Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification

<b>Background:</b> This study aimed to evaluate the effect of sample size on the development of a three-dimensional convolutional neural network (3DCNN) model for predicting the binary classification of three types of intracranial hemorrhage (ICH): intraparenchymal, subarachnoid, and sub...

Full description

Saved in:
Bibliographic Details
Main Authors: Chun-Chao Huang, Hsin-Fan Chiang, Cheng-Chih Hsieh, Bo-Rui Zhu, Wen-Jie Wu, Jin-Siang Shaw
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/2/216
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832588715491328000
author Chun-Chao Huang
Hsin-Fan Chiang
Cheng-Chih Hsieh
Bo-Rui Zhu
Wen-Jie Wu
Jin-Siang Shaw
author_facet Chun-Chao Huang
Hsin-Fan Chiang
Cheng-Chih Hsieh
Bo-Rui Zhu
Wen-Jie Wu
Jin-Siang Shaw
author_sort Chun-Chao Huang
collection DOAJ
description <b>Background:</b> This study aimed to evaluate the effect of sample size on the development of a three-dimensional convolutional neural network (3DCNN) model for predicting the binary classification of three types of intracranial hemorrhage (ICH): intraparenchymal, subarachnoid, and subdural (IPH, SAH, SDH, respectively). <b>Methods:</b> During the training, we compiled all images of each brain computed tomography scan into a single 3D image, which was then fed into the model to classify the presence of ICH. We divided the non-hemorrhage quantities into 20, 30, 40, 50, 100, and 150 and the ICH quantities into 20, 30, 40, and 50. Cross-validation was performed to compute the average area under the curve (AUC) over the last five iterations. The AUC and accuracy were used to evaluate the performance of the models. <b>Results:</b> Fifty patients, each with the three ICH types, and 150 non-hemorrhage cases were enrolled. Larger sample sizes achieved stable and acceptable performance in the artificial intelligence (AI) models, whereas training with a limited number of cases posed the risk of falsely high AUC values or accuracy. The overall trends and fluctuations in AUC values were similar between IPH and SDH but different for SAH. The accuracy of the results was relatively consistent among the three ICH types. <b>Conclusions:</b> The 3DCNN technique can be used to develop AI models capable of detecting ICH from limited case numbers. However, a minimal case number must be provided. The performance of AI models varies across different ICH types and is more stable with larger sample sizes.
format Article
id doaj-art-9f7ab19a429e4ef0bbb66e46ac5d1fbb
institution Kabale University
issn 2075-4418
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-9f7ab19a429e4ef0bbb66e46ac5d1fbb2025-01-24T13:29:08ZengMDPI AGDiagnostics2075-44182025-01-0115221610.3390/diagnostics15020216Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage ClassificationChun-Chao Huang0Hsin-Fan Chiang1Cheng-Chih Hsieh2Bo-Rui Zhu3Wen-Jie Wu4Jin-Siang Shaw5Department of Radiology, MacKay Memorial Hospital, Taipei 104, TaiwanDepartment of Radiology, MacKay Memorial Hospital, Taipei 104, TaiwanDepartment of Radiology, MacKay Memorial Hospital, Taipei 104, TaiwanInstitute of Mechatronic Engineering, National Taipei University of Technology, Taipei 106, TaiwanInstitute of Mechatronic Engineering, National Taipei University of Technology, Taipei 106, TaiwanInstitute of Mechatronic Engineering, National Taipei University of Technology, Taipei 106, Taiwan<b>Background:</b> This study aimed to evaluate the effect of sample size on the development of a three-dimensional convolutional neural network (3DCNN) model for predicting the binary classification of three types of intracranial hemorrhage (ICH): intraparenchymal, subarachnoid, and subdural (IPH, SAH, SDH, respectively). <b>Methods:</b> During the training, we compiled all images of each brain computed tomography scan into a single 3D image, which was then fed into the model to classify the presence of ICH. We divided the non-hemorrhage quantities into 20, 30, 40, 50, 100, and 150 and the ICH quantities into 20, 30, 40, and 50. Cross-validation was performed to compute the average area under the curve (AUC) over the last five iterations. The AUC and accuracy were used to evaluate the performance of the models. <b>Results:</b> Fifty patients, each with the three ICH types, and 150 non-hemorrhage cases were enrolled. Larger sample sizes achieved stable and acceptable performance in the artificial intelligence (AI) models, whereas training with a limited number of cases posed the risk of falsely high AUC values or accuracy. The overall trends and fluctuations in AUC values were similar between IPH and SDH but different for SAH. The accuracy of the results was relatively consistent among the three ICH types. <b>Conclusions:</b> The 3DCNN technique can be used to develop AI models capable of detecting ICH from limited case numbers. However, a minimal case number must be provided. The performance of AI models varies across different ICH types and is more stable with larger sample sizes.https://www.mdpi.com/2075-4418/15/2/216intracranial hemorrhagedeep learning3D convolutional neural networksartificial intelligenceCT
spellingShingle Chun-Chao Huang
Hsin-Fan Chiang
Cheng-Chih Hsieh
Bo-Rui Zhu
Wen-Jie Wu
Jin-Siang Shaw
Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification
Diagnostics
intracranial hemorrhage
deep learning
3D convolutional neural networks
artificial intelligence
CT
title Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification
title_full Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification
title_fullStr Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification
title_full_unstemmed Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification
title_short Impact of Dataset Size on 3D CNN Performance in Intracranial Hemorrhage Classification
title_sort impact of dataset size on 3d cnn performance in intracranial hemorrhage classification
topic intracranial hemorrhage
deep learning
3D convolutional neural networks
artificial intelligence
CT
url https://www.mdpi.com/2075-4418/15/2/216
work_keys_str_mv AT chunchaohuang impactofdatasetsizeon3dcnnperformanceinintracranialhemorrhageclassification
AT hsinfanchiang impactofdatasetsizeon3dcnnperformanceinintracranialhemorrhageclassification
AT chengchihhsieh impactofdatasetsizeon3dcnnperformanceinintracranialhemorrhageclassification
AT boruizhu impactofdatasetsizeon3dcnnperformanceinintracranialhemorrhageclassification
AT wenjiewu impactofdatasetsizeon3dcnnperformanceinintracranialhemorrhageclassification
AT jinsiangshaw impactofdatasetsizeon3dcnnperformanceinintracranialhemorrhageclassification