MDCT for Computerized Volumetry of Pneumothoraces in Pediatric Patients

Cai, Wenli; Lee, Edward Y.; Vij, Abhinav; Mahmood, Soran A.; Yoshida, Hiroyuki

doi:10.1016/j.acra.2010.11.008

Cited by 13 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By including scans with pneumothorax that also had additional pathologies from across several CT vendors with a wide variety of technical parameters, we evaluated the algorithm in a more realistic situation than other studies that excluded scans when the pneumothorax was not the only pathology [9]. While previous studies have suggested the use of specific reconstruction parameters to achieve the best results for their computerised pneumothorax volume quantification [11], there is still a large variability regarding the technical setup of chest CT scans in the clinical routine, necessitating the development of flexible and robust solutions.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning detection and quantification of pneumothorax in heterogeneous routine chest computed tomography

Röhrich

Schlegl²,

Bardach

et al. 2020

Eur Radiol Exp

View full text Add to dashboard Cite

Background: Automatically detecting and quantifying pneumothorax on chest computed tomography (CT) may impact clinical decision-making. Machine learning methods published so far struggle with the heterogeneity of technical parameters and the presence of additional pathologies, highlighting the importance of stable algorithms. Methods: A deep residual UNet was developed and evaluated for automated, volume-level pneumothorax grading (i.e., labelling a volume whether a pneumothorax was present or not), and pixel-level classification (i.e., segmentation and quantification of pneumothorax), on a retrospective series of routine chest CT data. Ground truth annotations were provided by radiologists. The fully automated pixel-level pneumothorax segmentation method was trained using 43 chest CT scans and evaluated on 9 chest CT scans with pixel-level annotation basis and 567 chest CT scans on a volume-level basis. Results: This method achieved a receiver operating characteristic area under the curve (AUC) of 0.98, an average precision of 0.97, and a Dice similarity coefficient (DSC) of 0.94. This segmentation performance resulted to be similar to the inter-rater segmentation accuracy of two radiologists, who achieved a DSC of 0.92. The comparison of manual and automated pneumothorax quantification yielded a Pearson correlation coefficient of 0.996. The volume-level pneumothorax grading accuracy was evaluated on 567 chest CT scans and yielded an AUC of 0.98 and an average precision of 0.95. Conclusions: We proposed a deep learning method for the detection and quantification of pneumothorax in heterogeneous routine clinical data that may facilitate the automated triage of urgent examinations and enable treatment decision support.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Some publications have shown the possibility of automatically quantifying the volume of pneumothorax, in both adults [9,10] and paediatric patients [11]. These studies have used specific functions in a multistep approach to achieve the final volume estimate.…”

Section: Introductionmentioning

confidence: 99%

Deep learning detection and quantification of pneumothorax in heterogeneous routine chest computed tomography

Röhrich

Schlegl²,

Bardach

et al. 2020

Eur Radiol Exp

View full text Add to dashboard Cite

show abstract

“…This may have led to the introduction of additional biases. Finally, while measuring the size of the pneumothorax with CT would have been more accurate than with CXR based formula,20,21 most patients had CT imaging after thoracostomy because of cost and patient convenience.…”

Section: Discussionmentioning

confidence: 99%

The Frequency of Reexpansion Pulmonary Edema after Trocar and Hemostat Assisted Thoracostomy in Patients with Spontaneous Pneumothorax

Cha

Kim

Ji³

et al. 2013

Yonsei Med J

View full text Add to dashboard Cite

PurposeSeveral risk factors for development of reexpansion pulmonary edema (REPE) after drainage of pneumothoraces have been reported, but the association between the method of thoracostomy and the development of REPE is unknown. The aim of this study was to compare the frequency of REPE after treatment of spontaneous pneumothorax with trocar or hemostat assisted closed thoracostomy.Materials and MethodsWe performed a prospective, observational study including 173 patients with spontaneous pneumothorax who visited the emergency department from January 2007 to December 2008. In 2007, patients were treated with hemostat-assisted drainage, whereas patients in 2008 were treated with trocar-assisted drainage. The main outcome was the development of REPE, determined by computed tomography of the chest 8 hours after closed thoracostomy. Outcomes in both groups were compared using univariate and multivariate analyses.ResultsNinety-two patients were included, 48 (42 males) of which underwent hemostat-assisted drainage and 44 (41 males) underwent trocar-assisted drainage. The groups were similar in mean age (24±10 vs. 26±14 respectively). The frequencies of REPE after hemostat- and trocar-assisted drainage were 63% (30 patients) and 86% (38 patients) respectively (p=0.009). In multivariate analysis, trocar-assisted drainage was the major contributing factor for developing REPE (odds ratio=5.7, 95% confidence interval, 1.5-21). Age, gender, size of pneumothorax, symptom duration and laboratory results were similar between the groups.ConclusionClosed thoracostomy using a trocar is associated with an increased risk of REPE compared with hemostat-assisted drainage in patients with spontaneous pneumothorax.

show abstract

“…We applied an automated method named pleural region geometric modeling (PR-GM) for extraction of the entire pleural region by modeling the pleural surface in MDCT images [27], which uses rib structures as references based on the fact that the pleural cavity is surrounded by the ribs and its surface is very close to the ribs.…”

Section: Methodsmentioning

confidence: 99%

MDCT quantification is the dominant parameter in decision-making regarding chest tube drainage for stable patients with traumatic pneumothorax

Cai¹,

Lee²,

Fikry³

et al. 2012

Computerized Medical Imaging and Graphics

Self Cite

View full text Add to dashboard Cite

It is commonly believed that the size of a pneumothorax is an important determinant of treatment decision, in particular regarding whether chest tube drainage (CTD) is required. However, the volumetric quantification of pneumothoraces has not routinely been performed in clinics. In this paper, we introduced an automated computer-aided volumetry (CAV) scheme for quantification of volume of pneumothoraces in chest multi-detect CT (MDCT) images. Moreover, we investigated the impact of accurate volume of pneumothoraces in the improvement of the performance in decision-making regarding CTD in the management of traumatic pneumothoraces. For this purpose, an occurrence frequency map was calculated for quantitative analysis of the importance of each clinical parameter in the decision-making regarding CTD by a computer simulation of decision-making using a genetic algorithm (GA) and a support vector machine (SVM). A total of 14 clinical parameters, including volume of pneumothorax calculated by our CAV scheme, was collected as parameters available for decision-making. The results showed that volume was the dominant parameter in decision-making regarding CTD, with an occurrence frequency value of 1.00. The results also indicated that the inclusion of volume provided the best performance that was statistically significant compared to the other tests in which volume was excluded from the clinical parameters. This study provides the scientific evidence for the application of CAV scheme in MDCT volumetric quantification of pneumothoraces in the management of clinically stable chest trauma patients with traumatic pneumothorax.

show abstract

MDCT for Computerized Volumetry of Pneumothoraces in Pediatric Patients

Cited by 13 publications

References 21 publications

Deep learning detection and quantification of pneumothorax in heterogeneous routine chest computed tomography

Deep learning detection and quantification of pneumothorax in heterogeneous routine chest computed tomography

The Frequency of Reexpansion Pulmonary Edema after Trocar and Hemostat Assisted Thoracostomy in Patients with Spontaneous Pneumothorax

MDCT quantification is the dominant parameter in decision-making regarding chest tube drainage for stable patients with traumatic pneumothorax

Contact Info

Product

Resources

About