Affiliation:
1. Department of Radiology, Renji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
2. Department of Radiology, Wayne State University, Detroit, MI, USA
Abstract
Background Breast cancer is the most common cancer in women worldwide. However, it remains a difficult diagnosis problem to differentiate between benign and malignant breast lesions, especially in small early breast lesions. Purpose To assess the diagnostic value of diffusion-weighted imaging (DWI) combined with T2-weighted imaging (T2WI) for small breast cancer characterization. Material and Methods Fifty-eight patients (65 lesions) with a lesion <2 cm in diameter underwent 3.0 Tesla breast magnetic resonance imaging (MRI) including DWI and histological analysis. Three observers with varying experience levels reviewed MRI. The probability of breast cancer in each lesion on MR images was recorded with a 5-point scale. Areas under the receiver-operating characteristic curve (AUCs) were compared by using the Z test; sensitivity and specificity were determined with the Z test after adjusting for data clustering. Results AUC of T2WI and DWI (Observer 1, 0.95; Observer 2, 0.91; Observer 3, 0.83) was greater than that of T2WI (Observer 1, 0.80; Observer 2, 0.74; Observer 3, 0.70) for all observers ( P < 0.0001 in all comparisons). Sensitivity of T2WI and DWI (Observer 1, 90%; Observer 2, 93%; and Observer 3, 86%) was greater than that of T2WI alone (Observer 1, 76%; Observer 2, 83%; Observer 3, 79%) for all observers ( P < 0.0001 in all comparisons). Specificity of T2WI and DWI was greater than that of T2WI alone for observer 1 (89% vs. 72%, P < 0.01) and observer 2 (94% vs. 78%, P < 0.001). Conclusion DWI combined with T2WI can improve the diagnostic performance of MRI in small breast cancer characterization. It should be considered selectively in the preoperative evaluation of patients with small lesions of the breast.
Subject
Radiology, Nuclear Medicine and imaging,General Medicine,Radiological and Ultrasound Technology
Cited by
14 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献