Analysis of domain shift in whole prostate gland, zonal and lesions segmentation and detection, using multicentric retrospective data.

Despite being one of the most prevalent forms of cancer, prostate cancer (PCa) shows a significantly high survival rate, provided there is timely detection and treatment. Computational methods can help make this detection process considerably faster and more robust. However, some modern machine-learning approaches require accurate segmentation of the prostate gland and the index lesion. Since performing manual segmentations is a very time-consuming task, and highly prone to inter-observer variability, there is a need to develop robust semi-automatic segmentation models. In this work, we leverage the large and highly diverse ProstateNet dataset, which includes 638 whole gland and 461 lesion segmentation masks, from 3 different scanner manufacturers provided by 14 institutions, in addition to other 3 independent public datasets, to train accurate and robust segmentation models for the whole prostate gland, zones and lesions. We show that models trained on large amounts of diverse data are better at generalizing to data from other institutions and obtained with other manufacturers, outperforming models trained on single-institution single-manufacturer datasets in all segmentation tasks. Furthermore, we show that lesion segmentation models trained on ProstateNet can be reliably used as lesion detection models.

Pulmonary side effects of nivolumab in a patient treated for gynecological malignancy.

We report a case of a 26-year-old female patient treated with nivolumab for peritoneal implants from small-cell ovarian cancer. Pulmonary changes were depicted on follow-up examinations, namely, consolidations and scattered ground glass opacities, without any remarkable pulmonary complaints or laboratory changes. Both infectious and progressive diseases were ruled out. The condition was attributed to nivolumab side effect, and total remission was documented, followed by corticosteroids administration. We present clinical details and imaging findings, followed by a discussion where pulmonary toxicity can often be mistaken with infection or secondary involvement.

MR imaging versus PET/CT for evaluation of pancreatic lesions.

PURPOSE: To retrospectively determine the diagnostic accuracy of magnetic resonance imaging (MRI) and combined positron emission tomography/computed tomography (PET/CT) in the differential diagnosis of benign and malignant pancreatic lesions. MATERIALS AND METHODS: Twenty-seven patients (15 women/12 men, mean age 56.5 years) with MR imaging and PET/CT studies performed to differentiate benign and malignant pancreatic lesions were identified between October 2008 and October 2010. Both MR and PET/CT data sets were retrospectively and blindly evaluated by two independent readers (4 readers total) with different degrees of experience, using a visual five-point score system. The results were correlated with final diagnosis obtained by histopathology. RESULTS: 17 patients had malignant diseases and 10 patients had benign diseases. Depending on the observer, the sensitivity, specificity, positive predictive value and negative predictive value of MRI varied between 88-94%, 50-80%, 75-89% and 71-89% respectively. Sensitivities, specificities, positive predictive values and negative predictive values of PET/CT were 73%, 56%, 73% and 56% respectively. The diagnostic accuracy of MR for the differential diagnosis of pancreatic lesions was 74-89%, compared with 67% for PET/CT. The weighted Cohen's kappa coefficient was 0.47 at MR and 0.53 at PET/CT. CONCLUSION: MRI achieved higher sensitivity and specificity in the differential diagnosis of pancreatic lesions.