1 code implementation • 31 Jan 2023 • Ange Lou, Shuyue Guan, Murray Loew
This paper proposes a Context Axial Reverse Attention Network (CaraNet) to improve the segmentation performance on small objects compared with several recent state-of-the-art models.
no code implementations • 8 Jan 2022 • Shuyue Guan, Murray Loew
Instead of using current deep-learning segmentation models (like the UNet and variants), we approach the segmentation problem using trained Convolutional Neural Network (CNN) classifiers, which automatically extract important features from images for classification.
no code implementations • 13 Oct 2021 • Han Chen, Yifan Jiang, Hanseok Ko, Murray Loew
Automatic segmentation of infected regions in computed tomography (CT) images is necessary for the initial diagnosis of COVID-19.
no code implementations • 11 Sep 2021 • Shuyue Guan, Murray Loew
To quantitatively measure the separability of datasets, we create an intrinsic measure -- the Distance-based Separability Index (DSI), which is independent of the classifier model.
1 code implementation • 16 Aug 2021 • Ange Lou, Shuyue Guan, Hanseok Ko, Murray Loew
Segmenting medical images accurately and reliably is important for disease diagnosis and treatment.
Ranked #9 on Medical Image Segmentation on ETIS-LARIBPOLYPDB
1 code implementation • 17 Jun 2021 • Shuyue Guan, Murray Loew
Without true labels, to design an effective CVI is as difficult as to create a clustering method.
1 code implementation • 10 May 2021 • Ange Lou, Shuyue Guan, Murray Loew
By comparison, CFPNet-M achieves comparable segmentation results on all five medical datasets with only 0. 65 million parameters, which is about 2% of U-Net, and 8. 8 MB memory.
2 code implementations • 22 Mar 2021 • Ange Lou, Murray Loew
Based on the CFP module, we built CFPNet for real-time semantic segmentation which applied a series of dilated convolution channels to extract effective features.
no code implementations • 5 Jan 2021 • Shuyue Guan, Murray Loew
We proposed three ML-learnable characteristics to verify learnable problems for ML models, such as DNNs, and explain why DNNs work for specific counting problems but cannot generally count connected components.
no code implementations • 23 Nov 2020 • Han Chen, Yifan Jiang, Murray Loew, Hanseok Ko
In this paper, we propose an unsupervised domain adaptation based segmentation network to improve the segmentation performance of the infection areas in COVID-19 CT images.
no code implementations • 31 Oct 2020 • Ange Lou, Shuyue Guan, Nada Kamona, Murray Loew
It was used to segment the breast area by using a set of breast IR images, collected in our pilot study by imaging breast cancer patients and normal volunteers with a thermal infrared camera (N2 Imager).
no code implementations • 26 Oct 2020 • Shuyue Guan, Murray Loew
In this study, we propose a novel theory based on space partitioning to estimate the approximate training accuracy for two-layer neural networks on random datasets without training.
1 code implementation • 16 Sep 2020 • Shuyue Guan, Murray Loew
We create the decision boundary complexity (DBC) score to define and measure the complexity of decision boundary of DNNs.
1 code implementation • 2 Sep 2020 • Shuyue Guan, Murray Loew
And, to have more CVIs is crucial because there is no universal CVI that can be used to measure all datasets, and no specific method for selecting a proper CVI for clusters without true labels.
no code implementations • 29 Jul 2020 • Yifan Jiang, Han Chen, Murray Loew, Hanseok Ko
However, training a deep-learning model requires large volumes of data, and medical staff faces a high risk when collecting COVID-19 CT data due to the high infectivity of the disease.
4 code implementations • 31 May 2020 • Ange Lou, Shuyue Guan, Murray Loew
The Convolution Neural Network (CNN) has brought a breakthrough in images segmentation areas, especially, for medical images.
1 code implementation • 27 May 2020 • Shuyue Guan, Murray Loew, Hanseok Ko
In machine learning, the performance of a classifier depends on both the classifier model and the dataset.
2 code implementations • 27 Feb 2020 • Shuyue Guan, Murray Loew
We characterize the performance of a GAN as an image generator according to three aspects: 1) Creativity: non-duplication of the real images.