Evaluation of Preprocessing Techniques for U-Net Based Automated Liver Segmentation (original) (raw)
Related papers
A U-Net Based Multi-scale Feature Extraction for Liver Tumour Segmentation in CT Images
Lecture notes in electrical engineering, 2022
A new method for automatic liver tumour segmentation from computed tomography (CT) scans based on deep neural network is presented. Two cascaded deep convolutional neural networks are used to segment the CT image of the abdominal cavity. The first U-net is used for coarse segmentation to obtain the approximate position of the liver and tumour. Using this as a prediction the original image is cropped to reduce its size in order to increase the segmentation accuracy. The second modified U-net is employed for accurate segmentation of the actual liver tumours. Residual modules and dense connections are added to U-net to help the network train faster while producing more accurate results. In addition, multi-dimensional information fusion is introduced to make the network more comprehensive in restoring information. The Liver Tumour Segmentation (LiTs) dataset is used to evaluate the relative segmentation performance obtaining an average dice score of 0.665 based our method.
A Multiple Layer U-Net, Un-Net, for Liver and Liver Tumor Segmentation in CT
IEEE Access
Medical image segmentation is one of the crucial tasks in diagnosis as well as pre-surgery. Recently, deep learning has significantly contributed to improving the efficiency of medical image extraction. The U-Net network has been a favored network model, which has been used as a platform architecture, for medical image segmentation. For the success of these studies, most of these models were primarily focused on the changing of the interconnection between the nodes in the network, and changing the structure of the convolution units. This would result in the ignorance of the output features of convolution units in the nodes. In this study, a U n-Net, an n-fold network architecture, was proposed based on the traditional U-Net. In the U n-Net model, the output features of the convolution units are taken as the skip connection. Therefore, the U n-Net network exploits the output features of the convolution units in the nodes. In this study, we investigated a U 2-Net and a U 3-Net for segmentation of the liver and liver tumors. Besides, dilated convolution (DC) and dense structure were also used in the nodes of our networks. The efficiency of our models was evaluated on two public datasets: LiTS and 3DIRCADb. The Dice's Similarity Coefficient (DSC) of our proposed models achieved 96.38% for liver segmentation and 73.69% for tumor segmentation on the LiTS dataset. For the 3DIRCADb dataset, the results achieved 96.45% in DSC for the liver segmentation and 73.34% for the tumor segmentation. The experimental results show that our proposed networks achieved better results than the recent networks. And it is convinced that our network would be useful for practical deployments.
2020
An automated and generally applicable method for segmentation is still in focus of medical image processing research. Since a few years artificial inteligence methods show promising results, especially with widely available scalable Deep Learning libraries. In this work, a five layer hybrid U-net is developed for slice-byslice segmentation of liver data sets. Training data is taken from the Medical Segmentation Decathlon database, providing 131 fully segmented volumes. A slice-oriented segmentation model is implemented utilizing deep learning algorithms with adaptions for variable parenchyma shape along the stacking direction and similarities between adjacent slices. Both are transformed for coronal and sagittal views. The implementation is on a GPU rack with TensorFlow and Keras. For a quantitative measure of segmentation accuracy, standardized volume and surface metrics are used. Results DSC=97.59, JI=95.29 and NSD=99.37 show proper segmentation comparable to 3D U-Nets and other state of the art. The development of a 2D-slice oriented segmentation is justified by short training time and less complexity and therefore massively reduced memory consumption. This work manifests the high potential of AI methods for general use in medical segmentation as fully-or semi-automated tool supervised by the expert user.
Segmentation of Liver Anatomy by Combining 3D U-Net Approaches
Applied Sciences, 2021
Accurate liver vessel segmentation is of crucial importance for the clinical diagnosis and treatment of many hepatic diseases. Recent state-of-the-art methods for liver vessel reconstruction mostly utilize deep learning methods, namely, the U-Net model and its variants. However, to the best of our knowledge, no comparative evaluation has been proposed to compare these approaches in the liver vessel segmentation task. Moreover, most research works do not consider the liver volume segmentation as a preprocessing step, in order to keep only inner hepatic vessels, for Couinaud representation for instance. For these reasons, in this work, we propose using accurate Dense U-Net liver segmentation and conducting a comparison between 3D U-Net models inside the obtained volumes. More precisely, 3D U-Net, Dense U-Net, and MultiRes U-Net are pitted against each other in the vessel segmentation task on the IRCAD dataset. For each model, three alternative setups that allow adapting the selected C...
Hepatic Tumor Detection using U-Net++ from Lungs CT Images
International Journal of Scientific and Research Publications, 2021
Cancer is today's one of the most harmful and deadly diseases. Early detection of cancer is essential as it helps radiologists and doctors perform treatment planning and surgery. Detection of tumors in the early stages is a difficult task. Artificial intelligencebased approaches which use deep learning and machine learning techniques assist radiologists in the prediction. The main aim of this research is to develop a fully automated tool for Liver Tumor Segmentation. Also, the purpose of this research is to examine the problem of lungs cancer detection from CT lungs images and propose an efficient approach that will segment out complex regions and improves the performance with lesser error rates. These techniques not only detect tumor but also provides other information (Survival rate, tumor statistics) which are very helpful.
A Deep Learning Approach for Liver and Tumor Segmentation in CT Images Using ResUNet
Bioengineering
According to the most recent estimates from global cancer statistics for 2020, liver cancer is the ninth most common cancer in women. Segmenting the liver is difficult, and segmenting the tumor from the liver adds some difficulty. After a sample of liver tissue is taken, imaging tests, such as magnetic resonance imaging (MRI), computer tomography (CT), and ultrasound (US), are used to segment the liver and liver tumor. Due to overlapping intensity and variability in the position and shape of soft tissues, segmentation of the liver and tumor from computed abdominal tomography images based on shade gray or shapes is undesirable. This study proposed a more efficient method for segmenting liver and tumors from CT image volumes using a hybrid ResUNet model, combining the ResNet and UNet models to address this gap. The two overlapping models were primarily used in this study to segment the liver and for region of interest (ROI) assessment. Segmentation of the liver is done to examine the ...
Biomedical Image Segmentation with Modified U-Net
Traitement du Signal
Image segmentation is an important field in image processing and computer vision, particularly in the development of methods to assist experts in the biomedical and medical fields. It plays a vital role in saving time and costs. One of the most successful and significant methods in image segmentation using deep learning is the U-Net model. In this paper, we propose U-Net11, a novel variant of U-Net that uses 11 convolutional layers and introduces some modifications to improve the segmentation performance. The classical U-Net model was developed and tested on three different datasets, outperforming the traditional U-Net approach. The U-Net11 model was evaluated for breast cancer segmentation, lung segmentation from CT images, and the nuclei segmentation dataset from the Data Science Bowl 2018 competition. These datasets are valuable due to their varying image quantities and the varying difficulty levels in segmentation tasks. The modified U-Net model has achieved Dice Similarity Coefficient scores of 69.09% on the breast cancer dataset, 95.02% on the lung segmentation dataset and 81.10% on the nuclei segmentation dataset, exceeding the performance of the classical U-Net model by 5%, 2% and 4% respectively. This difference in success rates is particularly significant for critical segmentation datasets. deep learning,
Effects of Multiple Filters on Liver Tumor Segmentation From CT Images
Frontiers in Oncology, 2021
Segmentation of liver tumors from Computerized Tomography (CT) images remains a challenge due to the natural variation in tumor shape and structure as well as the noise in CT images. A key assumption is that the performance of liver tumor segmentation depends on the characteristics of multiple features extracted from multiple filters. In this paper, we design an enhanced approach based on a two-class (liver, tumor) convolutional neural network that discriminates tumor as well as liver from CT images. First, the contrast and intensity values in CT images are adjusted and high frequencies are removed using Hounsfield units (HU) filtering and standardization. Then, the liver tumor is segmented from entire images with multiple filter U-net (MFU-net). Finally, a quantitative analysis is carried out to evaluate the segmentation results using three different methods: boundary-distance-based metrics, size-based metrics, and overlap-based metrics. The proposed method is validated on CT image...