Rule-based deep learning method for prognosis of neonatal hypoxic-ischemic encephalopathy by using susceptibility weighted image analysis
Rule-based deep learning method for prognosis of neonatal hypoxic-ischemic encephalopathy by using susceptibility weighted image analysis
Objective: susceptibility weighted imaging (SWI) of neonatal hypoxic-ischemic brain injury can provide assistance in the prognosis of neonatal hypoxic-ischemic encephalopathy (HIE). We propose a convolutional neural network model to classify SWI images with HIE.
Materials and methods: due to the lack of a large dataset, transfer learning method with fine-tuning a pre-trained ResNet 50 is introduced. We randomly select 11 datasets from patients with normal neurology outcomes (n = 31) and patients with abnormal neurology outcomes (n = 11) at 24 months of age to avoid bias in classification due to any imbalance in the data.
Results: we develop a rule-based system to improve the classification performance, with an accuracy of 0.93 ± 0.09. We also compute heatmaps produced by the Grad-CAM technique to analyze which areas of SWI images contributed more to the classification patients with abnormal neurology outcome.
Conclusion: such regions that are important in the classification accuracy can interpret the relationship between the brain regions affected by hypoxic-ischemic and neurodevelopmental outcomes of infants with HIE at the age of 2 years.
Heatmap, Hypoxic-ischemic, SWI image, Transfer learning
227-239
Tang, Zhen
583e802a-a29c-4af9-bdbf-a03af73c048d
Mahmoodi, Sasan
91ca8da4-95dc-4c1e-ac0e-f2c08d6ac7cf
Meng, Di
f598c27c-b4d0-402a-8096-9a9ca8791d9b
Darekar, Angela
e66f3c01-0825-4da9-af68-e645f4c57f62
Vollmer, Brigitte
044f8b55-ba36-4fb2-8e7e-756ab77653ba
22 January 2024
Tang, Zhen
583e802a-a29c-4af9-bdbf-a03af73c048d
Mahmoodi, Sasan
91ca8da4-95dc-4c1e-ac0e-f2c08d6ac7cf
Meng, Di
f598c27c-b4d0-402a-8096-9a9ca8791d9b
Darekar, Angela
e66f3c01-0825-4da9-af68-e645f4c57f62
Vollmer, Brigitte
044f8b55-ba36-4fb2-8e7e-756ab77653ba
Tang, Zhen, Mahmoodi, Sasan and Meng, Di
,
et al.
(2024)
Rule-based deep learning method for prognosis of neonatal hypoxic-ischemic encephalopathy by using susceptibility weighted image analysis.
Magnetic Resonance Materials in Physics, Biology and Medicine, 37 (2), .
(doi:10.1007/s10334-023-01139-2).
Abstract
Objective: susceptibility weighted imaging (SWI) of neonatal hypoxic-ischemic brain injury can provide assistance in the prognosis of neonatal hypoxic-ischemic encephalopathy (HIE). We propose a convolutional neural network model to classify SWI images with HIE.
Materials and methods: due to the lack of a large dataset, transfer learning method with fine-tuning a pre-trained ResNet 50 is introduced. We randomly select 11 datasets from patients with normal neurology outcomes (n = 31) and patients with abnormal neurology outcomes (n = 11) at 24 months of age to avoid bias in classification due to any imbalance in the data.
Results: we develop a rule-based system to improve the classification performance, with an accuracy of 0.93 ± 0.09. We also compute heatmaps produced by the Grad-CAM technique to analyze which areas of SWI images contributed more to the classification patients with abnormal neurology outcome.
Conclusion: such regions that are important in the classification accuracy can interpret the relationship between the brain regions affected by hypoxic-ischemic and neurodevelopmental outcomes of infants with HIE at the age of 2 years.
This record has no associated files available for download.
More information
Accepted/In Press date: 11 December 2023
e-pub ahead of print date: 22 January 2024
Published date: 22 January 2024
Keywords:
Heatmap, Hypoxic-ischemic, SWI image, Transfer learning
Identifiers
Local EPrints ID: 489437
URI: http://eprints.soton.ac.uk/id/eprint/489437
ISSN: 1352-8661
PURE UUID: 505160fd-43d8-4845-9081-aad5160bb6cc
Catalogue record
Date deposited: 24 Apr 2024 16:34
Last modified: 25 Apr 2024 01:42
Export record
Altmetrics
Contributors
Author:
Zhen Tang
Author:
Sasan Mahmoodi
Author:
Di Meng
Author:
Angela Darekar
Corporate Author: et al.
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics