Training on Polar Image Transformations Improves Biomedical Image Segmentation

Autor: Irena Galić, Marija Habijan, Marin Bencevic, Danilo Babin
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Technology and Engineering
General Computer Science
Computer science
020209 energy
ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION
Convolutional neural network
02 engineering and technology
medical image processing
Biomedical imaging
0202 electrical engineering
electronic engineering
information engineering
Medical imaging
Training
General Materials Science
Segmentation
Image segmentation
medical image segmentation
Artificial neural network
Medical diagnostic imaging
business.industry
General Engineering
Distributed object
Pattern recognition
semantic segmentation
TK1-9971
Transformation (function)
Data efficiency
convolutional neural network
Task analysis
Lesions
020201 artificial intelligence & image processing
Artificial intelligence
Electrical engineering. Electronics. Nuclear engineering
business
Neural networks
Zdroj: IEEE Access, Vol 9, Pp 133365-133375 (2021)
IEEE ACCESS
ISSN: 2169-3536
Popis: A key step in medical image-based diagnosis is image segmentation. A common use case for medical image segmentation is the identification of single structures of an elliptical shape. Most organs like the heart and kidneys fall into this category, as well as skin lesions, polyps, and other types of abnormalities. Neural networks have dramatically improved medical image segmentation results, but still require large amounts of training data and long training times to converge. In this paper, we propose a general way to improve neural network segmentation performance and data efficiency on medical imaging segmentation tasks where the goal is to segment a single roughly elliptically distributed object. We propose training a neural network on polar transformations of the original dataset, such that the polar origin for the transformation is the center point of the object. This results in a reduction of dimensionality as well as a separation of segmentation and localization tasks, allowing the network to more easily converge. Additionally, we propose two different approaches to obtaining an optimal polar origin: (1) estimation via a segmentation trained on non-polar images and (2) estimation via a model trained to predict the optimal origin. We evaluate our method on the tasks of liver, polyp, skin lesion, and epicardial adipose tissue segmentation. We show that our method produces state-of-the-art results for lesion, liver, and polyp segmentation and performs better than most common neural network architectures for biomedical image segmentation. Additionally, when used as a pre-processing step, our method generally improves data efficiency across datasets and neural network architectures.
Databáze: OpenAIRE
Externí odkaz: