MRI-based model for accurate prediction of P53 gene status in gliomas

Yulin Zhao; Fengning Liang; Yaru Cao; Teng Zhao; Lin Wang; Jinhui Xu; Hong Zhu; Yulin Zhao; Fengning Liang; Yaru Cao; Teng Zhao; Lin Wang; Jinhui Xu; Hong Zhu

doi:10.3934/era.2024142

Electronic Research Archive

2024, Volume 32, Issue 5: 3113-3129. doi: 10.3934/era.2024142

Previous Article Next Article

Research article Special Issues

MRI-based model for accurate prediction of P53 gene status in gliomas

1.
School of Medical Information and Engineering, Xuzhou Medical University, Xuzhou 221000, China
2.
Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo 14213, NY, USA

Received: 27 December 2023 Revised: 29 March 2024 Accepted: 12 April 2024 Published: 25 April 2024

The accurate diagnosis and treatment of gliomas depends largely on the understanding of the P53 gene status. In our study, we presented a robust deep learning model, CTD-RegNet (improved RegNet integrating CNN, vision transformer, and truth discovery), tailored for predicting P53 gene status in gliomas. Our model addressed common challenges of existing deep learning models, such as incomplete feature extraction and uncertainty. First, the model used the RegNet network as a basis for predicting P53 gene mutations by skillfully extracting heterogeneous features. Next, the RegNet network was enhanced by integrating the CNN and ViT modules to optimise feature extraction and computational efficiency. Finally, using the truth discovery algorithm, we iteratively refined model uncertainties, thereby improving prediction accuracy. Our experiments demonstrated the effectiveness of the CTD-RegNet model, achieving an impressive accuracy of 95.57% and an AUC score of 0.9789, outperforming existing P53 gene status prediction models. The non-invasive nature of our model minimised the economic burden and physical and psychological stress on patients, while providing critical insights for accurate clinical diagnosis and treatment of gliomas.
- glioma,
- P53,
- deep learning,
- truth discovery,
- uncertainty calibration
Citation: Yulin Zhao, Fengning Liang, Yaru Cao, Teng Zhao, Lin Wang, Jinhui Xu, Hong Zhu. MRI-based model for accurate prediction of P53 gene status in gliomas[J]. Electronic Research Archive, 2024, 32(5): 3113-3129. doi: 10.3934/era.2024142

Related Papers:

Abstract

The accurate diagnosis and treatment of gliomas depends largely on the understanding of the P53 gene status. In our study, we presented a robust deep learning model, CTD-RegNet (improved RegNet integrating CNN, vision transformer, and truth discovery), tailored for predicting P53 gene status in gliomas. Our model addressed common challenges of existing deep learning models, such as incomplete feature extraction and uncertainty. First, the model used the RegNet network as a basis for predicting P53 gene mutations by skillfully extracting heterogeneous features. Next, the RegNet network was enhanced by integrating the CNN and ViT modules to optimise feature extraction and computational efficiency. Finally, using the truth discovery algorithm, we iteratively refined model uncertainties, thereby improving prediction accuracy. Our experiments demonstrated the effectiveness of the CTD-RegNet model, achieving an impressive accuracy of 95.57% and an AUC score of 0.9789, outperforming existing P53 gene status prediction models. The non-invasive nature of our model minimised the economic burden and physical and psychological stress on patients, while providing critical insights for accurate clinical diagnosis and treatment of gliomas.

