Purpose: Reliable automated segmentation of the prostate is indispensable for image-guided prostate interventions. However, the segmentation task is challenging due to inhomogeneous intensity distributions, variation in prostate anatomy, among other problems. Manual segmentation can be time-consuming and is subject to inter-and intraobserver variation. We developed an automated deep learning-based method to address this technical challenge. Methods: We propose a three-dimensional (3D) fully convolutional networks (FCN) with deep supervision and group dilated convolution to segment the prostate on magnetic resonance imaging (MRI). In this method, a deeply supervised mechanism was introduced into a 3D FCN to effectively alleviate the common exploding or vanishing gradients problems in training deep models, which forces the update process of the hidden layer filters to favor highly discriminative features. A group dilated convolution which aggregates multiscale contextual information for dense prediction was proposed to enlarge the effective receptive field of convolutional neural networks, which improve the prediction accuracy of prostate boundary. In addition, we introduced a combined loss function including cosine and cross entropy, which measures similarity and dissimilarity between segmented and manual contours, to further improve the segmentation accuracy. Prostate volumes manually segmented by experienced physicians were used as a gold standard against which our segmentation accuracy was measured. Results: The proposed method was evaluated on an internal dataset comprising 40 T2-weighted prostate MR volumes. Our method achieved a Dice similarity coefficient (DSC) of 0.86 AE 0.04, a mean surface distance (MSD) of 1.79 AE 0.46 mm, 95% Hausdorff distance (95%HD) of 7.98 AE 2.91 mm, and absolute relative volume difference (aRVD) of 15.65 AE 10.82. A public dataset (PROMISE12) including 50 T2-weighted prostate MR volumes was also employed to evaluate our approach. Our method yielded a DSC of 0.88 AE 0.05, MSD of 1.02 AE 0.35 mm, 95% HD of 9.50 AE 5.11 mm, and aRVD of 8.93 AE 7.56. Conclusion: We developed a novel deeply supervised deep learning-based approach with a group dilated convolution to automatically segment the MRI prostate, demonstrated its clinical feasibility, and validated its accuracy against manual segmentation. The proposed technique could be a useful tool for image-guided interventions in prostate cancer.

No abstract

Purpose: Transrectal ultrasound (TRUS) is a versatile and real-time imaging modality that is commonly used in image-guided prostate cancer interventions (e.g., biopsy and brachytherapy). Accurate segmentation of the prostate is key to biopsy needle placement, brachytherapy treatment planning, and motion management. Manual segmentation during these interventions is time-consuming and subject to inter-and intraobserver variation. To address these drawbacks, we aimed to develop a deep learning-based method which integrates deep supervision into a three-dimensional (3D) patch-based V-Net for prostate segmentation. Methods and materials: We developed a multidirectional deep-learning-based method to automatically segment the prostate for ultrasound-guided radiation therapy. A 3D supervision mechanism is integrated into the V-Net stages to deal with the optimization difficulties when training a deep network with limited training data. We combine a binary cross-entropy (BCE) loss and a batch-based Dice loss into the stage-wise hybrid loss function for a deep supervision training. During the segmentation stage, the patches are extracted from the newly acquired ultrasound image as the input of the well-trained network and the well-trained network adaptively labels the prostate tissue. The final segmented prostate volume is reconstructed using patch fusion and further refined through a contour refinement processing. Results: Forty-four patients' TRUS images were used to test our segmentation method. Our segmentation results were compared with the manually segmented contours (ground truth). The mean prostate volume Dice similarity coefficient (DSC), Hausdorff distance (HD), mean surface distance (MSD), and residual mean surface distance (RMSD) were 0.92 AE 0.03, 3.94 AE 1.55, 0.60 AE 0.23, and 0.90 AE 0.38 mm, respectively. Conclusion: We developed a novel deeply supervised deep learning-based approach with reliable contour refinement to automatically segment the TRUS prostate, demonstrated its clinical feasibility, and validated its accuracy compared to manual segmentation. The proposed technique could be a useful tool for diagnostic and therapeutic applications in prostate cancer.

Multi-modality cancer treatments that include chemotherapy, radiation therapy, and targeted agents are highly effective therapies. Their use, especially in combination, is limited by the risk of significant cardiac toxicity. The current paradigm for minimizing cardiac morbidity, based on serial cardiac function monitoring, is suboptimal. An alternative approach based on biomarker testing, has emerged as a promising adjunct and a potential substitute to routine echocardiography. Biomarkers, most prominently cardiac troponins and natriuretic peptides, have been evaluated for their ability to describe the risk of potential cardiac dysfunction in clinically asymptomatic patients. Early rises in cardiac troponin concentrations have consistently predicted the risk and severity of significant cardiac events in patients treated with anthracycline-based chemotherapy. Biomarkers represent a novel, efficient, and robust clinical decision tool for the management of cancer therapy-induced cardiotoxicity. This article aims to review the clinical evidence that supports the use of established biomarkers such as cardiac troponins and natriuretic peptides, as well as emerging data on proposed biomarkers.

Importance The treatment of multiple brain metastases (MBM) from melanoma is controversial and includes surgical resection, stereotactic radiosurgery and whole brain radiation. Several new classes of agents have revolutionized the treatment of metastatic melanoma allowing for subsets of patients to have long-term survival. Given this, management of MBM from melanoma is continually evolving. Objective To review the current evidence regarding the treatment of MBM from melanoma. Evidence Review The Pubmed database was searched using combinations of search terms and synonyms for melanoma, brain metastases, radiation, chemotherapy, immunotherapy and targeted therapy published between January 1, 1995 and January 1, 2015. Articles were selected for inclusion based on targeted keyword searches, manual review of bibliographies, and whether the article was a clinical trial, large observational study, or retrospective study focusing on melanoma brain metastases. Of 2243 articles initially identified, 110 were selected for full review. Of these, the most pertinent 76 articles were included. Findings Patients with newly diagnosed MBM can be treated with various modalities, either alone or in combination. Level 1 evidence supports the use of radiosurgery alone, whole brain radiation therapy (WBRT), and radiosurgery with WBRT. Though the addition of WBRT to SRS improves the overall brain relapse rate, WBRT has no significant impact on overall survival and has detrimental neurocognitive outcomes. Cytotoxic chemotherapy has largely been ineffective; targeted therapies and immunotherapies have reported to have high response rates and deserve further attention in the setting of larger clinical trials. Further studies are needed to fully evaluate the efficacy of these novel regimens in combination with radiation therapy. Conclusions and Relevance At this time, the standard management for patients with MBM from melanoma includes SRS, WBRT, or combination of both. Emerging data exists to support the notion that SRS in combination with targeted therapies or immune therapy may obviate the need for whole brain radiation and prospective studies are required to fully evaluate the efficacy of these novel regimens in combination with radiation therapy.