Corpus ID: 236447740

A persistent homology-based topological loss for CNN-based multi-class segmentation of CMR

  title={A persistent homology-based topological loss for CNN-based multi-class segmentation of CMR},
  author={Nicholas Byrne and James R. Clough and Isra Valverde and G. Montana and Andrew P. King},
Multi-class segmentation of cardiac magnetic resonance (CMR) images seeks a separation of data into anatomical components with known structure and configuration. The most popular CNN-based methods are optimised using pixel wise loss functions, ignorant of the spatially extended features that characterise anatomy. Therefore, whilst sharing a high spatial overlap with the ground truth, inferred CNN-based segmentations can lack coherence, including spurious connected components, holes and voids… Expand


Explicit topological priors for deep-learning based image segmentation using persistent homology
We present a novel method to explicitly incorporate topological prior knowledge into deep learning based segmentation, which is, to our knowledge, the first work to do so. Our method uses the conceptExpand
Anatomically Constrained Neural Networks (ACNNs): Application to Cardiac Image Enhancement and Segmentation
This work proposes a generic training strategy that incorporates anatomical prior knowledge into CNNs through a new regularisation model, which is trained end-to-end and demonstrates how the learnt deep models of 3-D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies. Expand
Automatic 3D bi-ventricular segmentation of cardiac images by a shape-refined multi-task deep learning approach
A multi-task deep learning approach with atlas propagation is combined to develop a shape-refined bi-ventricular segmentation pipeline for short-axis CMR volumetric images that is robust and capable of producing accurate, high-resolution, and anatomically smooth bi- ventricular 3D models, despite the presence of artifacts in input CMR volumes. Expand
A topological encoding convolutional neural network for segmentation of 3D multiphoton images of brain vasculature using persistent homology
This work proposes a topological encoding convolutional neural network based on persistent homology to segment 3D multiphoton images of brain vasculature that outperforms state-of-the-art models in terms of the Dice coefficient and it is comparable in other metrics such as sensitivity. Expand
Topology Aware Fully Convolutional Networks for Histology Gland Segmentation
This work introduces the first deep network trained to encode geometric and topological priors of containment and detachment, and shows how the topology aware architecture outperforms competing methods by up to 10 % in both pixel-level accuracy and object-level Dice. Expand
Few-Shot Learning by a Cascaded Framework With Shape-Constrained Pseudo Label Assessment for Whole Heart Segmentation
A few-shot learning framework by combining ideas of semi-supervised learning and self-training for whole heart segmentation and achieving promising accuracy with a Dice score of 0.890 and a Hausdorff distance of 18.539 mm is proposed. Expand
Semi-supervised Learning for Segmentation Under Semantic Constraint
The proposed approach significantly reduces abnormalities produced during the segmentation of brain structures segmentation, and can be used in a semi-supervised way, opening a path to better generalization to unseen data. Expand
Topology-Preserving Deep Image Segmentation
A novel method that learns to segment with correct topology by designing a continuous-valued loss function that enforces a segmentation to have the same topology as the ground truth, i.e., having the same Betti number. Expand
Cardiac MRI Segmentation with Strong Anatomical Guarantees
With this method, CNNs can now produce results that are both within the inter-expert variability and always anatomically plausible, and can accommodate any cardiac segmentation method and convert its anatomical implausible results to plausible ones without affecting its overall geometric and clinical metrics. Expand
3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation
The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts and performs on-the-fly elastic deformations for efficient data augmentation during training. Expand