Published on Mon May 03 2021

Noisy Student learning for cross-institution brain hemorrhage detection

Emily Lin, Weicheng Kuo, Esther Yuh
0
0
0
Abstract

Computed tomography (CT) is the imaging modality used in the diagnosis of neurological emergencies, including acute stroke and traumatic brain injury. Advances in deep learning have led to models that can detect and segment hemorrhage on head CT. PatchFCN, one such supervised fully convolutional network (FCN), recently demonstrated expert-level detection of intracranial hemorrhage on in-sample data. However, its potential for similar accuracy outside the training domain is hindered by its need for pixel-labeled data from outside institutions. Also recently, a semi-supervised technique, Noisy Student (NS) learning, demonstrated state-of-the-art performance on ImageNet by moving from a fully-supervised to a semi-supervised learning paradigm. We combine the PatchFCN and Noisy Student approaches, extending semi-supervised learning to an intracranial hemorrhage segmentation task. Surprisingly, the NS model performance surpasses that of a fully-supervised oracle model trained with image-level labels on the same data. It also performs comparably to another recently reported supervised model trained on a labeled dataset 600x larger than that used to train the NS model. To our knowledge, we are the first to demonstrate the effectiveness of semi-supervised learning on a head CT detection and segmentation task.

Wed Nov 13 2019
Computer Vision
Extracting 2D weak labels from volume labels using multiple instance learning in CT hemorrhage detection
Multiple instance learning (MIL) is a supervised learning methodology that aims to allow models to learn instance class labels from bag class labels. MIL is gaining traction for learning from weak labels but has not been widely applied to 3D medical imaging.
0
0
0
Mon May 03 2021
Computer Vision
Weakly supervised deep learning-based intracranial hemorrhage localization
0
0
0
Sat Jun 29 2019
Computer Vision
Improved ICH classification using task-dependent learning
Intracranial hemorrhage (ICH) is among the most critical and timesensitive findings to be detected on Head CT. We present BloodNet, a deep learning architecture designed for optimal triaging of Head CTs.
0
0
0
Fri Jun 08 2018
Computer Vision
PatchFCN for Intracranial Hemorrhage Detection
This paper studies the problem of detecting and segmenting acute intracranial hemorrhage on head computed tomography (CT) scans. We propose to solve both tasks as a semantic segmentation problem using a patch-based fully convolutional network (PatchFCN)
0
0
0
Wed Mar 18 2020
Machine Learning
Train, Learn, Expand, Repeat
High-quality labeled data is essential to successfully train supervised machine learning models. Medical professionals who can expertly label the data are a scarce and expensive resource. We propose a recursive training strategy to perform semantic segmentation given only very few training samples with pixel-level annotations.
0
0
0
Sat Sep 08 2018
Machine Learning
Cost-Sensitive Active Learning for Intracranial Hemorrhage Detection
Deep learning for clinical applications is subject to stringent performance requirements. The enormous cost of labeling medical data makes this challenging. We show that our ensemble method compares favorably with the state-of-the-art.
0
0
0