An experimental study on the automatic segmentation of organs in computed tomography images based on transfer learning and data augmentation

Abstract

Abstract: Objective:To explore the role of transfer learning in the recognition, extraction, and automatic segmentation of anatomical structures in graphic software during machine deep learning, and the impact of data augmentation algorithms on the ability of transfer learning.
Method:The Efficient Net b1 neural network is used as the backbone network for deep learning on the platform of HyVision Ablation Planning V1.0 graphic software. 51 sets of VX2 rabbit liver cancer model computed tomography (CT) scan images are used for transfer learning training through data augmentation. The graphic software's existing functions of identifying, extracting, and automatically segmenting organs on human abdominal CT images are reproduced on animal models. The differences in Dice and normalized surface Dice (NSD) values under different learning models, as well as the differences between the image quality of 3D reconstruction and the doctor-marked animal model training set, are compared.
Result:From the model with "data augmentation without transfer learning" to the model with "data augmentation with transfer learning", the Dice index for automatic organ segmentation of VX2 rabbit CT scan images increased from 0.525 to 0.676, an increase of 28.76%, and the NSD index increased from 0.448 to 0.616, an increase of 37.50%. From the model with "no data augmentation but with transfer learning" to the model with "data augmentation with transfer learning", the Dice index for automatic organ segmentation of VX2 rabbit CT scan images increased from 0.502 to 0.676, an increase of 34.66%, and the NSD index increased from 0.459 to 0.616, an increase of 34.20%. This indicates that in the process of machine deep learning if studying new anatomical objects, transfer learning and data augmentation have equally important roles.
Conclusion:In the process of machine deep learning, the function of transfer learning can obtain better graphic recognition, extraction, and automatic segmentation results with the help of "data augmentation" algorithm.

Key words: Transfer learning, Data augmentation, Animal CT image segmentation, Neural network

Zhang Yumeng, Kang Wendi, Xi Junqing, Chi Chen, Yang Zhengqiang. An experimental study on the automatic segmentation of organs in computed tomography images based on transfer learning and data augmentation[J]. Electronic Journal of Liver Tumor, 2023, 10(3): 52-57.

References

[1] 王祎, 王峻峰, 杨超, 等. 数字医学技术在肝脏分段及解剖性肝切除中的应用现状[J/CD]. 中华肝脏外科手术学电子杂志, 2023, 12(1): 6.
[2] 何静, 陈胜男, 夏楠, 等. 计算机辅助手术系统在儿童肝脏原发性间叶肿瘤手术中的应用[J]. 肝胆胰外科杂志, 2022, 34(9): 513-518.
[3] RONNEBERGER O, FISCHER P, BROX T.U-Net: Convolutional Networks for Biomedical Image Segmentation[J]. Lecture Notes in Computer Science, 2015, 9351: 234-241.
[4] HITENDRA G, BHISHAM S, SHASHI S, et al.Spoofing detection system for e-health digital twin using EfficientNet Convolution Neural Network[J]. Multimedia Tools Appl, 2022, 81(19): 26873-26888.
[5] ZHUANG F, QI Z, DUAN K, et al.A Comprehensive Survey on Transfer Learning[J]. Proc IEEE, 2021, 109(1): 43-76.
[6] Morid M A, Borjali A, Del Fiol G.A Scoping Review Of Transfer Learning Research On Medical Image Analysis Using Imagenet[J]. Comput Biol Med, 2021, 128: 104115.
[7] NAN Y, JU J, HUA Q, et al.A-MobileNet: An approach of facial expression recognition[J]. Alex Eng J, 2022, 61(6): 4435-4444.
[8] JIA L, WANG Y, ZANG Y, et al.MobileNetV3 With CBAM for Bamboo Stick Counting[J]. IEEE Access, 2022, 10: 53963-53971.
[9] ROY A G, NAVAB N, WACHINGER C.Recalibrating Fully Convolutional Networks with Spatial and Channel 'Squeeze & Excitation' Blocks[J]. IEEE Trans Med Imaging, 2019, 38(2): 540-549.
[10] Arnon A, Brian T, David B, et al.Deep Residual Learning in Spiking Neural Networks[J]. Semantic Scholar, 2021.
[11] YAO Z, CAO Y, ZHENG S, et al. Cross-Iteration Batch Normalization[J]. Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2021, abs/2002.05712: 12331-12340.
[12] SU J, LIU Z, ZHANG J, et al.DV-Net: Accurate liver vessel segmentation via dense connection model with D-BCE loss function[J]. Knowl-Based Syst, 2021, 232: 107471.
[13] ABRAHAM N, KHAN N.Multimodal Segmentation with MGF-Net and the Focal Tversky Loss Function[J]. International MICCAI Brainlesion Workshop, 2019: 191-198.
[14] JHA S, SON L H, KUMAR R, et al.Neutrosophic image segmentation with Dice Coefficients[J]. Measurement, 2019, 134: 762-772.
[15] DOU H, KARIMI D, ROLLINS C K, et al.A Deep Attentive Convolutional Neural Network for Automatic Cortical Plate Segmentation in Fetal MRI[J]. IEEE Trans Med Imaging, 2021, 40(4): 1123-1133.
[16] THABTAH F, HAMMOUD S, KAMALOV F, et al.Data imbalance in classification: Experimental evaluation[J]. Inf Sci, 2020, 513: 429-441.
[17] ZHAO B, ZHANG X, LI H, et al.Intelligent fault diagnosis of rolling bearings based on normalized CNN considering data imbalance and variable working conditions[J]. Knowl-Based Syst, 2020, 199: 105971.
[18] PASSALIS N, TEFAS A, KANNIAINEN J, et al.Deep Adaptive Input Normalization for Time Series Forecasting[J]. IEEE Trans Neural Netw Learn Syst, 2020, 31(9): 3760-3765.
[19] HU C, ZHANG J, YUAN H, et al.Black swan event small-sample transfer learning (BEST-L) and its case study on electrical power prediction in COVID-19[J]. Appl Energy, 2022, 309: 118458.
[20] TAN Z, LUO L, ZHONG J.Knowledge transfer in evolutionary multi-task optimization: a survey[J]. Appl Soft Comput, 2023, 138: 110182.
[21] 叶晓敏, 谢富强. X线检查用于小动物临床应注意的问题[J]. 中国兽医杂志, 2006, 42(7): 2.
[22] DAVID J S, KYLE M G, YUKI H, et al.Integrated radiofrequency array and animal holder design for minimizing head motion during awake marmoset functional magnetic resonance imaging[J]. Neuroimage, 2019, 193: 126-138.
[23] 胡峻峰, 张志超, 赵亚凤, 等. 基于迁移学习的竹片缺陷识别[J]. 西北林学院学报, 2021, 36(5): 190-196.