An ensemble WideResNet learning-based approach for classification of multi-class colorectal cancer tissue types in histology images

Document Type : Research Article

Authors

1 Cancer Epidemiology Research Center, Aja University of Medical Sciences, Tehran, Iran

2 Department of Biomedical Engineering, Meybod University, Meybod, Iran

3 Infectious Diseases Research Center, Aja University of Medical Sciences, Tehran, Iran

Abstract

Histopathology Imaging (HI) plays a significant role in enhancing the prognosis of Colorectal Cancer (CRC), which ranks as the second leading cause of cancer-related deaths globally. Classifying colon cancer tissues with HI can be challenging due to differences in morphology, the presence of artifacts when recording microscopic images, and the lack of histological expertise. The use of WideResNet structure as a novel method for identifying textures in HIs is proposed using deep feature maps extracted from direct HIs and correlation matrices inputs. By using HSV (Hue, Saturation and Value) space, the first step is to decrease various artifacts during HI recording. After designing the ensemble model, the suggested structure has been evaluated and validated using the datasets CRC-5000 and NCT-CRC-HE-100K. On the CRC data set, the ensemble model had accuracy of 98.71% and 99.13%. Deep ensemble learning performed better than current methods in terms of computational and temporal performance, according to the results. Our proposed method is generalizable and has been tested on many of unseen HIs. Based on their data, pathologists can classify unseen Whole Slide Images (WSIs) images using only their lesion classes. The proposed tissue analysis mechanism reliably predicts CRC when pathological images are accurate.

