Segmentation of Intracranial Hemorrhage Using Semi-Supervised Multi-Task Attention-Based U-Net

Abstract

Intracranial Hemorrhage (ICH) has high rates of mortality, and risk factors associated with it are sometimes nearly impossible to avoid. Previous techniques to detect ICH using machine learning have shown some promise. However, due to a limited number of labeled medical images available, which often causes poor model accuracy in terms of the Dice coefficient, there is much to be improved. In this paper, we propose a modified u-net and curriculum learning strategy using a multi-task semi-supervised attention-based model, initially introduced by Chen et al., to segment ICH sub-groups from CT images. Using a modified inverse-sigmoid-based curriculum learning training strategy, we were able to stabilize Chen’s algorithm experimentally. This semi-supervised model produced higher Dice coefficient values in comparison to a supervised counterpart, regardless of the amount of labeled data used to train the model. Specifically, when training with 80% of the ground truth data, our semi-supervised model produced a Dice coefficient of 0.67, which was higher than 0.61, obtained by a comparable supervised model. This result also surpassed by a greater margin the one obtained by using the out-of-the-box u-net by Hssayeni et al.

A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT

“…Although several methods have been described in the literature to segment intracranial hemorrhage, they all suffer from some major drawbacks such as failing to balance the trade-off between accuracy and time-efficiency, sensitivity to initialization states, and lack of evaluation on CT images from multiple institutes [9][10][11][12][13][14][15][16][17] . Even the most recently developed deep learning solutions have often not been able to realize the targeted level of performance 18,19 . As one of the favorable methods standing out in the literature, the PItcHPERFeCT solution achieved a Dice similarity coefficient of 91% 28 .…”

“…Given the clinical significance, numerous attempts at developing computer-assisted automation tools for hematoma volume have been reported [13][14][15][16] . However, many of the techniques even till recent are either manual, semi-automatic, slow, or suffer from poor accuracy, especially for irregular hematoma shapes and sizes [13][14][15][16][17][18][19] . Great efforts have been geared towards building automated segmentation tools with use of artificial intelligence technologies, in particular the promising deep-learning algorithms 20,21 .…”

Multi-Task Deep Learning Model with an Attention Mechanism for Ship Accident Sentence Prediction

“…Hard-parameter sharing is a structure that shares the hidden layer of the neural network model and has an output layer optimized for each task, as shown in Figure 3. This method is the most basic structure of multi-task learning, and it can prevent overfitting of a specific task because the model determines the appropriate output for each task [10]. Soft-parameter sharing uses a different neural network model for each task, and multi-task learning is performed by regularizing the difference between the parameters of each model.…”

“…By performing multiple tasks with one common feature, neural networks determine the appropriate output for each task. Therefore, it has the advantage of preventing the overfitting of one specific task [10]. The factual information of the verdict is used as an input for the shared layer and is delivered to three prediction tasks through the attention mechanism.…”

Automatic Hemorrhage Detection From Color Doppler Ultrasound Using a Generative Adversarial Network (GAN)-Based Anomaly Detection Method

“…Recent advances in machine and deep learning methods for detection and localization tasks have shown promising results in a variety of healthcare applications including hemorrhage detection. However, most of existing hemorrhage detection approaches are either supervised [14] , [15] , [16] , semi-supervised [17] or weakly supervised [18] . Also, most of the existing works are focused on brain intracranial hemorrhage (ICH) detection in head CT scans, and NCTH detection still remains largely unexplored.…”