References

[1]	P. Sledzińska, M. Bebyn, J. Furtak, A. Koper, K. Koper, Current and promising treatment strategies in glioma, Rev. Neurosci., 34 (2022), 483–516. https://doi.org/10.1515/revneuro-2022-0060 doi: 10.1515/revneuro-2022-0060
[2]	Y. Iwatate, I. Hoshino, H. Yokota, F. Ishige, M. Itami, Y. Mori, et al., Radiogenomics for predicting p53 status, PD-L1 expression, and prognosis with machine learning in pancreatic cancer, Br. J. Cancer, 123 (2020), 1253–1261. https://doi.org/10.1038/s41416-020-0997-1 doi: 10.1038/s41416-020-0997-1
[3]	X. Sun, P. Pang, L. Lou, Q. Feng, Z. Ding, J. Zhou, Radiomic prediction models for the level of Ki-67 and p53 in glioma, J. Int. Med. Res., 48 (2020). https://doi.org/10.1177/0300060520914466 doi: 10.1177/0300060520914466
[4]	I. Ezawa, Y. Sawai, T. Kawase, A. Okabe, S. Tsutsumi, H. Ichikawa, et al., Novel p53 target gene FUCA1 encodes a fucosidase and regulates growth and survival of cancer cells, Cancer Sci., 107 (2016), 734–745. https://doi.org/10.1111/cas.12933 doi: 10.1111/cas.12933
[5]	D. N. Louis, P. Arie, W. Pieter, D. J. Brat, I. A. Cree, D. Figarella-Branger, et al., The 2021 WHO Classification of Tumors of the Central Nervous System: A summary, Neuro-Oncol., 23 (2021), 1231–1251. https://doi.org/10.1093/neuonc/noab106 doi: 10.1093/neuonc/noab106
[6]	K. Charnpreet, G. Urvashi, Artificial intelligence techniques for cancer detection in medical image processing: A review, Mater. Today Proc., 81 (2023), 806–809. https://doi.org/10.1016/j.matpr.2021.04.241 doi: 10.1016/j.matpr.2021.04.241
[7]	C. M. Moon, Y. Y. Lee, D. Y. Kim, W. Yoon, B. H. Baek, J. H. Park, et al., Preoperative prediction of Ki-67 and p53 status in meningioma using a multiparametric MRI-based clinical-radiomic model, Front. Oncol., 13 (2023), 1138069. https://doi.org/10.3389/fonc.2023.1138069 doi: 10.3389/fonc.2023.1138069
[8]	J. J. Jiang, L. M. Guan, Y. Guo, K. Xu, A preliminary study on the predictive efficacy of conventional T₂WI-based radiogenomics model for glioma p53 status, Chin. J. Clin. Med. Imaging, 32 (2021), 609–612.
[9]	I. T. Ashwini, J. T. Senders, S. Kremer, S. Devi, W. B. Gormley, O. Arnaout, et al., Survival prediction of glioblastoma patients-are we there yet? A systematic review of prognostic modeling for glioblastoma and its clinical potential, Neurosurgical Rev., 44 (2020), 2047–2057. https://doi.org/10.1007/s10143-020-01430-z doi: 10.1007/s10143-020-01430-z
[10]	B. Zhang, S. Qi, X. Pan, C. Li, Y. Yao, W. Qian, et al., Deep CNN model using CT radiomics feature mapping recognizes EGFR gene mutation status of lung adenocarcinoma, Front. Oncol., 10 (2021), 598721. https://doi.org/10.3389/fonc.2020.598721 doi: 10.3389/fonc.2020.598721
[11]	T. Noguchi, T. Ando, S. Emoto, H. Nozawa, K. Kawai, K. Sasaki, et al., Artificial intelligence program to predict p53 mutations in ulcerative colitis-associated cancer or dysplasia, Inflammatory Bowel Dis., 28 (2022), 1072–1080. https://doi.org/10.1093/ibd/izab350 doi: 10.1093/ibd/izab350
[12]	Y. S. Choi, S. Bae, J. H. Chang, S. G. Kang, S. H. Kim, J. Kim, et al., Fully automated hybrid approach to predict the IDH mutation status of gliomas via deep learning and radiomics, Neuro-Oncol., 23 (2021), 304–313. https://doi.org/10.1093/neuonc/noaa177 doi: 10.1093/neuonc/noaa177
[13]	Q. Xu, Q. Q. Xu, N. Shi, L. N. Dong, H. Zhu, K. Xu, A multitask classification framework based on vision transformer for predicting molecular expressions of glioma, Eur. J. Radiol., 157 (2022), 110560. https://doi.org/10.1016/j.ejrad.2022.110560 doi: 10.1016/j.ejrad.2022.110560
[14]	G. Madhuri, S. M. Kumar, O. Aparajita, GeneViT: Gene vision transformer with improved deepinsight for cancer classification, Comput. Biol. Med., 155 (2023), 106643. https://doi.org/10.1016/j.compbiomed.2023.106643 doi: 10.1016/j.compbiomed.2023.106643
[15]	C. Ma, Z. Huang, J. Xian, M. Gao, J. Xu, Improving uncertainty calibration of deep neural networks via truth discovery and geometric optimization, in Uncertainty in Artificial Intelligence, PMLR, (2021), 75–85. https://doi.org/10.48550/arXiv.2106.14662
[16]	L. R. Soenksen, T. Kassis, S. T. Conover, B. Marti-Fuster, J. S. Birkenfeld, J. Tucker-Schwartz, et al., Using deep learning for dermatologist-level detection of suspicious pigmented skin lesions from wide-field images, Sci. Transl. Med., 13 (2021), eabb3652. https://doi.org/10.1126/scitranslmed.abb3652 doi: 10.1126/scitranslmed.abb3652
