Computer-aided Detection of Pulmonary Embolism on CT Angiography: Initial Experience

Objective To evaluate the performance of a computer-aided detection (CAD) system for diagnosis of pulmonary embolism on computed tomography (CT) pulmonary angiography. Materials and Methods One hundred and four pulmonary CT angiograms for pulmonary emboli (PE) were reviewed both by radiologists and a CAD detection system (ImageChecker CT V2.0, R2 Technology Inc, Sunnyvale, CA). CT scans, read and reported by radiologists in a routine daily clinical setting, were later processed by the CAD system. The performance of the CAD system was analyzed. Results Forty-five PE were identified by the radiologists in 15 patients. The CAD system revealed 123 findings, interpreted by the system as PE. Twenty-six of them, detected in 8 patients, represented true-positive results. Ninety-seven (78.9%) CAD findings were not true PE and were defined as false-positive. Nineteen true PE in 7 patients were missed by the CAD system constituting 42% false-negative rate. Sensitivity of the CAD system was 53.3% and the specificity was 77.5%. The positive predictive value of CAD system was 28.5% and the negative predictive value was 90.7%. Conclusions With the evaluated CAD system, it is relatively simple and fast to check all detected findings and decide if they represent true PE. However, high false-negative results demand technologic improvement, to increase the sensitivity of the system. It is anticipated to become a promising supplement to the work and eyes of the radiologist in detecting PE on pulmonary CT angiography.

[1]  Martin Fiebich,et al.  Automatic detection of pulmonary nodules at spiral CT: clinical application of a computer-aided diagnosis system , 2002, European Radiology.

[2]  Berkman Sahiner,et al.  Computer-aided detection of breast masses on full field digital mammograms. , 2005, Medical physics.

[3]  K. Garg,et al.  Helical CT Scanning in the Diagnosis of Pulmonary Embolism , 2003, Respiration.

[4]  U. Schoepf Diagnosing pulmonary embolism: time to rewrite the textbooks , 2005, The International Journal of Cardiovascular Imaging.

[5]  D. Price Pulmonary embolism , 1976, Anaesthesia.

[6]  H MacMahon,et al.  Automated Segmentation and Visualization of the Pulmonary Vascular Tree in Spiral CT Angiography: An Anatomy-Oriented Approach Based on Three-Dimensional Image Analysis , 2001, Journal of computer assisted tomography.

[7]  Sumit K. Shah,et al.  Computer-aided lung nodule detection in CT: results of large-scale observer test. , 2005, Academic radiology.

[8]  M K Markey,et al.  Application of the mutual information criterion for feature selection in computer-aided diagnosis. , 2001, Medical physics.

[9]  D. Hosmer,et al.  A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study. , 1991, Archives of internal medicine.

[10]  K. Awai,et al.  Pulmonary nodules at chest CT: effect of computer-aided diagnosis on radiologists' detection performance. , 2004, Radiology.

[11]  J. Dalen Pulmonary embolism: what have we learned since Virchow? Natural history, pathophysiology, and diagnosis. , 2002, Chest.

[12]  Berkman Sahiner,et al.  Preliminary investigation of computer-aided detection of pulmonary embolism in three-dimensional computed tomography pulmonary angiography images. , 2005, Academic radiology.

[13]  U. Schoepf Computed tomography for pulmonary embolism diagnosis: the making of a reference standard , 2005, Journal of thrombosis and haemostasis : JTH.

[14]  K. O'Shaughnessy,et al.  Computer-aided detection in diagnostic mammography: detection of clinically unsuspected cancers. , 2004, AJR. American journal of roentgenology.

[15]  Maryellen L Giger,et al.  Update on the potential role of CAD in radiologic interpretations: are we making progress?. , 2005, Academic radiology.

[16]  Ning Wu,et al.  Lung Nodule Detection on Chest CT: Evaluation of a Computer-Aided Detection (CAD) System , 2005, Korean journal of radiology.

[17]  Kunio Doi,et al.  Computerized detection of pulmonary embolism in spiral CT angiography based on volumetric image analysis , 2002, IEEE Transactions on Medical Imaging.

[18]  Y. Kawata,et al.  Computer-aided diagnosis for pulmonary nodules based on helical CT images , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[19]  P. Cooperberg,et al.  Computer-aided detection in screening CT for pulmonary nodules. , 2006, AJR. American journal of roentgenology.

[20]  P. Taylor,et al.  Computer-aided detection in the United Kingdom National Breast Screening Programme: prospective study. , 2005, Radiology.

[21]  P. Hájek,et al.  Index of Terms , 1998 .

[22]  Jin Mo Goo,et al.  Computer-Aided Detection of Lung Nodules on Chest CT: Issues to be Solved before Clinical Use , 2005, Korean journal of radiology.

[23]  L. Wharton,et al.  THE MINOR FORMS OF PULMONARY EMBOLISM AFTER ABDOMINAL OPERATIONS: A CLINICAL AND ROENTGENOLOGIC STUDY , 1922 .