RESUMO
Purpose: The current study aimed to evaluate the efficiency of dynamic contrast-enhanced (DCE) MRI visual features in classifying benign liver nodules and hepatocellular carcinoma (HCC) using a machine learning model. Methods: 115 LI-RADS3, 137 LI-RADS4, and 140 LI-RADS5 nodules were included (392 nodules from 245 patients), which were evaluated by follow-up imaging for LR-3 and pathology results for LR-4 and LR-5 nodules. Data was collected retrospectively from 3 T and 1.5 T MRI scanners. All the lesions were categorized into 124 benign and 268 HCC lesions. Visual features included tumor size, arterial-phase hyper-enhancement (APHE), washout, lesion segment, mass/mass-like, and capsule presence. Gini-importance method extracted the most important features to prevent over-fitting. Final dataset was split into training(70%), validation(10%), and test dataset(20%). The SVM model was used to train the classifying algorithm. For model validation, 5-fold cross-validation was utilized, and the test data set was used to assess the final accuracy. The area under the curve and receiver operating characteristic curves were used to assess the performance of the classifier model. Results: For test dataset, the accuracy, sensitivity, and specificity values for classifying benign and HCC lesions were 82%,84%, and 81%, respectively. APHE, washout, tumor size, and mass/mass-like features significantly differentiated benign and HCC lesions with p-value < .001. Conclusions: The developed classification model employing DCE-MRI features showed significant performance of visual features in classifying benign and HCC lesions. Our study also highlighted the significance of mass and mass-like features in addition to LI-RADS categorization. For future work, this study suggests developing a deep-learning algorithm for automatic lesion segmentation and feature assessment to reduce lesion categorization errors.
RESUMO
BACKGROUND: Endometrial cancer (EC) is the eighth most prevalent cancer globally. T2-weighted magnetic resonance imaging (MRI) and diffusion-weighted imaging (DWI) help anatomical localization and local staging of lesions. The present study was performed to assess the diagnostic value of the simultaneous use of T2 and DWI techniques in EC evaluation. METHODS: Seventy-eight histopathological-proven EC cases were included in this study. Patients were assessed using a complete MRI exam, including T2 and DWI. The myometrial invasion, cervical, serosal or adnexal, vaginal or parametrial, and pelvic lymph node involvements and accuracy, sensitivity, specificity, and positive and negative predictive values were evaluated in each sequence distinctly and was compared with the pathology findings and full standard protocol using post-contrast multiphasic contrast-enhanced series. RESULTS: Deep myometrial invasion in EC cases was detected in 38.5% by T2-DWI and 37.2% by pathology. The pathology diagnosed cervical, serosal, and vaginal involvements and pelvic lymph node metastases in 20.5%, 7.7%, 6.4% and 11.5% of cases respectively, while the numbers for T2-DWI were 26.9%, 7.7%, 7.7%, and 15.4%, respectively. The accuracy, sensitivity, specificity, PPV, and NPV of T2-DWI in the diagnosis of myometrial invasion were 93.5%, 93.1%, 93.8%, 90%, and 93.8%, respectively. A slightly higher Kappa coefficient of DWI (0.973) in the diagnosis of myometrial invasion was identified compared to T2 (0.946). The T2-DWI technique had a 52.6% intraclass correlation coefficient in the diagnosis of IA stage. CONCLUSION: The simultaneous consideration of T2 and DWI technique may signify a noninvasive, rapid, safe, and accurate approach for precisely assessing myometrial invasion and EC staging. Elimination of intravenous contrast material result in prevention of contrast related side effects beside significant cost reduction for health care systems and patients with a comparable result to contrast enhanced MRI.
