Post-Processing in Cardiovascular Computed Tomography: Performance of a Client Server Solution versus a Stand-Alone Solution

Abstract

Purpose: To compare the performance of server-based (CSS) versus stand-alone post-processing software (ES) for the evaluation of cardiovascular CT examinations (cvCT) and to determine the crucial steps.

Materials and Methods: Data of 40 patients (20 patients for coronary artery evaluation and 20 patients prior to transcatheter aortic valve implantation [TAVI]) were evaluated by 5 radiologists with CSS and ES. Data acquisition was performed using a dual-source 128-row CT unit (SOMATOM Definition Flash, Siemens, Erlangen, Germany) and a 64-row CT unit (Brilliance 64, Philips, Hamburg, Germany). The following workflow was evaluated: Data loading, aorta and coronary segmentation, curved multiplanar reconstruction (cMPR) and 3 D volume rendering technique (3D-VRT), measuring of coronary artery stenosis and planimetry of the aortic annulus. The time requirement and subjective quality for the workflow were evaluated.

Results: The coronary arteries as well as the TAVI data could be evaluated significantly faster with CSS (5.5 ± 2.9 min and 8.2 ± 4.0 min, respectively) than with ES (13.9 ± 5.2 min and 15.2 ± 10.9 min, respectively, p≤ 0.01). Segmentation of the aorta (CSS: 1.9 ± 2.0 min, ES: 3.7 ± 3.3 min), generating cMPR of coronaries (CSS: 0.5 ± 0.2 min, ES: 5.1 ± 2.6 min), aorta and iliac vessels (CSS: 0.5 ± 0.4 min and 0.4 ± 0.4 min, respectively, ES: 1.6 ± 0.7 min and 2.8 ± 3 min, respectively) could be performed significantly faster with CSS than with ES with higher quality of cMPR, measuring of coronary stenosis and 3D-VRT (p < 0.05).

Conclusion: Evaluation of cvCT can be accomplished significantly faster and better with CSS than with ES. The segmentation remains the most time-consuming workflow step, so optimization of segmentation algorithms could improve performance even further.

Key points:

• With client-server-based (CSS) and stand-alone solutions (ES), cardiovascular CT datasets can be evaluated reliably.

• Evaluation of cardiovascular CT can be performed faster and better with CSS than with ES.

• In particular the generating of curved reconstructions is faster with CSS than with ES.

• Segmentation of data is a crucial step for semiautomatic software (CSS and ES).

Citation Format:

• Lücke C, Foldyna B, Andres C et al. Post-Processing in Cardiovascular Computed Tomography: Performance of a Client Server Solution versus a Stand-Alone Solution. Fortschr Röntgenstr 2014; 186: 1111 – 1121

Zusammenfassung

Ziel: Die Performance einer Client-Server- (CSS) und einer Einzelplatz-Softwarelösung (ES) bezüglich der Auswertung von kardiovaskulären CT-Studien (cvCT) zu vergleichen und zeitrelevante Arbeitsschritte zu bestimmen.

Material und Methoden: Die cvCT von insgesamt 40 Patienten (20 zur Koronarbeurteilung und 20 vor kathetergestütztem Aortenklappenersatz [TAVI]) wurden von 5 Untersuchern mit CSS und ES ausgewertet. Die Datenakquisition erfolgte an einem Dual-Source-128-Zeilen-CT (SOMATOM Definition Flash, Siemens, Erlangen, Deutschland) sowie an einem 64-Zeilen-CT (Brilliance 64, Philips, Best, Niederlande). Folgende Untersuchungsschritte wurden analysiert: Laden der Daten, Segmentierung von Aorta und Koronararterien, Anfertigung gekrümmter multiplanarer Reformatierungen (cMPR) und 3D-Rekonstruktionen (3D-VRT), Vermessung von Koronarstenosen und des Aortenklappenanulus. Die benötigte Zeit und subjektive Qualität der Arbeitsschritte wurde ermittelt.

Ergebnisse: Sowohl die Koronar- als auch die TAVI-Evaluation gelang mit CSS schneller (5,5 ± 2,9 min bzw. 8,2 ± 4,0 min) als mit ES (13,9 ± 5,2 min bzw. 15,2 ± 10,9 min, p≤ 0,01). Die Segmentierung der Aorta (CSS: 1,9 ± 2,0 min, ES: 3,7 ± 3,3 min), die Erstellung der cMPR der Koronararterien (CSS: 0,5 ± 0,2 min, ES: 5,1 ± 2,6 min), der Aorta und der Beckengefäße (CSS: 0,5 ± 0,4 min bzw. 0,4 ± 0,4 min, ES: 1,6 ± 0,7 min bzw. 2,8 ± 3 min) gelingt signifikant schneller mit CSS im Vergleich zur ES bei jeweils besserer Qualität der cMPR sowie der Vermessung und 3D-VRT der Koronararterien (p < 0,05).

Schlussfolgerung: Die Auswertung von cvCT mit CSS benötigt weniger Zeit und führt zu qualitativ besseren Ergebnissen als mit ES. Die Segmentierung der Daten verbleibt als zeitrelevanter Arbeitsschritt, sodass eine Optimierung von Segmentierungsalgorithmen zu weiteren deutlichen Zeitersparnissen führen könnte.