[17]	I. Radosavovic, R. P. Kosaraju, R. Girshick, K. He, P. Dollár, Designing network design spaces, in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2020), 10428–10436.
[18]	Q. Q. Xu, Q. Xu, H. C. Xu, Y. L. Zhao, K. Xu, H. Zhu, Intelligent prediction of glioma IDH1 mutation status based on CnViT, J. Shandong Univ. Eng. Ed., 53 (2023), 127–134.
[19]	S. Jiang, G. J. Zanazzi, S. Hassanpour, Predicting prognosis and IDH mutation status for patients with lower-grade gliomas using whole slide images, Sci. Rep., 11 (2021), 16849. https://doi.org/s41598-021-95948-x
[20]	Y. S. Choi, S. Bae, J. H. Chang, S. G. Kang, S. H. Kim, J. Kim, et al., Fully automated hybrid approach to predict the IDH mutation status of gliomas via deep learning and radiomics, Neuro-Oncol., 23 (2020), 304–313. https://doi.org/10.1093/neuonc/noaa177 doi: 10.1093/neuonc/noaa177
[21]	M. B. Taha, M. T. Li, D. Boley, C. C. Chen, J. Sun, Detection of isocitrate dehydrogenase mutated glioblastomas through anomaly detection analytics, Neurosurgery, 89 (2021), 323–328. https://doi.org/10.1093/neuros/nyab130 doi: 10.1093/neuros/nyab130
[22]	R. K. Kawaguchi, M. Takahashi, M. Miyake, M. Kinoshita, S. Takahashi, K. Ichimura, et al., Assessing versatile machine learning models for glioma radiogenomic studies across hospitals, Cancers, 13 (2021), 3611. https://doi.org/10.3390/cancers13143611 doi: 10.3390/cancers13143611
[23]	S. Xie, R. Girshick, P. Dollár, Z. Tu, K. He, Aggregated residual transformations for deep neural networks, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2017), 1492–1500. https://doi.org/10.1109/CVPR.2017.634
[24]	S. Tummala, S. Kadry, S. A. C. Bukhari, H. T. Rauf, Classification of brain tumor from magnetic resonance imaging using vision transformers ensembling, Curr. Oncol., 29 (2020), 7498–7511. https://doi.org/10.3390/curroncol29100590 doi: 10.3390/curroncol29100590
[25]	L. B. Ammar, K. Gasmi, I. B. Ltaifa, ViT-TB: ensemble learning based ViT model for tuberculosis recognition, Cybern. Syst., 55 (2022), 634–653. https://doi.org/10.1080/01969722.2022.2162736 doi: 10.1080/01969722.2022.2162736
[26]	C. Guo, G. Pleiss, Y. Sun, K. Q. Weinberger, On calibration of modern neural networks, in International Conference on Machine Learning, (2017), 1321–1330. https://doi.org/10.48550/arXiv.1706.04599
[27]	B. Murugesan, B. Liu, A. Galdran, I. B. Ayed, J. Dolz, Calibrating segmentation networks with margin-based label smoothing, Med. Image Anal., 87 (2023), 102826. https://doi.org/10.1016/j.media.2023.102826 doi: 10.1016/j.media.2023.102826
[28]	T. Buddenkotte, L. E. Sanchez, M. Crispin-Ortuzar, R. Woitek, C. McCague, J. D. Brenton, et al., Calibrating ensembles for scalable uncertainty quantification in deep learning-based medical image segmentation, Comput. Biol. Med., 163 (2023), 107096. https://doi.org/10.1016/j.compbiomed.2023.107096 doi: 10.1016/j.compbiomed.2023.107096
[29]	C. Kevin, E. Ralph, Improved calibration of building models using approximate Bayesian calibration and neural networks, J. Build. Perform. Simul., 16 (2023), 291–307. https://doi.org/10.1080/19401493.2022.2137236 doi: 10.1080/19401493.2022.2137236
[30]	H. Xu, H. Zhang, Q. Li, T. Qin, Z. Zhang, A data-semantic-conflict-based multi-truth discovery algorithm for a programming site, Comput. Mater. Continuum., 68 (2021), 2681–2691. https://doi.org/10.32604/cmc.2021.016188 doi: 10.32604/cmc.2021.016188
[31]	H. Ding, J. Xu, Learning the truth vector in high dimensions, J. Comput. Syst. Sci., 109 (2020), 78–94. https://doi.org/10.1016/j.jcss.2019.12.002 doi: 10.1016/j.jcss.2019.12.002
[32]	J. J. Cao, C. Chang, N. F. Weng, J. Q. Tao, C. Jiang, Truth value discovery based on neural network coding, J. Comput. Syst. Sci., 43 (2021). https://doi.org/10.3969/j.issn.1007-130X.2021.09.004 doi: 10.3969/j.issn.1007-130X.2021.09.004
[33]	A. Kumar, P. Liang, T. Ma, Verified uncertainty calibration, Adv. Neural Inf. Process. Syst., 32 (2019). https://doi.org/10.48550/arXiv.1909.10155 doi: 10.48550/arXiv.1909.10155
[34]	J. Z. Liu, Z. Lin, S. Padhy, D. Tran, T. Bedrax-Weiss, B. Lakshminarayanan, Simple and principled uncertainty estimation with deterministic deep learning via distance awareness, Adv. Neural Inf. Process. Syst., 33 (2020), 7498–7512. https://doi.org/10.48550/arXiv.2006.10108 doi: 10.48550/arXiv.2006.10108

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)