Kong, Min J.

Abstract INTRODUCTION DSC-MRI perfusion methods are commonly used to evaluate both primary and metastatic brain cancer with the creation of maps of relative cerebral blood volume (rCBV). Recently, guidelines were established to ensure a standard DSC-MRI acquisition protocol that reduces inter-site variability. The purpose of this study was to initiate the determination of rCBV benchmark values using the DSC-MRI consensus protocol in treatment-naïve brain metastases. METHODS Patients from three sites with treatment-naïve, contrast-enhancing brain metastases on MRI were considered for inclusion in this retrospective study. The MRIs included pre- and post-contrast T1w(T1+C) images obtained after administration of gadolinium-based contrast agent (GBCA) (0.1 mmol/kg), which serves as the recommended preload for the DSC-MRI data collection. A 2nd GBCA dose (0.1 mmol/kg) was administered 40-60sec after the collection of baseline GRE-EPI images using recommended settings (FA=60o, TE/TR=30ms/1100-1250ms) for 120s. Calibrated pre/post T1w difference maps (dT1) were used for delineation of enhancing lesion, and standardized (calibrated) rCBV (sRCBV) created. Mean sRCBV for metastases were compared to normal appearing white matter (NAWM) and treatment-naive glioblastoma (GBM) from a previous study. Pairwise comparisons were performed using the Mann-Whitney nonparametric test. RESULTS N=52 patients with primary histology of lung (n=27); breast (n=6); skin (n=7); gastrointestinal (GI: n=3) and genitourinary (GU: n=9) cancers were included in comparison to GBM (n=31). The mean sRCBV for all metastases (1.77+/-1.05) is significantly lower (p=0.0003) than for GBM (2.67+/-1.34) but with both statistically greater (p< 0.0001) than NAWM (0.706+/-0.163). Individually, lung (1.47+/- 0.61), breast (2.275+/-0.87), skin (2.10+/-1.22), GI (1.91+/- 0.64) and GU (2.01+/-0.63) mean sRCBV are statistically greater than for NAWM. CONCLUSION Using the consensus DSC-MRI acquisition protocol confirms use of sRCBV to identify biologically active, treatment-naive brain metastases setting benchmark values for future applications.

e16550 Background: Improved computational power and modern algorithms have generated significant interest in radiomics for cancer diagnosis and staging. Here we assess the performance of deep learning (DL) models as a means for feature extraction in combination with supervised machine learning (ML) algorithms for accurate staging and chemotherapy response assessment of bladder cancer. Methods: Deidentified grayscale CT images from bladder cancer patients scheduled to undergo radical cystectomy were included in this retrospective study. These images were manually annotated with two regional masks (normal region and cancer region). Five DL models- namely, AlexNet, GoogleNet, InceptionV3, ResNet-50, and XceptionNet pre-trained on the ImageNet dataset, a public dataset, were then fine-tuned on our bladder CT scan data to extract features. Through feature selection process, the subset of the features was used to build ML classifiers for classification. The classification was performed using five different ML classifiers, namely k-Nearest Neighbor (KNN), Naïve-Bayes (NB), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), and Decision Tree (DT). The classification task was performed with 10-fold cross-validation, and each of the experiments contained a different but not mutually exclusive subset of samples. The evaluation metrics include accuracy, sensitivity, specificity, precision, and F1-score. Results: A total of 200 deidentified grayscale CT images of 100 patients with histologically proven bladder cancer, were included in this study. For experiment (1) normal vs. cancer, the LDA classifier on XceptionNet based features provides the best performance with an accuracy of 86.07%, sensitivity of 96.75%, specificity of 69.65%, precision of 83.07%, and F1-score of 89.39%. For experiment (2) non-muscle invasive Cancer (NMIBC) vs. muscle invasive bladder cancer (MIBC), the LDA classifier on XceptionNet based features provided the best performance with an accuracy of 79.72%, sensitivity of 66.62%, specificity of 87.39%, precision of 75.58%, and F1-score of 70.81%. For experiment (3) T0 lesion vs. MIBC, the LDA classifier on XceptionNet based features provides the best performance with an accuracy of 74.96%, sensitivity of 80.51%, specificity of 70.22%, precision of 69.78%, and F1-score of 74.73%. Conclusions: Our proposed model has shown good results in differentiating normal vs cancer and promising performance in differentiating T0 vs MIBC after chemotherapy treatment. We are expanding our dataset to further improve the performance in differentiating T0 vs MIBC. In addition, we will investigate the applicability of GAN for data augmentation to address data limit. We believe the hybrid DL and ML framework may facilitates radiologists' decisions and clinical decision-making in patients with bladder cancer.