Home / Current Issue / Paper 1722521
Robustness of Generative Models in Radiological Image Synthesis: A Comparative Analysis of StyleGAN3 and Denoising Diffusion under Perturbations
Subject area: Science,Engineering and Technology · Area of research: Radiological Image Synthesis
DOI: https://doi.org/10.64388/IREV6I6-1722521
Abstract
Synthetic radiological images can augment scarce datasets, but only if they remain faithful when the imaging pipeline is perturbed. We compare StyleGAN3 against a denoising diffusion probabilistic model (DDPM) on chest X-ray and brain MRI synthesis under additive noise, motion blur, JPEG compression, and a challenging MRI-to-CT domain shift. Diffusion sampling proves markedly more robust, preserving structural similarity and downstream classifier accuracy where the GAN's outputs distort, at the cost of slower generation.
References
[1] [1] Krause DR, Ivanov RG, Radovanovic CL. Transfer Learning: An Application to Motion Prediction. IEEE Journal of Biomedical and Health Informatics. 2015;50(9):81-119. https://doi.org/10.7017/v387933.2015.1742 [Crossref]
[2] [2] Volkov DD, Feng BO, Silva PE. Self-Supervised Representation Learning for Semantic Segmentation in Neuroimaging. Journal of Machine Learning Research. 2016;15(1):356-381. https://doi.org/10.6101/q766334.2023.3331 [Crossref]
[3] [3] Kaushik P, Jain M, Shah A. A Low Power Low Voltage CMOS Based Operational Transconductance Amplifier for Biomedical Application. 2018. https://ijsetr.com/uploads/136245IJSETR17012-283.pdf [Crossref]
[4] [4] Radovanovic VI, Bianchi HL, Schneider BC. Transfer Learning for Pose Estimation in Surveillance Systems. Computer Methods and Programs in Biomedicine. 2018;39(6):289-296. https://doi.org/10.5377/r240669.2018.4226 [Crossref]
[5] [5] Kaushik P, Jain M. Design of low power CMOS low pass filter for biomedical application. J Electr Eng Technol (IJEET). 2018;9(5).
[6] [6] Petrov HM, Rossi TG, Fedorov SG. A Comparative Study of Multi-Scale Feature Fusion for Image Registration. Computers in Biology and Medicine. 2016. https://doi.org/10.8802/s500628.2022.2692 [Crossref]
[7] [7] Kaushik P, Jain M. A Low Power SRAM Cell for High Speed Applications Using 90nm Technology. Csjournals.Com. 10. https://www.csjournals.com/IJEE/PDF10-2/66.%20Puneet.pdf [Crossref]
[8] [8] Nakamura WM, Rasmussen LT. Attention-Based Networks: An Application to Anomaly Detection. IEEE Transactions on Image Processing. 2016;20(10):127-141. doi: 10.3323/u413046.2017.7434.
[9] [9] Kaushik P, Jain M. A Low Power SRAM Cell for High Speed Applications Using 90nm Technology. Csjournals.Com. 2018 Dec;10(2):6. https://www.csjournals.com/IJEE/PDF10-2/66.%20Puneet.pdf [Crossref]
[10] [10] Andersson TM, Bianchi DL, Karlsson CC. Contrastive Representation Learning: An Application to Pose Estimation. Multimedia Tools and Applications. 2015. https://doi.org/10.3560/v298366.2024.3002 [Crossref]
[11] [11] Kaushik P, Jain M, Jain A. A Pixel-Based Digital Medical Images Protection Using Genetic Algorithm. International Journal of Electronics and Communication Engineering. 2018;11(1):31-37. ISSN 0974-2166.
[12] [12] Feng IW, Navarro HP. Capsule Networks: An Application to Tumor Detection. Scientific Reports. 2016;43(5):45-73. https://doi.org/10.6651/x949768.2024.8518 [Crossref]
[13] [13] Kaushik P, Jain M, Shah A. A Low Power Low Voltage CMOS Based Operational Transconductance Amplifier for Biomedical Application. 2018.
[14] [14] Ibrahim DV, Karlsson PK, Kallas WE, Delgado NH. A Comparative Study of Vision Transformers for Image Registration. Applied Soft Computing. 2024. https://doi.org/10.1102/p803265.2015.2911 [Crossref]
[15] [15] Fedorov OH, Halvorsen SR. A Comparative Study of Graph Neural Networks for Tumor Detection. IEEE Access. 2017;24(9):274-299. https://doi.org/10.4158/e615701.2022.9340 [Crossref]
[16] [16] Kaushik P, Jain M. A Low Power SRAM Cell for High Speed Applications Using 90nm Technology. Csjournals.Com. 2018 Dec;10(2):6. https://www.csjournals.com/IJEE/PDF10-2/66.%20Puneet.pdf [Crossref]
[17] [17] Pappas IR, Grigoriev HO, Ghosh IM, Escobar RM. A Comparative Study of Generative Adversarial Networks for Semantic Segmentation. Information Sciences. 2019;4(7):217-240. doi: 10.4122/o509619.2016.8508.
