The Genetic Algorithm (GA) is one of the most popular heuristic methods due to its natural and fast implementation. However, at the same time, it has the disadvantage of poor optimization. To improve performance, it’s necessary avoid stuck in local maximums throught choosing proper methods and parameters that vary for each application. In photonic devices, although the GA has been recently used to optimize passive silicon Y-branches, its performance is still trailing behind other optimization algorithms based on swarms, for instance. In this work, we present a new three-part heuristic method for optimizing Y-branches. We used the Finite-difference Time-domain (FDTD) method and the Particle Swarm Optimization (PSO) to generate an optimal data set as initial population for the GA. Considering an adequate population model, we demonstrate improvement in the performance for the design of a Y-branch through the GA. Next, we used a variation of a gradient-based search method to fine-tune the final parameters to find the absolute maximum. As a result, we produced new non-intuitive Y-branch devices with on-chip areas smaller than 2µm2 and excess loss down to 0.05 dB @1550 nm for the TE mode. A complete study of fabrication feasibility and uv-lithography typical fabrication errors and its effects on the bandwidth will be shown at the time of the conference. Our method will be compared against other widely-used heuristic methods in photonic device design in terms of number of iterations.
Breast parenchymal density is considered a strong indicator of cancer risk. However, measures of breast density are often qualitative and require the subjective judgment of radiologists. This work proposes a supervised algorithm to automatically assign a BI-RADS breast density score to a digital mammogram. The algorithm applies principal component analysis to the histograms of a training dataset of digital mammograms to create four different spaces, one for each BI-RADS category. Scoring is achieved by projecting the histogram of the image to be classified onto the four spaces and assigning it to the closest class. In order to validate the algorithm, a training set of 86 images and a separate testing database of 964 images were built. All mammograms were acquired in the craniocaudal view from female patients without any visible pathology. Eight experienced radiologists categorized the mammograms according to a BIRADS score and the mode of their evaluations was considered as ground truth. Results show better agreement between the algorithm and ground truth for the training set (kappa=0.74) than for the test set (kappa=0.44) which suggests the method may be used for BI-RADS classification but a better training is required.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.