no code implementations • 17 Oct 2022 • Stefano B. Blumberg, Daniele Raví, Mou-Cheng Xu, Matteo Figini, Iasonas Kokkinos, Daniel C. Alexander
Transposed convolution is crucial for generating high-resolution outputs, yet has received little attention compared to convolution layers.
1 code implementation • 30 Aug 2022 • Ashkan Pakzad, Mou-Cheng Xu, Wing Keung Cheung, Marie Vermant, Tinne Goos, Laurens J De Sadeleer, Stijn E Verleden, Wim A Wuyts, John R Hurst, Joseph Jacob
We compare our ATN model with a state-of-the-art GAN-based network (simGAN) using a) qualitative assessment; b) assessment of the ability of ATN and simGAN based CT airway metrics to predict mortality in a population of 113 patients with IPF.
1 code implementation • 8 Aug 2022 • Mou-Cheng Xu, Yukun Zhou, Chen Jin, Marius de Groot, Daniel C. Alexander, Neil P. Oxtoby, Yipeng Hu, Joseph Jacob
Secondly, we propose a semi-supervised medical image segmentation method purely based on the original pseudo labelling, namely SegPL.
1 code implementation • 19 Mar 2022 • Mou-Cheng Xu, Yu-Kun Zhou, Chen Jin, Stefano B Blumberg, Frederick J Wilson, Marius deGroot, Daniel C. Alexander, Neil P. Oxtoby, Joseph Jacob
We propose MisMatch, a novel consistency-driven semi-supervised segmentation framework which produces predictions that are invariant to learnt feature perturbations.
2 code implementations • 23 Oct 2021 • Mou-Cheng Xu, Yukun Zhou, Chen Jin, Marius de Groot, Neil P. Oxtoby, Daniel C. Alexander, Joseph Jacob
The state-of-the-art SSL methods in image classification utilise consistency regularisation to learn unlabelled predictions which are invariant to input level perturbations.
no code implementations • 24 Oct 2020 • Mou-Cheng Xu
Brain shift makes the pre-operative MRI navigation highly inaccurate hence the intraoperative modalities are adopted in surgical theatre.
3 code implementations • 31 Jul 2020 • Le Zhang, Ryutaro Tanno, Mou-Cheng Xu, Chen Jin, Joseph Jacob, Olga Ciccarelli, Frederik Barkhof, Daniel C. Alexander
Recent years have seen increasing use of supervised learning methods for segmentation tasks.
1 code implementation • 29 Jul 2020 • Chen Jin, Ryutaro Tanno, Mou-Cheng Xu, Thomy Mertzanidou, Daniel C. Alexander
We demonstrate on three publicly available high-resolution image datasets that the foveation module consistently improves segmentation performance over the cases trained with patches of fixed FoV/resolution trade-off.
1 code implementation • 29 Jul 2020 • Mou-Cheng Xu, Neil P. Oxtoby, Daniel C. Alexander, Joseph Jacob
We compared our methods with state-of-the-art attention mechanisms in medical imaging, including self-attention, spatial-attention and spatial-channel mixed attention.