct

Download Full-text

First results of spatial reconstruction and quantification of COVID-19 chest CT infiltrates using lung CT analyzer and 3D slicer

British Journal of Surgery ◽

10.1093/bjs/znab202.077 ◽

2021 ◽

Vol 108 (Supplement_4) ◽

Author(s):

R Bumm ◽

A Lasso ◽

N Kawel-Böhm ◽

A Wäckerlin ◽

P Ludwig ◽

...

Keyword(s):

3D Reconstruction ◽

Ct Scans ◽

Ct Reconstruction ◽

Hounsfield Units ◽

Data Set ◽

3D Slicer ◽

First Results ◽

Patients At Risk ◽

Ct Data ◽

Lung Ct

Abstract Objective Lung CT scans are early diagnostic tests in evaluation of COVID-19 patients. Data are usually analyzed visually and the extent of infiltrations can only roughly be estimated. The aim of the present study was to create a software to spatially visualize and quantify infiltrated and collapsed areas in lung CT scans and set these volumes into relation with non-affected lung areas. Methods A new software "Lung CT Analyzer" (LCTA, 1) was created from scratch in an international team-effort within the 3D medical imaging software 3D Slicer (2). LCTA consists of two components: "Lung CT Segmenter" implements an intuitive and semiautomatic workflow for the generation of lung masks. LCTA then uses masked thresholds of Hounsfield units to detect non-affected versus affected (emphysematous, infiltrated, and collapsed) areas of the lung. Intrapulmonary vessels are subtracted from the other volumes. Segment volumes are expressed in milliliters and displayed in 3D. COVID-Q was defined as affected divided by non-affected volume and can be calculated separately for both lungs. 3D Slicer and LCTA are open source, freely available and maintained on Github. Results CT data of twelve patients with moderate to severe COVID-19 (9 m, 3 f) were selected for the present retrospective study. All scans were performed shortly after admission. Thresholds of Hounsfield units (HU) for areas of interest were defined prior to the study and processing was identical for all patients. The median time effort for 3D reconstruction was 8 minutes per patient. For more detailed results please see the enclosed table. A 3D Slicer demo data set (Control) has been included for comparison. Conclusion The COVID-19 pandemic promoted fast-paced innovations such as LCTA in our hospital. LCTA was feasible, reproducible and easy to perform. COVID-Q correlated with COVID-19 lung involvement in all cases. All fatal cases showed COVID-Q values of > 2.0. LCTA enabled the serial 3D reconstruction of infiltrated and collapsed lung areas in lung CT scans. The procedure may be of great help in the future analysis of pulmonary infiltrates of any cause. In COVID-19 disease, volumetric lung CT reconstruction could result in the definition of new prognostic factors, identify patients “at-risk” in the ICU, and be useful for follow-up. (1) Lung CT Analyzer: https://github.com/rbumm/SlicerLungCTAnalyzer (2) 3D Slicer: http://slicer.org

Download Full-text

DeepLung: 3D Deep Convolutional Nets for Automated Pulmonary Nodule Detection and Classification

10.1101/189928 ◽

2017 ◽

Cited By ~ 26

Author(s):

Wentao Zhu ◽

Chaochun Liu ◽

Wei Fan ◽

Xiaohui Xie

Keyword(s):

State Of The Art ◽

Gradient Boosting ◽

Gradient Boosting Machine ◽

Nodule Detection ◽

Pulmonary Nodule Detection ◽

Public Dataset ◽

Ct Data ◽

Nodule Classification ◽

Lung Ct ◽

3D Networks

AbstractIn this work, we present a fully automated lung CT cancer diagnosis system, DeepLung. DeepLung contains two parts, nodule detection and classification. Considering the 3D nature of lung CT data, two 3D networks are designed for the nodule detection and classification respectively. Specifically, a 3D Faster R-CNN is designed for nodule detection with a U-net-like encoder-decoder structure to effectively learn nodule features. For nodule classification, gradient boosting machine (GBM) with 3D dual path network (DPN) features is proposed. The nodule classification subnetwork is validated on a public dataset from LIDC-IDRI, on which it achieves better performance than state-of-the-art approaches, and surpasses the average performance of four experienced doctors. For the DeepLung system, candidate nodules are detected first by the nodule detection subnetwork, and nodule diagnosis is conducted by the classification subnetwork. Extensive experimental results demonstrate the DeepLung is comparable to the experienced doctors both for the nodule-level and patient-level diagnosis on the LIDC-IDRI dataset.

Download Full-text

CT radiomic models to distinguish COVID-19 pneumonia from other interstitial pneumonias

La radiologia medica ◽

10.1007/s11547-021-01370-8 ◽

2021 ◽

Author(s):

Nicolò Cardobi ◽

Giulio Benetti ◽

Giuseppe Cardano ◽

Cinzia Arena ◽

Claudio Micheletto ◽

...

Keyword(s):

Interstitial Pneumonia ◽

Respiratory Symptoms ◽

False Negative ◽

Classification Performance ◽

Nasopharyngeal Swab ◽

Training Set ◽

Median Response ◽

Ct Data ◽

Lung Ct ◽

Model Approach

Abstract Purpose To classify COVID-19, COVID-19-like and non-COVID-19 interstitial pneumonia using lung CT radiomic features. Material and Methods CT data of 115 patients with respiratory symptoms suspected for COVID-19 disease were retrospectively analyzed. Based on the results of nasopharyngeal swab, patients were divided into two main groups, COVID-19 positive (C +) and COVID-19 negative (C−), respectively. C− patients, however, presented with interstitial lung involvement. A subgroup of C−, COVID-19-like (CL), were considered as highly suggestive of COVID pneumonia at CT. Radiomic features were extracted from the whole lungs. A dual machine learning (ML) model approach was used. The first one excluded CL patients from the training set, eventually included on the test set. The second model included the CL patients also in the training set. Results The first model classified C + and C− pneumonias with AUC of 0.83. CL median response (0.80) was more similar to C + (0.92) compared to C− (0.17). Radiomic footprints of CL were similar to the C + ones (possibly false negative swab test). The second model, however, merging C + with CL patients in the training set, showed a slight decrease in classification performance (AUC = 0.81). Conclusion Whole lung ML models based on radiomics can classify C + and C− interstitial pneumonia. This may help in the correct management of patients with clinical and radiological stigmata of COVID-19, however presenting with a negative swab test. CL pneumonia was similar to C + pneumonia, albeit with slightly different radiomic footprints.

Download Full-text