ABSTRACT
PURPOSE: Placental site trophoblastic tumor (PSTT) is a rare form of gestational trophoblastic neoplasm with few previous imaging case reports. We report multiparametric MRI findings in four cases of PSTT with special emphasis on the "pseudo-myometrial thinning" underlying the tumor. METHODS: We reviewed multiparametric MRI and pathologic findings in four cases of PSTT from four institutions. Signal intensity, enhancement pattern, margins, and location of the tumors were evaluated, and myometrial thickness underlying the tumor and normal myometrial thickness contralateral to the tumor were measured on MRI. The myometrial thickness underlying the tumor was also measured in the resected specimen and compared with the myometrial thickness measured on MRI using the Friedman test. RESULTS: All tumors showed heterogeneous signal intensity on T1-weighted imaging, T2-weighted imaging (T2WI), and diffusion-weighted imaging. Three of the four tumors had a hypervascular area on dynamic contrast-enhanced (DCE) MRI. A hypointense rim on T2WI and DCE-MRI was seen in all tumors. All tumors protruded into the uterine cavity to varying degrees and extended into the myometrium close to the serosa. The myometrial thickness underlying the tumor measured on MRI (median thickness, 1.2 mm) was significantly thinner than that measured on pathology (median thickness, 9.5 mm) and normal myometrial thickness contralateral to the tumor on MRI (median thickness, 10.3 mm) (P = 0.02), and there was no significant difference between the latter two. CONCLUSIONS: The thickness of the myometrium underlying the tumor on MRI was approximately one tenth of the thickness on pathology. Thus, the tumors appeared to have almost transmural invasion even when pathologically located within the superficial myometrium. This "pseudo-thinning" of the underlying myometrium and the hypointense rim on MRI could be caused by focal compression of the myometrium by the tumor, possibly due to the fragility of the myometrium at the placental site.
ABSTRACT
PURPOSE: To evaluate the diagnostic performance of deep learning using the Residual Networks 50 (ResNet50) neural network constructed from different segmentations for distinguishing malignant and benign non-mass enhancement (NME) on breast magnetic resonance imaging (MRI) and conduct a comparison with radiologists with various levels of experience. MATERIALS AND METHODS: A total of 84 consecutive patients with 86 lesions (51 malignant, 35 benign) presenting NME on breast MRI were analyzed. Three radiologists with different levels of experience evaluated all examinations, based on the Breast Imaging-Reporting and Data System (BI-RADS) lexicon and categorization. For the deep learning method, one expert radiologist performed lesion annotation manually using the early phase of dynamic contrast-enhanced (DCE) MRI. Two segmentation methods were applied: a precise segmentation was carefully set to include only the enhancing area, and a rough segmentation covered the whole enhancing region, including the intervenient non-enhancing area. ResNet50 was implemented using the DCE MRI input. The diagnostic performance of the radiologists' readings and deep learning were then compared using receiver operating curve analysis. RESULTS: The ResNet50 model from precise segmentation achieved diagnostic accuracy equivalent [area under the curve (AUC) = 0.91, 95% confidence interval (CI) 0.90, 0.93] to that of a highly experienced radiologist (AUC = 0.89, 95% CI 0.81, 0.96; p = 0.45). Even the model from rough segmentation showed diagnostic performance equivalent to a board-certified radiologist (AUC = 0.80, 95% CI 0.78, 0.82 vs. AUC = 0.79, 95% CI 0.70, 0.89, respectively). Both ResNet50 models from the precise and rough segmentation exceeded the diagnostic accuracy of a radiology resident (AUC = 0.64, 95% CI 0.52, 0.76). CONCLUSION: These findings suggest that the deep learning model from ResNet50 has the potential to ensure accuracy in the diagnosis of NME on breast MRI.