Date of Award
Spring 5-11-2024
Degree Name
Master of Science (MS)
Degree Type
Thesis
Abstract
Machine learning is commonly used in biomedical image analysis, as it allows automated image segmentation and identification that minimizes the need for tedious human involvement. Drosophila melanogaster is often used as a cardiac disease model, where optical coherence microscopy (OCM) is used to image and analyze its beating dynamics. As OCM often generates a large volume of images, automated image segmentation is necessary to quantify the heart beating efficiently. Our most recent heart segmentation model, FlyNet 2.0+, is a fully convolutional LSTM U-Net model. However, the performance of the model diminishes in the presence of artifacts, such as image reflection and heart movement, resulting in time-consuming manual intervention for mask correction. Therefore, we developed the FlyNet 3.0 model with integrated attention gates in skip connections between each level of the LSTM U-Net model. The attention model adaptively adjusts and automatically learns to focus on the target structure, the heart area. Compared to the previous model, Flynet 3.0 increases the prediction intersection over union (IOU) accuracy from 0.86 to 0.89 for images with reflection artifacts and from 0.81 to 0.89 for those depicting heart movement. Furthermore, we have expanded the functionalities of OCM analyses through automated and dynamic heart wall thickness measurements, which we have validated using a Drosophila model of cardiac hypertrophy.
Language
English (en)
Chair
Chao Zhou
Committee Members
Song Hu Quing Zhu