Deep learning architectures for automated image segmentation

TL;DR

This paper introduces novel deep learning architectures for improved image segmentation in natural and medical images, demonstrating superior performance over existing methods through extensive evaluation on multiple datasets.

Contribution

The paper presents two new deep learning architectures: a dilated dense encoder-decoder with spatial pyramid pooling for salient object segmentation and a dilated dense UNet for medical image lesion segmentation, both outperforming existing models.

Findings

Outperforms state-of-the-art in salient object segmentation on three datasets.

Achieves superior lesion localization and segmentation in medical images.

Provides insights into 3D medical image segmentation considerations.

Abstract

Image segmentation is widely used in a variety of computer vision tasks, such as object localization and recognition, boundary detection, and medical imaging. This thesis proposes deep learning architectures to improve automatic object localization and boundary delineation for salient object segmentation in natural images and for 2D medical image segmentation. First, we propose and evaluate a novel dilated dense encoder-decoder architecture with a custom dilated spatial pyramid pooling block to accurately localize and delineate boundaries for salient object segmentation. The dilation offers better spatial understanding and the dense connectivity preserves features learned at shallower levels of the network for better localization. Tested on three publicly available datasets, our architecture outperforms the state-of-the-art for one and is very competitive on the other two. Second, we propose and evaluate a custom 2D dilated dense UNet architecture for accurate lesion localization and segmentation in medical images. This architecture can be utilized as a stand-alone segmentation framework or used as a rich feature extracting backbone to aid other models in medical image segmentation. Our architecture outperforms all baseline models for accurate lesion localization and segmentation on a new dataset. We furthermore explore the main considerations that should be taken into account for 3D medical image segmentation, among them preprocessing techniques and specialized loss functions.