Abstract
Based on the Nottingham Histological Grading (NHG) system, precise and accurate segmentation of breast tumor regions is crucial, typically in the scoring decision in terms of tubule formation. The main purposes of this study are to propose a robust segmentation framework that could produce repeatable, reproducible, and cost-effective measurable outputs. A novel spatial neighborhood intensity constraint (SNIC) and knowledge-based clustering framework for breast tumor region segmentation are proposed in this study. The proposed clustering framework constitutes of five different stages. These stages are color normalization, segmentation and removal of nucleus cells, SNIC, FCM with knowledge-based initial centroids selection, and post-processing. The main contribution of the proposed clustering framework lies within its simplicity yet powerful clustering ability producing precise segmentation in breast tumor regions using histopathology images. The SNIC is meant to remove the original RGB intensity of the nucleus cells and replace it with new intensity using the spatial information from the neighborhood pixels. Additionally, a knowledge-based initial centroids selection method is implemented to facilitate the fuzzy clustering algorithm. The proposed SNIC and knowledge-based initial centroids selection methods are expected to ease the clustering stage producing complementary results. For hypothesis validation purposes, justifications and validations are performed to validate the applicability of the proposed SNIC and knowledge-based initial centroids selection. Both of the proposed methods are found robust and plausible. The proposed framework achieved overall Accuracy (Acc), F1 score (F1), Area Overlap Measure (AOM), and Combined Equal Importance (CEI) of 91.2%, 92.1%, 85.7%, and 90.1%, respectively. For benchmarking purposes, the output results are compared to four recent works. The proposed framework demonstrates its robustness by outperformed these works achieving better results in Acc, F1, AOM, and CEI by 5.8%, 5.9%, 8.5%, and 7.0% higher than that of the best performance work amongst the four recent works.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig12_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig13_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig14_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11042-022-12129-2/MediaObjects/11042_2022_12129_Fig15_HTML.png)
Similar content being viewed by others
References
Amal KRG, Arun LC (2017) A complete color normalization method on pathological images. Int J Adv Res Innov Ideas Educ 2:60–68
Arai K, Kadoya N, Kato T, Endo H, Komori S, Abe Y, Nakamura T, Wada H, Kikuchi Y, Takai Y, Jingu K (2017) Feasibility of CBCT-based proton dose calculation using a histogram-matching algorithm in proton beam therapy. Phys Medica 33:68–76. https://doi.org/10.1016/j.ejmp.2016.12.006
Arjmand A, Meshgini S, Afrouzian R, Farzamnia A (2019) Breast tumor segmentation using K-means clustering and cuckoo search optimization. Int Conf Comput Knowl Eng ICCKE. https://doi.org/10.1109/ICCKE48569.2019.8964794
Ashiba HI (2021) A proposed framework for diagnosis and breast cancer detection. Mult Tools Appl 80:9333–9369. https://doi.org/10.1007/s11042-020-10131-0
Cebeci Z, Yildiz F (2015) Comparison of K-means and fuzzy C-means algorithms on different cluster structures. J Agric Inf 3:13–23. https://doi.org/10.17700/jai.2015.6.3.196
Das DK, Dutta PK (2019) Efficient automated detection of mitotic cells from breast histological images using deep convolution neutral network with wavelet decomposed patches. Comput Biol Med 104:29–42. https://doi.org/10.1016/j.compbiomed.2018.11.001
Elston CW, Ellis IO (1991) Pathological prognostic factors in breast cancer. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 19(5):403–410