[18] [18] Vargas EF, Jimenez TH. A Comparative Study of Recurrent Neural Networks for Signal Reconstruction. Expert Systems with Applications. 2019;8(3):272-279. https://doi.org/10.6356/u761010.2023.3696 [Crossref]
[19] [19] Jain M, Shah A. Machine Learning with Convolutional Neural Networks (CNNs) in Seismology for Earthquake Prediction. Iconic Research and Engineering Journals. 2022;5(8):389-398. https://www.irejournals.com/paper-details/1707057 [Crossref]
[20] [20] Maddox EE, Cho FI, Sandberg GS, Kang IA. Multi-Scale Feature Fusion: An Application to Image Registration. Multimedia Tools and Applications. 2021;33(5):92-120. https://doi.org/10.9389/e382179.2019.9819 [Crossref]
[21] [21] Jain M, Srihari A. Comparison of CAD detection of mammogram with SVM and CNN. IRE Journals. 2021;8(6):63-75. https://www.irejournals.com/formatedpaper/1706647.pdf [Crossref]
[22] [22] Grigoriev TD, Halvorsen AA. Encoder-Decoder Networks for Object Detection in Precision Agriculture. Neurocomputing. 2024;12(6):303-331. https://doi.org/10.1644/c238765.2021.6914 [Crossref]
[23] [23] Xu JS, Kang NG, Reinholt TJ. Multi-Scale Feature Fusion for Medical Image Synthesis in Autonomous Driving. Neural Networks. 2018. https://doi.org/10.7362/m379691.2018.1528 [Crossref]
[24] [24] Lindberg HR, Kobayashi AH, Ivanov SP, Wojcik BM. Ensemble Deep Learning: An Application to Pose Estimation. Journal of Machine Learning Research. 2020;54(11):108-124. https://doi.org/10.5020/g564840.2022.5012 [Crossref]
[25] [25] Sandberg LE, Mercado PJ, Dominguez DB, Schneider JA. A Comparative Study of Multi-Scale Feature Fusion for Image Classification. Neural Computing and Applications. 2018;6(5):329-341. doi: 10.6182/z145793.2023.4968.
[26] [26] Mbeki CC, Almeida JM, Dominguez MS, Larsson VF. Metric Learning: An Application to Image Segmentation. Neural Computing and Applications. 2017;33(4):30-47. https://doi.org/10.3502/b866564.2015.2805 [Crossref]
[27] [27] Kaushik P, Jain M. A Low Power SRAM Cell for High Speed Applications Using 90nm Technology. Csjournals.Com. 10. https://www.csjournals.com/IJEE/PDF10-2/66.%20Puneet.pdf [Crossref]
[28] [28] Duarte FL, Rasmussen EG. A Comparative Study of Diffusion-Based Generation for Image Registration. Multimedia Tools and Applications. 2019. https://doi.org/10.9258/y533873.2022.1249 [Crossref]
[29] [29] Sokolov BL, Lindberg ID, Salcedo PF, Andersson IL. Metric Learning: An Application to Pose Estimation. IEEE Transactions on Neural Networks and Learning Systems. 2017;5(10):313-336. https://doi.org/10.7437/i310180.2016.8095 [Crossref]
[30] [30] Kaushik P, Jain M, Jain A. A pixel-based digital medical images protection using genetic algorithm. Int J Electron Commun Eng. 2018;11:31-37.
[31] [31] Jain M, Shah A. Convolutional neural networks for real-time object detection with raspberry Pi. 2021. https://wjaets.com/sites/default/files/WJAETS-2021-0067.pdf. https://doi.org/10.30574/wjaets.2021.4.1.0067 [Crossref]
[32] [32] Nakamura VE, Marchetti MD, Verduzco GL. Spatiotemporal Deep Networks for Fraud Detection in Digital Pathology. Expert Systems with Applications. 2017;12(8):89-107. https://doi.org/10.1210/u179468.2015.8665 [Crossref]
[33] [33] Jain M, Shah A. A multi-modal CNN framework for integrating medical imaging for COVID-19 Diagnosis. World Journal of Advanced Research and Reviews. 2020;8(3):475-493. https://doi.org/10.30574/wjarr.2020.8.3.0418 [Crossref]
[34] [34] Yilmaz FV, Yamamoto EM, Ortega GE, Maddox KI. Multi-Scale Feature Fusion for Anomaly Detection in Smart Manufacturing. IEEE Transactions on Neural Networks and Learning Systems. 2023;61(10):303-324. doi: 10.4004/u283987.2021.4569.