Keywords

Main Subjects

References

References:
1.Zhou, C., Jin, Y., Chen, Y., et al. “Histopathologyclassification and localization of colorectal cancer usingglobal labels by weakly supervised deep learning”,Comput. Med. Imaging Graph., 88, 101861 (2021).https://doi.org/10.1016/j.compmedimag.2021.101861.
2.Tripathy, A., Dash, J., Kancharla, S., et al. “Probiotics:a promising candidate for management of colorectalcancer”, Cancers, 13(13), p. 3178 (2021). https://doi.org/10.1016/j.compmedimag.2021.101861.
3.van den Berg, I., Coebergh van den Braak, R.R., vanVugt, J.L., et al. “Actual survival after resection ofprimary colorectal cancer: Results from a prospectivemulticenter study”, World J. Surg. Oncol., 19(1), pp.1-10 (2021). https://doi.org/10.1186/s12957-021-02207-4.
4.Ottaiano, A., Santorsola, M., Circelli, L., et al.“Hypertension, type 2 diabetes, obesity, and p53 mutations negatively correlate with metastatic colorectal cancerpatients’ survival”, Front. Med., 10, 1091634 (2023).https://doi.org/10.3389/fmed.2023.1091634.
5.Jelski, W. and Mroczko, B. “Biochemical markers ofcolorectal cancer-present and future”, Cancer Manag.Res., 22, pp. 4789-97 (2020). https://doi.org/10.3389/fmed.2023.1091634.
6.Metze, K., Adam, R., and Florindo, J.B. “The fractaldimension of chromatin-a potential molecular markerfor carcinogenesis, tumor progression and prognosis”,Expert Rev. Mol. Diagn., 19(4), pp. 299-312 (2019). https://doi.org/10.1080/14737159.2019.1597707.
7.Sirinukunwattana, K., Raza, S.E., Tsang, Y.W., et al.“Locality sensitive deep learning for detection andclassification of nuclei in routine colon cancerhistology images”, IEEE Transactions on MedicalImaging, 35(5), pp.1196-1206 (2016). https://doi.org/10.1109/tmi.2016.2525803.
8.Jia, Z., Huang, X., Eric, I., et al. “Constrained deepweak supervision for histopathology imagesegmentation”, IEEE Transactions on MedicalImaging, 36(11), pp. 2376-2388 (2017). https://doi.org/10.1109/TMI.2017.2724070.
9.Sung, H., Ferlay, J., Siegel, R. L., et al. “Global cancerstatistics 2020: GLOBOCAN estimates of incidenceand mortality worldwide for 36 cancers in 185countries”, CA: A Cancer Journal for Clinicians,71(3), pp. 209-249 (2021). https://doi.org/10.3322/caac.21660.
10.Gadermayr, M., Dombrowski, A.K., Klinkhammer, B.M.,et al. “CNN cascades for segmenting sparse objects ingigapixel whole slide images”, Computerized MedicalImaging and Graphics, 71, pp. 40-48 (2019). https://doi.org/10.1016/j.compmedimag.2018.11.002.
11.Hatipoglu, N. and Bilgin, G. “Cell segmentation inhistopathological images with deep learningand Biological Engineering and Computing, 55(10), pp. 1829-1848 (2017). https://doi.org/10.1007/s11517-017-1630-1.
12.Kaushal, C., Bhat, S., Koundal, D., et al. “Recenttrends in computer assisted diagnosis (CAD) systemfor breast cancer diagnosis using histopathologicalimages”, Irbm, 40(4), pp. 211-227 (2019). https://doi.org/10.1016/j.irbm.2019.06.001.
13.Talo, M. “Automated classification of histopathologyimages using transfer learning”, Artificial Intelligencein Medicine, 101, 101743 (2019). https://doi.org/10.1016/j.artmed.2019.101743.
14.Mazo, C., Bernal, J., Trujillo, M., et al. “Transferlearning for classification of cardiovascular tissues inhistological images”, Computer Methods andPrograms in Biomedicine, 165, pp. 69-76 (2018). https://doi.org/10.1016/j.cmpb.2018.08.006.
15.George, K., Faziludeen, S., and Sankaran, P. “Breastcancer detection from biopsy images using nucleusguided transfer learning and belief based fusion”,Computers in Biology and Medicine, 124, 103954(2020). https://doi.org/10.1016/j.compbiomed.2020.103954.
16.Celik, Y., Talo, M., Yildirim, O., et al. “Automatedinvasive ductal carcinoma detection based using deeptransfer learning with whole-slide images”, PatternRecognition Letters, 133, pp. 232-239 (2020).https://doi.org/10.1016/j.patrec.2020.03.011.
17.Rezaee, K., Badiei, A., and Meshgini, S. “A hybrid deeptransfer learning-based approach for COVID-19classification in chest X-ray images”, In 2020 27thNational and 5th International Iranian Conference onBiomedical Engineering (ICBME), pp. 234-241 (2020).https://doi.org/10.1109/ICBME51989.2020.9319426.
18.Kleczek, P., Jaworek-Korjakowska, J., and Gorgon,M. “A novel method for tissue segmentation in high-resolution H&E-stained histopathological whole-slideimages”, Computerized Medical Imaging andGraphics, 79, 101686 (2020). https://doi.org/10.1016/j.compmedimag.2019.101686.
19.Amores, J. “Multiple instance classification: Review,taxonomy and comparative study”, ArtificialIntelligence, 201, pp. 81-105 (2013). https://doi.org/10.1016/j.artint.2013.06.003.
20.Haj-Hassan, H., Chaddad, A., Harkouss, et al.“Classifications of multispectral colorectal cancertissues using convolution neural network”, Journal ofPathology Informatics, 8, pp. 1-7 (2017). https://doi.org/10.4103/jpi.jpi_47_16.
