Computertomographische Untersuchungen mit computerassistierter Detektion von pulmonalen Rundherden bei Hund und Katze

Zusammenfassung

Ziel war, den diagnostischen Wert der computerassistierten Detektion (CAD) von pulmonalen Rundherden in der Tiermedizin zu überprüfen. Hierzu wurde die Detektionsrate der computerassistierten Methode mit der Detektionsrate zweier unabhängiger Untersucher in Bezug auf die Sensitivität und die Rate falsch positiver Befunde verglichen. Material und Methoden: In die Studie wurden 51 Hunde und 16 Katzen einbezogen, bei denen zuvor mittels computertomographischer Untersuchung pulmonale Rundherde diagnostiziert worden waren. Zwei unabhängige Untersucher erfassten bei jedem Patienten die Anzahl der Rundherde ≥ 3 mm und setzten anschließend das Programm zur automatischen Rundherddetektion ein. Im letzten Schritt wurde ein Konsensus der beiden Untersucher unter Kenntnis der CAD-Ergebnisse getroffen. Zur Rundherdauswertung diente ein kommerziell erhältliches Programm zur computerassistierten Detektion. Ergebnisse: Insgesamt wurden 586 pulmonale Rundherde diagnostiziert. Für Untersucher 1 ergab sich eine Sensitivität von 89,2%, für Untersucher 2 betrug sie 87,4%. Die CAD erreichte eine Sensitivität von 69,4%. Mithilfe der CAD ließ die Sensitivität von Untersucher 1 auf 94,7% und die Sensitivität von Untersucher 2 auf 90,8% steigern. Schlussfolgerung und klinische Relevanz: Das verwendete CAD-System wies eine moderate Sensitivität von 69,4% auf. Trotz gewichtiger Limitationen wie einer hohen Anzahl an falsch positiven und falsch negativen Befunden konnte die CAD die Sensitivität beider Untersucher erhöhen, sodass sie zumindest eine unterstützende Funktion besitzt.

Summary

Objective: The aim of this study was to assess the potential benefit of computer-assisted detection (CAD) of pulmonary nodules in veterinary medicine. Therefore, the CAD rate was compared to the detection rates of two individual examiners in terms of its sensitivity and falsepositive findings. Materials and methods: We included 51 dogs and 16 cats with pulmonary nodules previously diagnosed by computed tomography. First, the number of nodules ≥ 3 mm was recorded for each patient by two independent examiners. Subsequently, each examiner used the CAD software for automated nodule detection. With the knowledge of the CAD results, a final consensus decision on the number of nodules was achieved. The software used was a commercially available CAD program. Results: The sensitivity of examiner 1 was 89.2%, while that of examiner 2 reached 87.4%. CAD had a sensitivity of 69.4%. With CAD, the sensitivity of examiner 1 increased to 94.7% and that of examiner 2 to 90.8%. Conclusion and clinical relevance: The CAD-system, which we used in our study, had a moderate sensitivity of 69.4%. Despite its severe limitations, with a high level of false-positive and false-negative results, CAD increased the examiners’ sensitivity. Therefore, its supportive role in diagnostics appears to be evident.

