Published on Mon Aug 30 2021

Robust Interactive Semantic Segmentation of Pathology Images with Minimal User Input

Mostafa Jahanifar, Neda Zamani Tajeddin, Navid Alemi Koohbanani, Nasir Rajpoot

Automatic deep learning models for semantic segmentation require lots of annotated training data. We propose an efficient interactive segmentation network that requires minimum input from the user to accurately annotate different tissue types in the histology image.

1
0
0
Abstract

From the simple measurement of tissue attributes in pathology workflow to designing an explainable diagnostic/prognostic AI tool, access to accurate semantic segmentation of tissue regions in histology images is a prerequisite. However, delineating different tissue regions manually is a laborious, time-consuming and costly task that requires expert knowledge. On the other hand, the state-of-the-art automatic deep learning models for semantic segmentation require lots of annotated training data and there are only a limited number of tissue region annotated images publicly available. To obviate this issue in computational pathology projects and collect large-scale region annotations efficiently, we propose an efficient interactive segmentation network that requires minimum input from the user to accurately annotate different tissue types in the histology image. The user is only required to draw a simple squiggle inside each region of interest so it will be used as the guiding signal for the model. To deal with the complex appearance and amorph geometry of different tissue regions we introduce several automatic and minimalistic guiding signal generation techniques that help the model to become robust against the variation in the user input. By experimenting on a dataset of breast cancer images, we show that not only does our proposed method speed up the interactive annotation process, it can also outperform the existing automatic and interactive region segmentation models.

Thu Aug 26 2021
Computer Vision
Evaluating Transformer based Semantic Segmentation Networks for Pathological Image Segmentation
Transformer neural networks have shown the unique merit of capturing the global long distance dependencies across the entire image. Segmenter, Swin-Transformer and TransUNet came out as the best performers among the twelve evaluated models.
2
0
1
Mon Jun 22 2020
Machine Learning
HookNet: multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images
0
0
0
Sun Apr 25 2021
Computer Vision
MIDeepSeg: Minimally Interactive Segmentation of Unseen Objects from Medical Images Using Deep Learning
0
0
0
Wed Dec 09 2020
Machine Learning
AIDE: Annotation-efficient deep learning for automatic medical image segmentation
Annotation-effIcient Deep lEarning (AIDE) can handle imperfect datasets with an elaborately designed cross-model self-correcting mechanism. For three clinical datasets containing 11,852 breast images, AIDE consistently produces segmentation maps comparable to those generated by the fully
0
0
0
Fri Aug 06 2021
Computer Vision
Efficient and Generic Interactive Segmentation Framework to Correct Mispredictions during Clinical Evaluation of Medical Images
Researchers propose conditional inference technique for DNNs. Method takes the intervention by a medical expert as test time constraints and performs conditional inference conditioned upon these constraints. Method can be useful to clinicians for post-surgical follow-up with minimal intervention.
0
0
0
Wed Oct 11 2017
Computer Vision
Interactive Medical Image Segmentation using Deep Learning with Image-specific Fine-tuning
Convolutional neural networks (CNNs) have achieved state-of-the-art performance for automatic medical image segmentation. However, they have not demonstrated sufficiently accurate and robust results for clinical use. We propose a novel deep learning-based framework for interactive segmentation by incorporating CNNs into a bounding box.
0
0
0
Wed Feb 11 2015
Machine Learning
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Batch Normalization allows us to use much higher learning rates and be less careful about initialization. It also acts as a regularizer, in some cases eliminating the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalized achieves the same
2
342
1,163
Fri Jun 25 2021
Computer Vision
Semantic annotation for computational pathology: Multidisciplinary experience and best practice recommendations
Computational Pathology (CPath) offers an integrated solution to utilize information embedded in pathology WSIs. Despite the crucial role of annotations in CPath projects, there are no well-defined guidelines or best practices on how they should be carried out.
2
9
21
Mon Oct 16 2017
Neural Networks
Searching for Activation Functions
The choice of activation functions in deep networks has a significant effect on the training dynamics and task performance. Currently, the most successful and widely-used activation function is the Rectified Linear Unit (ReLU) We propose using automatic search techniques to discover new activation functions.
1
2
7
Tue May 28 2019
Machine Learning
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. We propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient.
4
0
3
Sat Jan 13 2018
Computer Vision
MobileNetV2: Inverted Residuals and Linear Bottlenecks
MobileNetV2 is based on an inverted residual structure where input and output of the residual block are thin bottleneck layers opposite to traditional residual models. The architecture uses lightweight depthwise convolutions to filter features in the intermediate expansion layer.
1
3
3
Wed Aug 12 2020
Computer Vision
Self-Path: Self-supervision for Classification of Pathology Images with Limited Annotations
High-resolution pathology images lend themselves well to `data hungry' deep learning algorithms. We propose a self-supervised CNN approach to leverage unlabeled data for learning generalizable and domain invariantrepresentations.
0
0
0