21.Iizuka, O., Kanavati, F., Kato, K., et al. “Deep learning models for histopathological classification of gastricand colonic epithelial tumours”, Scientific Reports,10(1), pp. 1-11 (2020). https://doi.org/10.1038/s41598-020-58467-9.
22.Kwak, M.S., Lee, H.H., Yang, J.M., et al. “Deepconvolutional neural network-based lymph nodemetastasis prediction for colon cancer usinghistopathological images”, Frontiers in Oncology, 10,p.3053 (2020). https://doi.org/10.3389/fonc.2020.619803.
23.Korbar, B., Olofson, A.M., Miraflor, A.P., et al. “Deeplearning for classification of colorectal polyps onwhole-slide images”, Journal of PathologyInformatics, 8, pp. 1-30 (2017). https://doi.org/10.4103/jpi.jpi_34_17.
24.Manivannan, S., Li, W., Zhang, J., et al. “Structureprediction for gland segmentation with hand-craftedand deep convolutional features”, IEEE Transactionson Medical Imaging, 37(1), pp. 210-221 (2017). https://doi.org/10.1109/TMI.2017.2750210.
25.Ho, C., Zhao, Z., Chen, X.F, et al. “A promising deeplearning-assistive algorithm for histopathologicalscreening of colorectal cancer”, Scientific Reports,12(1), pp. 1-9 (2022). https://doi.org/10.1038/s41598-022-06264-x.
26.Chen, H., Li, C., Li, X, et al. “IL-MCAM: Aninteractive learning and multi-channel attentionmechanism-based weakly supervised colorectalhistopathology image classification approach”,Computers in Biology and Medicine, 143, 105265(2022). https://doi.org/10.1016/j.compbiomed.2022.105265.
27.Wang, K.S., Yu, G., Xu, C., et al. “Accurate diagnosisof colorectal cancer based on histopathology imagesusing artificial intelligence”, BMC Medicine, 19(1),pp. 1-12 (2021). https://doi.org/10.1186/s12916-021-01942-5.
28.Riasatian, A., Babaie, M., Maleki, D., et al. “Fine-tuning and training of densenet for histopathologyimage representation using tcga diagnostic slides”,Medical Image Analysis, 70, p. 102032 (2021). https://doi.org/10.1016/j.media.2021.102032.
29.Kassani, S.H., Kassani, P.H., Wesolowski, M.J, et al.“Deep transfer learning based model for colorectalcancer histopathology segmentation: A comparativestudy of deep pre-trained models”, Int. J. Med.Inform., 159, 104669 (2022). https://doi.org/10.1016/j.ijmedinf.2021.104669.
30.Raju, M.S. and Rao, B.S. “Colorectal multi-classimage classification using deep learning models”,Bull. Electr. Eng. Inform., 11(1), pp. 195-200 (2022). https://doi.org/10.11591/eei.v11i1.3299.
31.Bokhorst, J.M., Nagtegaal, I.D., Fraggetta, F., et al.“Deep learning for multi-class semantic segmentationenables colorectal cancer detection and classificationin digital pathology images”, Sci. Rep.,13(1), p. 8398(2023). https://doi.org/10.1038/s41598-023-35491-z.
32.Khazaee Fadafen, M. and Rezaee, K. “Ensemble-basedmulti-tissue classification approach of colorectal cancerhistology images using a novel hybrid deep learningframework”, Sci. Rep., 13(1), p. 8823 (2023). https://doi.org/10.1038/s41598-023-35431-x.
33.Ha, I., Kim, H., Park, S., et al. “Image retrieval usingBIM and features from pretrained VGG network forindoor localization”, Building and Environment, 140,pp. 23-31 (2018). https://doi.org/10.1016/j.buildenv.2018.05.026.
34.Li, Z., Liu, F., Yang, W., et al. “A survey ofconvolutional neural networks: Analysis applicationsand prospects”, CoRR, vol. abs/2004.02806 (2020). https://doi.org/10.1109/TNNLS.2021.3084827.
35.Zagoruyko, S. and Komodakis, N. “Wide residualnetworks”, arXiv preprint arXiv:1605, 07146 (2015). https://doi.org/10.48550/arXiv.1605.07146.
36.Kather, J.N., Weis, C.A., Bianconi, F., et al. “Multi-class texture analysis in colorectal cancer histology”,Scientific Reports, 6(1), pp. 1-11 (2016). https://doi.org/10.1038/srep27988.
37.Dhal, K.G., Ray, S., Das, S., et al. “Hue-preservingand gamut problem-free histopathology imageenhancement”, Iranian Journal of Science andTechnology, Transactions of Electrical Engineering,43(3), pp. 645-672 (2019). https://doi.org/10.1007/s40998-019-00175-w.
38.Ghosh, S., Bandyopadhyay, A., Sahay, S., et al.“Colorectal histology tumor detection using ensembledeep neural network”, Engineering Applications ofArtificial Intelligence, 100, 104202 (2021). https://doi.org/10.1016/j.engappai.2021.104202.
39.Hamida, A.B., Devanne, M., Weber, J., et al. “Deeplearning for colon cancer histopathological images analysis”, Computers in Biology and Medicine, 136, 104730 (2021). https://doi.org/10.1016/j.compbiomed.2021.104730.
40.Kather, J.N., Krisam, J., Charoentong, P., et al.“Predicting survival from colorectal cancer histologyslides using deep learning: A retrospective multicenterstudy”, PLoS Medicine, 16(1), e1002730 (2021). https://doi.org/10.1371/journal.pmed.1002730.
Scientia Iranica
Volume 32, Issue 5
Transactions on Computer Science & Engineering and Electrical Engineering
March and April 2025 Article ID:6790

Files

History

  • Receive Date: 13 May 2022
  • Revise Date: 13 September 2023
  • Accept Date: 31 December 2023

Share

How to cite

Statistics