Schlüsselwörter

Computertomographie - Neoplasie - Diagnostik - bildgebende Verfahren

Keywords

Computed tomography - neoplasia - diagnostics - imaging techniques

• Literatur

• 1 Achenbach T. Computerunterstützte Diagnostik in der Thoraxradiologie – aktuelle Schwerpunkte und Techniken. RöFo 2003; 1471-1481.
• 2 Alexander K, Joly H, Blond L, D’Anjou M-A, Nadeau M-È, Olive J, Beauchamp G. A comparison of computed tomography, computed radiography, and film-screen radiography for the detection of canine pulmonary nodules. Vet Radiol Ultrasound 2012; 53: 258-265.
• 3 Armato 3rd SG, Giger ML, Moran CJ, Blackburn JT, Doi K, MacMahon H. Computerized detection of pulmonary nodules on CT scans. Radiographics 1999; 19: 1303-1311.
• 4 Armbrust LJ, Biller DS, Bamford A, Chun R, Garrett LD, Sanderson MW. Comparison of three-view thoracic radiography and computed tomography for detection of pulmonary nodules in dogs with neoplasia. J Am Vet Med Assoc 2012; 240: 1088-1094.
• 5 Austin JH, Romney BM, Goldsmith LS. Missed bronchogenic carcinoma: radiographic findings in 27 patients with a potentially resectable lesion evident in retrospect. Radiology 1992; 182: 115-122.
• 6 Awai K, Murao K, Ozawa A, Komi M, Hayakawa H, Hori S, Nishimura Y. Pulmonary nodules at chest CT: effect of computer-aided diagnosis on radiologists’ detection performance. Radiology 2004; 230: 347-352.
• 7 Bae KT, Kim JS, Na YH, Kim KG, Kim JH. Pulmonary nodules: automated detection on CT images with morphologic matching algorithm – preliminary results. Radiology 2005; 236: 286-293.
• 8 Beigelman-Aubry C, Raffy P, Yang W, Castellino RA, Grenier PA. Computer-aided detection of solid lung nodules on follow-up MDCT screening: evaluation of detection, tracking, and reading time. Am J Roentgenol 2007; 189: 948-955.
• 9 Beyer F, Zierott L, Fallenberg EM, Juergens KU, Stoeckel J, Heindel W, Wormanns D. Comparison of sensitivity and reading time for the use of computer-aided detection (CAD) of pulmonary nodules at MDCT as concurrent or second reader. Eur Radiol 2007; 17: 2941-2947.
• 10 Bogoni L, Ko JP, Alpert J, Anand V, Fantauzzi J, Florin CH, Koo CW, Mason D, Rom W, Shiau M, Salganicoff M, Naidich DP. Impact of a computeraided detection (CAD) system integrated into a picture archiving and communication system (PACS) on reader sensitivity and efficiency for the detection of lung nodules in thoracic CT exams. J Digit Imaging 2012; 25: 771-781.
• 11 Brown MS, Goldin JG, Suh RD, McNitt-Gray MF, Sayre JW, Aberle DR. Lung micronodules: automated method for detection at thin-section CT – initial experience. Radiology 2003; 226: 256-262.
• 12 Christe A, Leidolt L, Huber A, Steiger P, Szucs-Farkas Z, Roos JE, Heverhagen JT, Ebner L. Lung cancer screening with CT: Evaluation of radiologists and different computer assisted detection software (CAD) as first and second readers for lung nodule detection at different dose levels. Eur J Radiol 2013; 82: e873-e878.
• 13 Das M, Muhlenbruch G, Mahnken AH, Flohr TG, Gundel L, Stanzel S, Kraus T, Gunther RW, Wildberger JE. Small pulmonary nodules: effect of two computer-aided detection systems on radiologist performance. Radiology 2006; 241: 564-571.
• 14 Godoy MC, Kim TJ, White CS, Bogoni L, de Groot P, Florin C, Obuchowski N, Babb JS, Salganicoff M, Naidich DP, Anand V, Park S, Vlahos I, Ko JP. Benefit of computer-aided detection analysis for the detection of subsolid and solid lung nodules on thin-and thick-section CT. Am J Roentgenol 2013; 200: 74-83.
• 15 Jeong YJ, Yi CA, Lee KS. Solitary pulmonary nodules: detection, characterization, and guidance for further diagnostic workup and treatment. Am J Roentgenol 2007; 188: 57-68.
• 16 Joly H, d’Anjou MA, Alexander K, Beauchamp G. Comparison of singleslice computed tomography protocols for detection of pulmonary nodules in dogs. Vet Radiol Ultrasound 2009; 50: 279-284.
• 17 Kim JS, Kim JH, Cho G, Bae KT. Automated detection of pulmonary nodules on CT images: effect of section thickness and reconstruction interval – initial results. Radiology 2005; 236: 295-299.
• 18 Lee Y, Hara T, Fujita H, Itoh S, Ishigaki T. Automated detection of pulmonary nodules in helical CT images based on an improved template-matching technique. IEEE Trans Med Imaging 2001; 20: 595-604.
• 19 Li Q, Li F, Doi K. Computerized detection of lung nodules in thin-section CT images by use of selective enhancement filters and an automated rulebased classifier. Acad Radiol 2008; 15: 165-175.
• 20 Marten K, Engelke C. Computer-aided detection and automated CT volumetry of pulmonary nodules. Eur Radiol 2007; 17: 888-901.
• 21 Muhm JR, Miller WE, Fontana RS, Sanderson DR, Uhlenhopp MA. Lung cancer detected during a screening program using four-month chest radiographs. Radiology 1983; 148: 609-615.
• 22 Nemanic S, London CA, Wisner ER. Comparison of thoracic radiographs and single breath-hold helical CT for detection of pulmonary nodules in dogs with metastatic neoplasia. J Vet Intern Med 2006; 20: 508-515.
• 23 Niesterok C, Köhler C, Ludewig E, Alef M, Oechtering G, Kiefer I. Vergleich von Projektionsradiographie und Computertomographie zur Detektion pulmonaler Rundherde bei Hund und Katze. Tierärztl Prax 2013; 41 (K): 155-162.
• 24 Otoni CC, Rahal SC, Vulcano LC, Ribeiro SM, Hette K, Giordano T, Doiche DP, Amorim RL. Survey radiography and computerized tomography imaging of the thorax in female dogs with mammary tumors. Acta Vet Scand 2010; 52: 20.
• 25 Prather AB, Berry CR, Thrall DE. Use of radiography in combination with computed tomography for the assessment of noncardiac thoracic disease in the dog and cat. Vet Radiol Ultrasound 2005; 46: 114-121.
• 26 Preteux FNM. et al. Algorithms for automated evaluation of pulmonary lesions by high resolution CT via image analysis. Proc Radiol Soc of North America. 1989
• 27 Wiemker R, Rogalla P, Blaffert T, Sifri D, Hay O, Shah E, Truyen R, Fleiter T. Aspects of computer-aided detection (CAD) and volumetry of pulmonary nodules using multislice CT. Br J Radiol 2005; 78: S46-S56.
• 28 Wormanns D, Fiebich M, Saidi M, Diederich S, Heindel W. Automatic detection of pulmonary nodules at spiral CT: clinical application of a computer-aided diagnosis system. Eur Radiol 2002; 12: 1052-1057.
• 29 Wormanns D, Kohl G, Klotz E, Marheine A, Beyer F, Heindel W, Diederich S. Volumetric measurements of pulmonary nodules at multi-row detector CT: in vivo reproducibility. Eur Radiol 2004; 14: 86-92.