Key words

cardiac - vascular - CT - post-processing software - performance

• References

• 1 Achenbach S, Marwan M, Schepis T et al. High-pitch spiral acquisition: a new scan mode for coronary CT angiography. Journal of cardiovascular computed tomography 2009; 3: 117-121
• 2 Boehm T, Handgraetinger O, Link J et al. Evaluation of radiological workstations and web-browser-based image distribution clients for a PACS project in hands-on workshops. European radiology 2004; 14: 908-914
• 3 Kotter E, Langer M. Computer aided detection and diagnosis in radiology. European radiology 2011; 21: 590-592
• 4 Tsiflikas I, Biermann C, Thomas C et al. Carotid artery stenosis: performance of advanced vessel analysis software in evaluating CTA. European journal of radiology 2012; 81: 2255-2259
• 5 Kopetsch T. Entwicklung der Altersstruktur und der Zahl der Ärzte in Deutschland unter besonderer Berücksichtigung der Radiologen. Deutscher Röntgenkongress 2010; 182: S66
• 6 Andres C, Gutberlet M, Lehmkuhl L. Preoperative CT imaging in infectious endocarditis before urgent indication for heart valve replacement: whole body computerized tomography including coronary evaluation. Fortschr Röntgenstr 2013; 185: 373-375
• 7 Achenbach S, Barkhausen J, Beer M et al. Konsensusempfehlungen der DRG/DGK/DGPK zum Einsatz der Herzbildgebung mit Computertomografie und Magnetresonanztomografie. Fortschr Röntgenstr 2012; 184: 345-368
• 8 Arbab-Zadeh A, Miller JM, Rochitte CE et al. Diagnostic accuracy of computed tomography coronary angiography according to pre-test probability of coronary artery disease and severity of coronary arterial calcification. The CORE-64 (Coronary Artery Evaluation Using 64-Row Multidetector Computed Tomography Angiography) International Multicenter Study. Journal of the American College of Cardiology 2012; 59: 379-387
• 9 Holmes Jr DR, Mack MJ, Kaul S et al. 2012 ACCF/AATS/SCAI/STS Expert Consensus Document on Transcatheter Aortic Valve Replacement: Developed in collaboration with the American Heart Association, American Society of Echocardiography, European Association for Cardio-Thoracic Surgery, Heart Failure Society of America, Mended Hearts, Society of Cardiovascular Anesthesiologists, Society of Cardiovascular Computed Tomography, and Society for Cardiovascular Magnetic Resonance. The Annals of thoracic surgery 2012; 93: 1340-1395
• 10 Lehmkuhl L, Foldyna B, Von Aspern K et al. Inter-individual variance and cardiac cycle dependency of aortic root dimensions and shape as assessed by ECG-gated multi-slice computed tomography in patients with severe aortic stenosis prior to transcatheter aortic valve implantation: is it crucial for correct sizing?. The international journal of cardiovascular imaging 2013; 29: 693-703
• 11 Kempfert J, Van Linden A, Lehmkuhl L et al. Aortic annulus sizing: echocardiographic versus computed tomography derived measurements in comparison with direct surgical sizing. European Journal of Cardio-Thoracic Surgery 2012; 42: 627-633
• 12 Gerhards A, Raab P, Herber S et al. Software-assisted CT-postprocessing of the carotid arteries. Fortschr Röntgenstr 2004; 176: 870-874
• 13 Beyer F, Zierott L, Fallenberg EM et al. Comparison of sensitivity and reading time for the use of computer-aided detection (CAD) of pulmonary nodules at MDCT as concurrent or second reader. European radiology 2007; 17: 2941-2947
• 14 Dewey M, Schnapauff D, Laule M et al. Multislice CT coronary angiography: evaluation of an automatic vessel detection tool. Fortschr Röntgenstr 2004; 176: 478-483
• 15 Busch S, Johnson TR, Nikolaou K et al. Visual and automatic grading of coronary artery stenoses with 64-slice CT angiography in reference to invasive angiography. European radiology 2007; 17: 1445-1451
• 16 Cordeiro MA, Lardo AC, Brito MS et al. CT angiography in highly calcified arteries: 2D manual vs. modified automated 3D approach to identify coronary stenoses. The international journal of cardiovascular imaging 2006; 22: 507-516
• 17 Reimann AJ, Tsiflikas I, Brodoefel H et al. Efficacy of computer aided analysis in detection of significant coronary artery stenosis in cardiac using dual source computed tomography. The international journal of cardiovascular imaging 2009; 25: 195-203
• 18 Diehm N, Baumgartner I, Silvestro A et al. Automated software supported versus manual aorto-iliac diameter measurements in CT angiography of patients with abdominal aortic aneurysms: assessment of inter- and intraobserver variation. VASA Zeitschrift für Gefäßkrankheiten 2005; 34: 255-261
• 19 Lutz AM, Willmann JK, Pfammatter T et al. Evaluation of aortoiliac aneurysm before endovascular repair: comparison of contrast-enhanced magnetic resonance angiography with multidetector row computed tomographic angiography with an automated analysis software tool. Journal of vascular surgery 2003; 37: 619-627
• 20 Blanke P, Spira EM, Ionasec R et al. Semiautomated Quantification of Aortic Annulus Dimensions on Cardiac CT for TAVR. J Am Coll Cardiol Img 2014; 7: 320-322