A Malich1, D Hentrich, A Hansch, J Böttcher, U Teichgräber. 1. Institut für Radiologie, Südharz-Krankenhaus Nordhausen, Dr.-R.-Koch-Str. 39, 99734, Nordhausen, Deutschland. ansgar.malich@shk-ndh.de
Abstract
BACKGROUND: A recently developed CAD software which highlights intravascular thrombotic structures from multislice computed tomography (MSCT) data was tested regarding feasibility, interobserver reliability and effect on radiology reports. MATERIAL AND METHODS: The CAD system ImageChecker® CT-Lung was tested in a randomized double-blinded study on 160 MSCT datasets (standardized technical conditions) performed for suspected pulmonary embolism (PE). The CAD data and images were analyzed by three radiologists in an independent and blinded fashion. RESULTS: The data from all 160 cases could be analyzed and 604 CAD prompts were set. Using the CAD analysis significantly more PEs were found in the peripheral pulmonary arterial circulation than described in the initial report. In 38 cases the 3 radiologists in consensus scored the images with the CAD adjunct as PE positive in peripheral vessels, which were initially reported as negative. Despite differences in the evaluation between two radiologists the amended assessment of the imaging data using the CAD softwear was reliable. There was a significant correlation between D-dimer values and the number of embolic structures detected by the CAD analysis. CONCLUSION: The recently developed CAD system is a useful adjunct as second reader to detect subtle emboli in peripheral vessels of MSCT datasets.
BACKGROUND: A recently developed CAD software which highlights intravascular thrombotic structures from multislice computed tomography (MSCT) data was tested regarding feasibility, interobserver reliability and effect on radiology reports. MATERIAL AND METHODS: The CAD system ImageChecker® CT-Lung was tested in a randomized double-blinded study on 160 MSCT datasets (standardized technical conditions) performed for suspected pulmonary embolism (PE). The CAD data and images were analyzed by three radiologists in an independent and blinded fashion. RESULTS: The data from all 160 cases could be analyzed and 604 CAD prompts were set. Using the CAD analysis significantly more PEs were found in the peripheral pulmonary arterial circulation than described in the initial report. In 38 cases the 3 radiologists in consensus scored the images with the CAD adjunct as PE positive in peripheral vessels, which were initially reported as negative. Despite differences in the evaluation between two radiologists the amended assessment of the imaging data using the CAD softwear was reliable. There was a significant correlation between D-dimer values and the number of embolic structures detected by the CAD analysis. CONCLUSION: The recently developed CAD system is a useful adjunct as second reader to detect subtle emboli in peripheral vessels of MSCT datasets.
Authors: John Eng; Jerry A Krishnan; Jodi B Segal; Dennis T Bolger; Leonardo J Tamariz; Michael B Streiff; Mollie W Jenckes; Eric B Bass Journal: AJR Am J Roentgenol Date: 2004-12 Impact factor: 3.959
Authors: Kevin N Blackmon; Charles Florin; Luca Bogoni; Joshua W McCain; James D Koonce; Heon Lee; Gorka Bastarrika; Christian Thilo; Philip Costello; Marcos Salganicoff; U Joseph Schoepf Journal: Eur Radiol Date: 2011-01-13 Impact factor: 5.315
Authors: John A Heit; W Michael O'Fallon; Tanya M Petterson; Christine M Lohse; Marc D Silverstein; David N Mohr; L Joseph Melton Journal: Arch Intern Med Date: 2002-06-10