OBJECTIVES: To evaluate a software algorithm highlighting vascular iodine distribution in dual energy (DE) computed tomography angiography (CTA) for the diagnosis of pulmonary embolism (PE). MATERIAL AND METHODS: Pulmonary DE-CTA of 16 patients with PE and 16 patients without PE were analyzed using a software algorithm highlighting vascular iodine distribution. The algorithm color-codes lung vessels depending on their local iodine distribution on a 2-color scale. The diagnostic performance of the software for the detection of PE was assessed on patient and segmental basis by consensus reading of 2 blinded radiologists. The reading of the standard CTA data by an independent third radiologist and clinical follow-up served as the standard of reference for the diagnosis of PE. RESULTS: Of 576 analyzed segments CTA revealed 88 diseased lung segments with 1 or more emboli. The software correctly highlighted 62 segments as positive. Twenty-six segments with PE were not highlighted. Seventy-five segments were highlighted false positive. All 16 patients with PE were identified as positive, but 1 of these patients had no true positive finding on a segmental basis and was therefore classified as false negative. Twenty-three segments in 8 patients without PE were highlighted as positive. Sensitivity, specificity, positive predictive value, and negative predictive value of the software algorithm were 93.8%, 50%, 65.2%, 88.9% per patient and 70.5%, 84.6%, 45.3%, 94.1% per segment, respectively. CONCLUSION: Additional review of the DE-CTA with a dedicated software algorithm highlighting the vascular iodine distribution has a high negative predictive value important for exclusion of segmental PE.
OBJECTIVES: To evaluate a software algorithm highlighting vascular iodine distribution in dual energy (DE) computed tomography angiography (CTA) for the diagnosis of pulmonary embolism (PE). MATERIAL AND METHODS: Pulmonary DE-CTA of 16 patients with PE and 16 patients without PE were analyzed using a software algorithm highlighting vascular iodine distribution. The algorithm color-codes lung vessels depending on their local iodine distribution on a 2-color scale. The diagnostic performance of the software for the detection of PE was assessed on patient and segmental basis by consensus reading of 2 blinded radiologists. The reading of the standard CTA data by an independent third radiologist and clinical follow-up served as the standard of reference for the diagnosis of PE. RESULTS: Of 576 analyzed segments CTA revealed 88 diseased lung segments with 1 or more emboli. The software correctly highlighted 62 segments as positive. Twenty-six segments with PE were not highlighted. Seventy-five segments were highlighted false positive. All 16 patients with PE were identified as positive, but 1 of these patients had no true positive finding on a segmental basis and was therefore classified as false negative. Twenty-three segments in 8 patients without PE were highlighted as positive. Sensitivity, specificity, positive predictive value, and negative predictive value of the software algorithm were 93.8%, 50%, 65.2%, 88.9% per patient and 70.5%, 84.6%, 45.3%, 94.1% per segment, respectively. CONCLUSION: Additional review of the DE-CTA with a dedicated software algorithm highlighting the vascular iodine distribution has a high negative predictive value important for exclusion of segmental PE.
Authors: Long Jiang Zhang; Guang Ming Lu; Felix G Meinel; Andrew D McQuiston; James G Ravenel; U Joseph Schoepf Journal: Eur Radiol Date: 2015-03-13 Impact factor: 5.315
Authors: Long Jiang Zhang; Chang Sheng Zhou; U Joseph Schoepf; Hui Xue Sheng; Sheng Yong Wu; Aleksander W Krazinski; Justin R Silverman; Felix G Meinel; Yan E Zhao; Zong Jun Zhang; Guang Ming Lu Journal: Eur Radiol Date: 2013-06-13 Impact factor: 5.315
Authors: Robbert W van Hamersvelt; Martin J Willemink; Pim A de Jong; Julien Milles; Alain Vlassenbroek; Arnold M R Schilham; Tim Leiner Journal: Eur Radiol Date: 2017-01-25 Impact factor: 5.315