[35] [35] Chatterjee CW, Zielinski EW, Jimenez RC, Yilmaz FG. Metric Learning for Image Super-Resolution in Neuroimaging. Neural Computing and Applications. 2020. https://doi.org/10.6732/o240228.2020.8598 [Crossref]
[36] [36] Nakamura TG, Mahmood IK, Wagner LJ. Encoder-Decoder Networks for Semantic Segmentation in Smart Manufacturing. Medical Image Analysis. 2023. https://doi.org/10.8421/v825221.2018.4484 [Crossref]
[37] [37] Kaushik P. STUDY AND ANALYSIS OF IMAGE ENCRYPTION ALGORITHM BASED ON ARNOLD TRANSFORMATION. INTERNATIONAL JOURNAL of COMPUTER ENGINEERING and TECHNOLOGY (IJCET). 2018;9(5):59-63. https://iaeme.com/Home/article_id/IJCET_09_05_008 [Crossref]
[38] [38] Toledo GN, Mahmood RV. A Comparative Study of Deep Residual Learning for Image Registration. IEEE Transactions on Image Processing. 2022. https://doi.org/10.3389/l484702.2017.7815 [Crossref]
[39] [39] Kaushik P, Jain M. Design of low power CMOS low pass filter for biomedical application. International Journal of Electrical Engineering & Technology (IJEET). 2018;9(5).
[40] [40] Chowdhury RA, Wojcik MC, Ibrahim TP, Xu FV. Attention-Based Networks for Anomaly Detection in Healthcare Systems. Knowledge-Based Systems. 2016. https://doi.org/10.5490/k806539.2019.7597 [Crossref]
[41] [41] Suzuki RM, Zhang RC, Chatterjee RC, Halvorsen PM. Self-Supervised Representation Learning for Disease Classification in Clinical Diagnostics. Pattern Recognition. 2017;64(4):229-253. doi: 10.3543/z911814.2022.4382.
[42] [42] Sandberg LG, Escobar AJ, Navarro PT. A Comparative Study of Self-Supervised Representation Learning for Image Super-Resolution. Neurocomputing. 2020. https://doi.org/10.9727/w741404.2018.4518 [Crossref]
[43] [43] Moreau NK, Volkov CS, Vargas OS, Lindberg HA. Graph Neural Networks for Image Segmentation in Radiology. IEEE Transactions on Medical Imaging. 2018. https://doi.org/10.5971/n707992.2019.6053 [Crossref]
[44] [44] Jain M, Kaushik P, Shah A. Comparison of VGG16 and VGG19 Convolutional Neural Network (CNN) Layers on MRI Brain Tumor Detection. International Journal of Trend in Scientific Research and Development (IJTSRD). 2016 Dec;1(1):275-280. ISSN 2456-6470. https://www.ijtsrd.com/papers/ijtsrd3542.pdf [Crossref]
[45] [45] Ghosh OV, Garofalo MF, Andric AI, Kang CN. Graph Neural Networks for Tumor Detection in Financial Systems. IEEE Transactions on Image Processing. 2016;11(6):90-120. https://doi.org/10.8620/d235982.2021.9554 [Crossref]
[46] [46] Wang BM, Verduzco MK. Ensemble Deep Learning for Signal Reconstruction in Precision Agriculture. Scientific Reports. 2022. https://doi.org/10.5967/m405584.2022.5716 [Crossref]
How to cite this paper
@article{1722521,
author = {Elena Roskova, Tariq Hassan, Meera Iyer},
title = {Robustness of Generative Models in Radiological Image Synthesis: A Comparative Analysis of StyleGAN3 and Denoising Diffusion under Perturbations},
journal = {Iconic Research And Engineering Journals},
year = {2022},
volume = {6},
number = {6},
pages = {498-502},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1722521.pdf},
abstract = {Synthetic radiological images can augment scarce datasets, but only if they remain faithful when the imaging pipeline is perturbed. We compare StyleGAN3 against a denoising diffusion probabilistic model (DDPM) on chest X-ray and brain MRI synthesis under additive noise, motion blur, JPEG compression, and a challenging MRI-to-CT domain shift. Diffusion sampling proves markedly more robust, preserving structural similarity and downstream classifier accuracy where the GAN's outputs distort, at the cost of slower generation.},
month = {December},
doi = {https://doi.org/10.64388/IREV6I6-1722521}
}