This paper presents a SVM based computer-aided diagnosis (CAD) system for the characterization of clustered microcalcifications in digitized mammograms. First, the region of interest (ROI) in mammogram is enhanced usi...
详细信息
This paper presents a SVM based computer-aided diagnosis (CAD) system for the characterization of clustered microcalcifications in digitized mammograms. First, the region of interest (ROI) in mammogram is enhanced using morphological enhancement (MORPHeN) method. Second, pixels in potential microcalcification regions are segmented out by using edge detection and morphological operations. Third, features based on shape, texture and statistical properties areextracted from each region. Finally, these features are fed to a SVM based classifier for identifying the clusters as either benign or malignant. The SVM with RbF kernel gave A z =0.9803 with 97% accuracy and the SVM with polynomial kernel gave A z =0.9541 with 95% accuracy
Accuracy requirements are usually determined as a percentage of the specification range of the measured part or process. Setting accuracy requirements in this manner results in a wide and unpredictab.e range of false ...
Accuracy requirements are usually determined as a percentage of the specification range of the measured part or process. Setting accuracy requirements in this manner results in a wide and unpredictab.e range of false rejection and acceptance probabilities. This causes extra costs due to either: 1) over specification of measurement systems accuracy requirements;2) time, effort, retesting, and resolution of false rejections;or 3) system degradation caused by false acceptance of out-of-specification parts. Achieving a consistent and known risk of false acceptance is only possib.eby considering the measured process C(pk), the process's mean in relation to the center of the specification range, and the measurement system error distribution. This paper presents a method for calculating the probabilities of false rejection and false acceptance for a normal process which is measured with, alternately, uniform and normally distributed error. It is shown that under most conditions uniform error causes 20% to 30% higher false rejection and acceptance probabilities. Thus, knowledge of measurement error distribution could provide lower total production cost.
A ship design methodology is presented for developing hull forms that attain improved performance in both seakeeping and resistance. Contrary to traditional practice, the methodology starts with developing a seakeepin...
A ship design methodology is presented for developing hull forms that attain improved performance in both seakeeping and resistance. Contrary to traditional practice, the methodology starts with developing a seakeeping-optimized hull form without making concessions to other performance considerations, such as resistance. The seakeeping-optimized hull is then modified to improve other performance characteristics without degrading the seakeeping. Presented is a point-design example produced by this methodology. Merits of the methodology and the point design are assessed on thebasis of theoretical calculations and model experiments. This methodology is an integral part of the Hull Form Design System (HFDS) being developed for computer-supported naval ship design. The modularized character of HFDS and its application to hull form development are discussed.
暂无评论