M B I Lobbes1, H K P Mulder2, M Rousch3, W H Backes2, J E Wildberger2, C R L P N Jeukens2. 1. Maastricht University Medical Center, Departments of Radiology and Nuclear Medicine, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands; GROW School for Oncology and Developmental Biology, P.O. Box 616, 6200 MD Maastricht, The Netherlands. Electronic address: marc.lobbes@mumc.nl. 2. Maastricht University Medical Center, Departments of Radiology and Nuclear Medicine, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands. 3. Maastricht University Medical Center, Department of Pathology, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands.
Abstract
PURPOSE: Contrast-enhanced spectral mammography (CESM) is diagnostically superior to full-field digital mammography. An important improvement for CESM would be the ability to quantify enhancement. In this proof-of-concept study we present a method for quantifying CESM enhancement. METHODS AND MATERIALS: We developed a custom-made quantifier tool (I-STRIP) containing five chambers, each filled with a different iodine mass thickness (IMT). CESM grey values of the recombined image (CGV) in the I-STRIP were used to quantify breast IMT. We evaluated the I-STRIP's accuracy using a dedicated breast phantom containing chambers with known IMT's. Furthermore, we tested the effect of the I-STRIP on image quality and clinical use in five patients. Retrospectively, we studied whether current CESM protocols could distinguish between malignant and benign lesions in terms of CGV. RESULTS: Phantom experiments showed that quantification was independent of chamber height and size, phantom thickness and I-STRIP position for different IMT's (1.5, 3.0 and 7.5 mg l/cm2). Near the phantom's periphery accuracy was found to be lower due to the breast-within-breast artifact. In the clinical setting (n = 5), workflow and image quality were not influenced by the I-STRIP. The mean IMT of these invasive breast cancers was 2.1 mg l/cm2 (range 1.3-3.4 mg l/cm2). Malignant lesions showed significantly higher CGV's than benign lesions (p = 0.002). CONCLUSION: We showed in both phantom and clinical experiments that CESM quantification is feasible, without influencing workflow or image quality. The current CESM imaging protocol seems to be able to distinguish between benign and malignant breast lesions in terms of CGV.
PURPOSE: Contrast-enhanced spectral mammography (CESM) is diagnostically superior to full-field digital mammography. An important improvement for CESM would be the ability to quantify enhancement. In this proof-of-concept study we present a method for quantifying CESM enhancement. METHODS AND MATERIALS: We developed a custom-made quantifier tool (I-STRIP) containing five chambers, each filled with a different iodine mass thickness (IMT). CESM grey values of the recombined image (CGV) in the I-STRIP were used to quantify breast IMT. We evaluated the I-STRIP's accuracy using a dedicated breast phantom containing chambers with known IMT's. Furthermore, we tested the effect of the I-STRIP on image quality and clinical use in five patients. Retrospectively, we studied whether current CESM protocols could distinguish between malignant and benign lesions in terms of CGV. RESULTS: Phantom experiments showed that quantification was independent of chamber height and size, phantom thickness and I-STRIP position for different IMT's (1.5, 3.0 and 7.5 mg l/cm2). Near the phantom's periphery accuracy was found to be lower due to the breast-within-breast artifact. In the clinical setting (n = 5), workflow and image quality were not influenced by the I-STRIP. The mean IMT of these invasive breast cancers was 2.1 mg l/cm2 (range 1.3-3.4 mg l/cm2). Malignant lesions showed significantly higher CGV's than benign lesions (p = 0.002). CONCLUSION: We showed in both phantom and clinical experiments that CESM quantification is feasible, without influencing workflow or image quality. The current CESM imaging protocol seems to be able to distinguish between benign and malignant breast lesions in terms of CGV.