A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging
Medical imaging is crucial for the detection and diagnosis of breast cancer. Artificial intelligence and computer vision have rapidly become popular in medical image analyses thanks to technological advancements. To improve the effectiveness and efficiency of medical diagnosis and treatment, signifi...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/4/616 |
| _version_ | 1797297548301434880 |
|---|---|
| author | Lucas Jian Hoong Leow Abu Bakr Azam Hong Qi Tan Wen Long Nei Qi Cao Lihui Huang Yuan Xie Yiyu Cai |
| author_facet | Lucas Jian Hoong Leow Abu Bakr Azam Hong Qi Tan Wen Long Nei Qi Cao Lihui Huang Yuan Xie Yiyu Cai |
| author_sort | Lucas Jian Hoong Leow |
| collection | DOAJ |
| description | Medical imaging is crucial for the detection and diagnosis of breast cancer. Artificial intelligence and computer vision have rapidly become popular in medical image analyses thanks to technological advancements. To improve the effectiveness and efficiency of medical diagnosis and treatment, significant efforts have been made in the literature on medical image processing, segmentation, volumetric analysis, and prediction. This paper is interested in the development of a prediction pipeline for breast cancer studies based on 3D computed tomography (CT) scans. Several algorithms were designed and integrated to classify the suitability of the CT slices. The selected slices from patients were then further processed in the pipeline. This was followed by data generalization and volume segmentation to reduce the computation complexity. The selected input data were fed into a 3D U-Net architecture in the pipeline for analysis and volumetric predictions of cancer tumors. Three types of U-Net models were designed and compared. The experimental results show that Model 1 of U-Net obtained the highest accuracy at 91.44% with the highest memory usage; Model 2 had the lowest memory usage with the lowest accuracy at 85.18%; and Model 3 achieved a balanced performance in accuracy and memory usage, which is a more suitable configuration for the developed pipeline. |
| first_indexed | 2024-03-07T22:22:54Z |
| format | Article |
| id | doaj.art-845e7a439cc743d59f0d91a6f5d3c799 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-07T22:22:54Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-845e7a439cc743d59f0d91a6f5d3c7992024-02-23T15:26:18ZengMDPI AGMathematics2227-73902024-02-0112461610.3390/math12040616A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer ImagingLucas Jian Hoong Leow0Abu Bakr Azam1Hong Qi Tan2Wen Long Nei3Qi Cao4Lihui Huang5Yuan Xie6Yiyu Cai7School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, SingaporeSchool of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, SingaporeNational Cancer Center, Singapore 168583, SingaporeNational Cancer Center, Singapore 168583, SingaporeSchool of Computing Science, University of Glasgow, Glasgow G12 8RZ, UKSchool of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, SingaporeSchool of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, SingaporeSchool of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, SingaporeMedical imaging is crucial for the detection and diagnosis of breast cancer. Artificial intelligence and computer vision have rapidly become popular in medical image analyses thanks to technological advancements. To improve the effectiveness and efficiency of medical diagnosis and treatment, significant efforts have been made in the literature on medical image processing, segmentation, volumetric analysis, and prediction. This paper is interested in the development of a prediction pipeline for breast cancer studies based on 3D computed tomography (CT) scans. Several algorithms were designed and integrated to classify the suitability of the CT slices. The selected slices from patients were then further processed in the pipeline. This was followed by data generalization and volume segmentation to reduce the computation complexity. The selected input data were fed into a 3D U-Net architecture in the pipeline for analysis and volumetric predictions of cancer tumors. Three types of U-Net models were designed and compared. The experimental results show that Model 1 of U-Net obtained the highest accuracy at 91.44% with the highest memory usage; Model 2 had the lowest memory usage with the lowest accuracy at 85.18%; and Model 3 achieved a balanced performance in accuracy and memory usage, which is a more suitable configuration for the developed pipeline.https://www.mdpi.com/2227-7390/12/4/616convolutional neural network3D computed tomography scanbreast cancer3D U-Net architecture3D volumetric prediction pipeline |
| spellingShingle | Lucas Jian Hoong Leow Abu Bakr Azam Hong Qi Tan Wen Long Nei Qi Cao Lihui Huang Yuan Xie Yiyu Cai A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging Mathematics convolutional neural network 3D computed tomography scan breast cancer 3D U-Net architecture 3D volumetric prediction pipeline |
| title | A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging |
| title_full | A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging |
| title_fullStr | A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging |
| title_full_unstemmed | A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging |
| title_short | A Convolutional Neural Network-Based Auto-Segmentation Pipeline for Breast Cancer Imaging |
| title_sort | convolutional neural network based auto segmentation pipeline for breast cancer imaging |
| topic | convolutional neural network 3D computed tomography scan breast cancer 3D U-Net architecture 3D volumetric prediction pipeline |
| url | https://www.mdpi.com/2227-7390/12/4/616 |
| work_keys_str_mv | AT lucasjianhoongleow aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT abubakrazam aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT hongqitan aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT wenlongnei aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT qicao aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT lihuihuang aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT yuanxie aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT yiyucai aconvolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT lucasjianhoongleow convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT abubakrazam convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT hongqitan convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT wenlongnei convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT qicao convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT lihuihuang convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT yuanxie convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging AT yiyucai convolutionalneuralnetworkbasedautosegmentationpipelineforbreastcancerimaging |