Dynamic contrast-enhanced magnetic resonance imaging (MRI) has been established as a valuable tool for the detection of breast cancer. There is evidence suggesting that diffusion-weighted imaging (DWI) may be useful to distinguish between malignant and benign breast lesions. We seek to evaluate the ability of DWI to differentiate between malignant and benign breast lesions at 3 T. Dynamic contrast-enhanced MRI and DWI of the breasts were performed in 31 female patients (age: mean, 46 years; range, 34-69 years) with suspected breast lesions on mammography and ultrasound using a 3-T scanner (MAGNETOM Tim Trio; Siemens Medical Solutions, Erlangen, Germany). Each lesion was assigned as either malignant or benign, blinded to the results of mammography and ultrasound, according to their imaging characteristics on contrast-enhanced MRI, DWI, and apparent diffusion coefficient (ADC) measurements. Tissue samples were obtained from all lesions by either needle or excision biopsy. Using histological results as the gold standard, the diagnostic accuracies of the dynamic contrast-enhanced MRI, DWI, and ADC were calculated and compared. All breast lesions (n = 31) were identified on both the dynamic contrast-enhanced MRI and DWI scans. The threshold ADC value was determined to be 0.00121 mm2/s, below which a lesion was considered malignant. The sensitivities/specificities of the dynamic contrast-enhanced MRI, qualitative DWI, and quantitative ADC were 95%/91%, 95%/63.6%, and 90%/91%, respectively. The differences in sensitivities, specificities, positive and negative predictive values, and diagnostic accuracies between the 3 examinations were statistically insignificant. Diffusion-weighted imaging at 3 T is highly sensitive in the detection of malignant breast lesions even with qualitative assessment alone, whereas ADC measurement offers quantitative assessment and increases the specificity to more than 90%. Further studies involving a larger cohort size and a wider spectrum of breast lesions are indicated.
1 May 2019 in Journal of Cardiovascular Computed Tomography
Radiomics has the potential to aid prostate cancer (PC) diagnoses and prediction by analyzing and modeling quantitative features extracted from clinical imaging. However, its reliability has been a concern, possibly due to its high-dimensional nature. This study aims to quantitatively investigate the impact of randomly generated irrelevant features on MRI radiomics feature selection, modeling, and performance by progressively adding randomly generated features. Two multiparametric-MRI radiomics PC datasets were used (dataset 1 (n = 260), dataset 2 (n = 100)). The endpoint was to differentiate pathology-confirmed clinically significant (Gleason score (GS) ≥ 7) from insignificant (GS < 7) PC. Random features were generated at 12 levels with a 10% increment from 0% to 100% and an additional 5%. Three feature selection algorithms and two classifiers were used to build the models. The area under the curve and accuracy were used to evaluate the model’s performance. Feature importance was calculated to assess features’ contributions to the models. The metrics of each model were compared using an ANOVA test with a Bonferroni correction. A slight tendency to select more random features with the increasing number of random features introduced to the datasets was observed. However, the performance of the radiomics-built models was not significantly affected, which was partially due to the higher contribution of radiomics features toward the models compared to the random features. These reliability effects also vary among datasets. In conclusion, while the inclusion of additional random features may still slightly impact the performance of the feature selection, it may not have a substantial impact on the MRI radiomics model performance.