Automatic breast tumor detection in ABVS images based on convolutional neural network and superpixel patterns

Breast cancer is one of the most common female malignancies, as well as the second leading cause of mortality for women. Early detection and treatment can dramatically decrease the mortality rate. Recently, automated breast volume scanner (ABVS) has become one of the most frequently used diagnose methods for breast tumor screening because of its operator-independent and reproducible advantages. However, it is a challenging job to obtain the tumors’ accurate locations and shapes by reviewing hundreds of ABVS slices. In this paper, a novel computer-aided detection (CADe) system is developed to reduce clinicians’ reading time and improve the efficiency. The CADe system mainly contains three parts: tumor candidate acquisition, false-positive reduction and tumor segmentation. Firstly, a local phase-based approach is built to obtain breast tumor candidates for further recognition. Subsequently, a convolutional neural network (CNN) is applied to reduce false positives (FPs). The introduction of CNN can help to avoid complicated feature extraction as well as elevate the accuracy and efficiency. Finally, superpixel-based segmentation is used to outline the breast tumor. Here, superpixel-based local binary pattern (SLBP) is proposed to assist the segmentation, which improves the performance. The methods were evaluated on a clinical ABVS dataset whose abnormal cases were manually labeled by an experienced radiologist. The experiment results were mainly composed of two parts. At the FP reduction stage, the proposed CNN achieved 100% and 78.12% sensitivity with FPs/case of 2.16 and 0. At the segmentation stage, our SLBP obtained 82.34% true positive, 15.79% false positive and 83.59% Dice similarity. In summary, the proposed CADe system demonstrated promising potential to detect and outline breast tumors in ABVS images.

     

Bulging in incremental sheet forming of cold bonded multi-layered Cu clad sheet: Influence of forming conditions and bending

Bulge is a defect that causes geometrical inaccuracy and premature failure in the innovative incremental sheet forming (ISF) process. This study has two-fold objectives: (1) knowing the bulging behavior of a Cu clad tri-layered steel sheet as a function of forming conditions, and (2) analyzing the bending effect on bulging in an attempt to identify the associated mechanism. A series of ISF tests and bending analysis are performed to realize these objectives. From the cause-effect analysis, it is found that bulge formation in the layered sheet is sensitive to forming conditions in a way that bulging can be minimized utilizing annealed material and performing ISF with larger tool diameter and step size. The bending under tension analysis reveals that the formation of bulge is an outgrowth of bending moment that the forming tool applies on the sheet during ISF. Furthermore, the magnitude of bending moment depending upon the forming conditions varies from 0.046 to 10.24 N·m/m and causes a corresponding change in the mean bulge height from 0.07 to 0.91 mm. The bending moment governs bulging in layered sheet through a linear law. These findings lead to a conclusion that the bulge defect can be overcome by controlling the bending moment and the formula proposed can be helpful in this regards.     

Reoptimization framework and policy analysis for maritime inventory routing under uncertainty

We study a maritime inventory routing problem, in which shipments between production and consumption nodes are carried out by a fleet of vessels. The vessels have specific capacities and can be chartered under different agreements. The inventory levels of all consumption nodes and some production nodes should be maintained within specified bounds; for the remaining production nodes, orders should be picked up within pre-defined time windows. We propose a discrete-time mixed-integer programming model. In the face of new information and uncertainty, this optimization model has to be re-solved, as the horizon is rolled forward. We discuss how to account for different sources of uncertainty. We present a rolling-horizon reoptimization framework that allows us to study different policies that impact the quality of the implemented solution, so we can identify the optimal set of policies.