Annemarie Bakai

Automated volumetry of solid pulmonary nodules in a phantom: accuracy across different CT scanner technologies.

Objectives:The accuracy of automated volumetry for pulmonary nodules in a phantom using different CT scanner technologies from single-slice spiral CT (SSCT) to 64-slice multidetector-row CT (MDCT) was compared. Materials and Methods:A lung phantom with 5 different categories of pulmonary nodules was scanned using a single-slice spiral CT, a 4-slice MDCT, a 16-slice MDCT and a 64-slice MDCT. Each category comprised of 7–9 nodules each (total n = 40) with different known volumes. Standard dose and low dose protocols were performed using thin and thick collimation. Image data were reconstructed at the thinnest slice thickness. Data sets were analyzed with a dedicated volumetry software. Volumes of all nodules were calculated and compared. Results:Mean absolute percentage error (APE) for all nodules was 8.65% (±7.29%) for the SSCT, 10.26% (±8.25%) for the 4-slice MDCT, 8.19% (±7.57%) for the 16-slice MDCT and 7.89% (±7.39%) for the 64-slice MDCT. There was statistically significant influence of the scanner type, protocol, anatomic location, and nodule volume on APE, but overall, APEs were comparable. Conclusion:Computer-aided volumetry showed accurate measurements in all tested scanner types. This finding has important implications for nodule assessment and follow-up.

Fuzzy pulmonary vessel segmentation in contrast enhanced CT data

Pulmonary vascular tree segmentation has numerous applications in medical imaging and computer-aided diagnosis (CAD), including detection and visualization of pulmonary emboli (PE), improved lung nodule detection, and quantitative vessel analysis. We present a novel approach to pulmonary vessel segmentation based on a fuzzy segmentation concept, combining the strengths of both threshold and seed point based methods. The lungs of the original image are first segmented and a threshold-based approach identifies core vessel components with a high specificity. These components are then used to automatically identify reliable seed points for a fuzzy seed point based segmentation method, namely fuzzy connectedness. The output of the method consists of the probability of each voxel belonging to the vascular tree. Hence, our method provides the possibility to adjust the sensitivity/specificity of the segmentation result a posteriori according to application-specific requirements, through definition of a minimum vessel-probability required to classify a voxel as belonging to the vascular tree. The method has been evaluated on contrast-enhanced thoracic CT scans from clinical PE cases and demonstrates overall promising results. For quantitative validation we compare the segmentation results to randomly selected, semi-automatically segmented sub-volumes and present the resulting receiver operating characteristic (ROC) curves. Although we focus on contrast enhanced chest CT data, the method can be generalized to other regions of the body as well as to different imaging modalities.

Computer-aided detection of pulmonary embolism: Influence on radiologists’ detection performance with respect to vessel segments

The purpose was to assess the sensitivity of a CAD software prototype for the detection of pulmonary embolism in MDCT chest examinations with regard to vessel level and to assess the influence on radiologists’ detection performance. Forty-three patients with suspected PE were included in this retrospective study. MDCT chest examinations with a standard PE protocol were acquired at a 16-slice MDCT. All patient data were read by three radiologists (R1, R2, R3), and all thrombi were marked. A CAD prototype software was applied to all datasets, and each finding of the software was analyzed with regard to vessel level. The standard of reference was assessed in a consensus read. Sensitivity for the radiologists and CAD software was assessed. Thirty-three patients were positive for PE, with a total of 215 thrombi. The mean overall sensitivity for the CAD software alone was 83% (specificity, 80%). Radiologist sensitivity was 77% = R3, 82% = R2, and R1 = 87%. With the aid of the CAD software, sensitivities increased to 98% (R1), 93% (R2), and 92% (R3) (p<0.0001). CAD performance at the lobar level was 87%, at the segmental 90% and at the subsegmental 77%. With the use of CAD for PE, the detection performance of radiologists can be improved.

Computer-aided measurements of pulmonary emphysema in chest multidetector-row spiral computed tomography: effect of image reconstruction parameters.

Objective: To evaluate the effect of different image reconstruction parameters on quantitative automated measurements of pulmonary emphysema in chest multidetector-row spiral computed tomography. Materials and Methods: Thirty patients with known emphysema underwent multidetector-row spiral computed tomography. Retrospective reconstruction with a soft tissue kernel (Siemens B20 at 1-mm, 2-mm, and 3-mm slices) and 4 alternative kernel grades (from smooth to sharp: Siemens B30, B40, B50, B60 at 1-mm slices) was performed. Total lung volume, emphysema volume (EV), 15th percentile density, and 4 EV clusters were quantified. Results were compared with those of standard algorithm B20/1-mm slices. Results: Differences in total lung volume were less than 0.2%. Alternative kernel grades resulted in a significantly increased average EV. The 15th percentile density showed a significant average difference for all alternative algorithms. The large emphysema cluster showed a significant change for reconstruction algorithms B50, B60, B20/2 mm and B20/3 mm. Conclusions: Pulmonary EV is significantly affected by different reconstruction algorithms.

Halbautomatische Segmentierung von Pulmonalgefäßen in CT Daten als Referenz zur Validierung automatischer Verfahren

Das Segmentieren von Pulmonalgefasen in Computertomographie (CT) Daten wurde schon vielfach behandelt und wird z.B. bei der computerunterstutzten Detektion von Lungenembolien angewendet. Vielen Segmentierverfahren fehlt jedoch eine quantitative Validierung aufgrund mangelnder Referenzsegmentierungen. Wir stellen ein System zur halbautomatischen Segmentierung von Blutgefasen in definierten Bereichen der Lunge basierend auf dem Random-Walker-Algorithmus vor. Durch Initialisierung der Methode mittels automatisch generierter Saatpunkte wird die Effizienz des Verfahrens erhoht und die erforderliche Benutzerinteraktion reduziert. Die resultierenden Segmentierungen konnen zur Validierung von automatischen Verfahren verwendet werden. Exemplarisch evaluieren wir ein vollautomatisches Segmentierverfahren basierend auf dem Fuzzy-Connectedness-Algorithmus.