Annotations of Gene Pathways Images in Biomedical Publications Using Siamese Network

Authors: Micheal Olaolu Arowolo, Muhammad Azam, Fei He, Mihail Popescu, Dong Xu

Abstract:

As the quantity of biological articles rises, so does the number of biological route figures. Each route figure shows gene names and relationships. Manually annotating pathway diagrams is time-consuming. Advanced image understanding models could speed up curation, but they must be more precise. There is rich information in biological pathway figures. The first step to performing image understanding of these figures is to recognize gene names automatically. Classical optical character recognition methods have been employed for gene name recognition, but they are not optimized for literature mining data. This study devised a method to recognize an image bounding box of gene name as a photo using deep Siamese neural network models to outperform the existing methods using ResNet, DenseNet and Inception architectures, the results obtained about 84% accuracy.

Keywords: Biological pathway, gene identification, object detection, Siamese network, ResNet.

References:

[1] N. Rosário-Ferreira et al., “The Treasury Chest of Text Mining: Piling Available Resources for Powerful Biomedical Text Mining,” BioChem, vol. 1, no. 2, pp. 60–80, Jul. 2021, doi: 10.3390/biochem1020007.
[2] K. Hanspers, A. Riutta, M. Summer-Kutmon, and A. R. Pico, “Pathway information extracted from 25 years of pathway figures,” Genome Biol, vol. 21, no. 1, p. 273, Dec. 2020, doi: 10.1186/s13059-020-02181-2.
[3] S. Kraus et al., “Literature reviews as independent studies: guidelines for academic practice,” Review of Managerial Science, vol. 16, no. 8, pp. 2577–2595, Nov. 2022, doi: 10.1007/s11846-022-00588-8.
[4] J. Egger et al., “Medical deep learning—A systematic meta-review,” Comput Methods Programs Biomed, vol. 221, p. 106874, Jun. 2022, doi: 10.1016/j.cmpb.2022.106874.
[5] C. G. Cess and S. D. Finley, “Representation learning for a generalized, quantitative comparison of complex model outputs,” Aug. 2022.
[6] J. Schmidt, M. R. G. Marques, S. Botti, and M. A. L. Marques, “Recent advances and applications of machine learning in solid-state materials science,” NPJ Comput Mater, vol. 5, no. 1, p. 83, Aug. 2019, doi: 10.1038/s41524-019-0221-0.
[7] C. Fotis, N. Meimetis, A. Sardis, and L. G. Alexopoulos, “DeepSIBA: chemical structure-based inference of biological alterations using deep learning,” Mol Omics, vol. 17, no. 1, pp. 108–120, 2021, doi: 10.1039/D0MO00129E.
[8] C. J. Kelly, A. Karthikesalingam, M. Suleyman, G. Corrado, and D. King, “Key challenges for delivering clinical impact with artificial intelligence,” BMC Med, vol. 17, no. 1, p. 195, Dec. 2019, doi: 10.1186/s12916-019-1426-2.
[9] J. M. Vaz and S. Balaji, “Convolutional neural networks (CNNs): concepts and applications in pharmacogenomics,” Mol Divers, vol. 25, no. 3, pp. 1569–1584, Aug. 2021, doi: 10.1007/s11030-021-10225-3.
[10] A. S. Lundervold and A. Lundervold, “An overview of deep learning in medical imaging focusing on MRI,” Z Med Phys, vol. 29, no. 2, pp. 102–127, May 2019, doi: 10.1016/j.zemedi.2018.11.002.
[11] M. Javaid, A. Haleem, R. Pratap Singh, R. Suman, and S. Rab, “Significance of machine learning in healthcare: Features, pillars and applications,” International Journal of Intelligent Networks, vol. 3, pp. 58–73, 2022, doi: 10.1016/j.ijin.2022.05.002.
[12] F. He et al., “Identifying Genes and Their Interactions from Pathway Figures and Text in Biomedical Articles,” in 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), IEEE, Dec. 2021, pp. 398–405. doi: 10.1109/BIBM52615.2021.9669391.
[13] R. Anders, H. Kristina, and R., P. Alexander, “Identifying Genes in Published Pathway Figure Images,” bioRxiv, 2018.
[14] K. A. Tran, O. Kondrashova, A. Bradley, E. D. Williams, J. V. Pearson, and N. Waddell, “Deep learning in cancer diagnosis, prognosis and treatment selection,” Genome Med, vol. 13, no. 1, p. 152, Dec. 2021, doi: 10.1186/s13073-021-00968-x.
[15] F. Alharbi and A. Vakanski, “Machine Learning Methods for Cancer Classification Using Gene Expression Data: A Review,” Bioengineering, vol. 10, no. 2, p. 173, Jan. 2023, doi: 10.3390/bioengineering10020173.
[16] V. Singh, S.-S. Chen, M. Singhania, B. Nanavati, A. kumar kar, and A. Gupta, “How are reinforcement learning and deep learning algorithms used for big data based decision making in financial industries–A review and research agenda,” International Journal of Information Management Data Insights, vol. 2, no. 2, p. 100094, Nov. 2022, doi: 10.1016/j.jjimei.2022.100094.
[17] B. Mandal, A. Okeukwu, and Y. Theis, “Masked Face Recognition using ResNet-50,” Apr. 2021.
[18] R. Zeng and M. Liao, “Developing a Multi-Layer Deep Learning Based Predictive Model to Identify DNA N4-Methylcytosine Modifications,” Front Bioeng Biotechnol, vol. 8, Apr. 2020, doi: 10.3389/fbioe.2020.00274.
[19] J. Zhang and A. Zhang, “Deep learning-based multi-model approach on electron microscopy image of renal biopsy classification,” BMC Nephrol, vol. 24, no. 1, p. 132, May 2023, doi: 10.1186/s12882-023-03182-6.
[20] I. P. E. Fábia, Freitas. de M. Erikson, and S. R. M. Marcella, “Person Re-Identication Using Convolutional Neural Network and Autoencoder Embedded on Frameworks Based on Siamese and Triplet Networks,” Res Sq, pp. 1–21, 2020.
[21] F. Ren and S. Xue, “Intention Detection Based on Siamese Neural Network with Triplet Loss,” IEEE Access, vol. 8, pp. 82242–82254, 2020, doi: 10.1109/ACCESS.2020.2991484.
[22] B. Lodhi and J. Kang, “Multipath-DenseNet: A Supervised ensemble architecture of densely connected convolutional networks,” Inf Sci (N Y), vol. 482, pp. 63–72, May 2019, doi: 10.1016/j.ins.2019.01.012.