Fouad S, Randell D, Galton A, Mehanna H, Landini G (2017) Unsupervised superpixel-based segmentation of histopathological images with consensus clustering. Commun Comput Inf Sci 723:767–779. https://doi.org/10.1007/978-3-319-60964-5_67
Ganesan P, Sathish BS, Sajiv G (2016) Automatic segmentation of fruits in CIELuv color space image using hill climbing optimization and fuzzy C-means clustering. IEEE WCTFTR 2016: Proc World Conf Futur Trends Res Innov Soc Welf. https://doi.org/10.1109/STARTUP.2016.7583960
George K, Faziludeen S, Sankaran P, Joseph P (2020) Breast cancer detection from biopsy images using nucleus guided transfer learning and belief based fusion. Comput Biol Med 124:103954. https://doi.org/10.1016/j.compbiomed.2020.103954
Gosain A, Dahiya S (2016) Performance analysis of various fuzzy clustering algorithms: a review. Procedia Comput Sci 79:100–111. https://doi.org/10.1016/j.procs.2016.03.014
Haralick RM, Dinstein I, Shanmugam K (1973) Textural features for image classification. IEEE Trans Syst Man Cybern 3:610–621. https://doi.org/10.1109/TSMC.1973.4309314
Khan AM, El-daly H, Rajpoot N (2012) RanPEC: random projections with ensemble clustering for segmentation of tumor areas in breast histology images. Med Image Underst Anal
Khan AM, El-Daly H, Simmons E, Rajpoot NM (2013) HyMaP: a hybrid magnitude-phase approach to unsupervised segmentation of tumor areas in breast cancer histology images. J Pathol Inform 4:S1. https://doi.org/10.4103/2153-3539.109802
Li X, Plataniotis KN (2015) A complete color normalization approach to histopathology images using color cues computed from saturation-weighted statistics. IEEE Trans Biomed Eng 62:1862–1873. https://doi.org/10.1109/TBME.2015.2405791
Linder N, Konsti J, Turkki R, Rahtu E, Lundin M, Nordling S, Haglund C, Ahonen T, Pietikäinen M, Lundin J (2012) Identification of tumor epithelium and stroma in tissue microarrays using texture analysis. Diagn Pathol 7(22):1–11. https://doi.org/10.1186/1746-1596-7-22
Liu Q, Zhou B, Li S, Li AP, Zou P, Jia Y (2016) Community detection utilizing a novel multi-swarm fruit fly optimization algorithm with hill-climbing strategy. Arab J Sci Eng 41(3):807–828. https://doi.org/10.1007/s13369-015-1905-5
Majeed H, Nguyen T, Kandel M, Marcias V, Do M, Tangella K, Balla A, Popescu G (2016) Automatic tissue segmentation of breast biopsies imaged by QPI. Quant Phase Imaging II. https://doi.org/10.1117/12.2209142
Mekhmoukh A, Mokrani K (2015) Improved fuzzy C-means based particle swarm optimization (PSO) initialization and outlier rejection with level set methods for MR brain image segmentation. Comput Methods Prog Biomed 122:266–281. https://doi.org/10.1016/j.cmpb.2015.08.001
Monaco J, Hipp J, Lucas D, Smith S, Balis U, Madabhushi A (2012) Image segmentation with implicit color standardization using spatially constrained expectation maximization: detection of nuclei. Lect Notes Comput Sci. https://doi.org/10.1007/978-3-642-33415-3_45
Nickfarjam AM, Soltaninejad S, Tajeripour F (2014) A novel supervised bi-level thresholding technique based on particle swarm optimization. Arab J Sci Eng 39(2):753–766. https://doi.org/10.1007/s13369-013-0638-6
Nobuyuki O (1979) A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybern. https://doi.org/10.1109/TSMC.1979.4310076
Pantanowitz L, Farahani N, Parwani A (2015) Whole slide imaging in pathology: advantages, limitations, and emerging perspectives. Pathol Lab Med Int 7(23):23–33. https://doi.org/10.2147/PLMI.S59826
Qu AP, Chen JM, Wang LW, Yuan JP, Yang F, Xiang QM, Maskey N, Yang GF, Liu J, Li Y (2015) Segmentation of hematoxylin-eosin stained breast cancer histopathological images based on pixel-wise SVM classifier. Sci China Inf Sci 58:1–13. https://doi.org/10.1007/s11432-014-5277-3
Rakha EA, Reis-Filho JS, Baehner F, Dabbs DJ, Decker T, Eusebi V, Fox SB, Ichihara S, Jacquemier J, Lakhani SR, Palacios J, Richardson AL, Schnitt SJ, Schmitt FC, Tan PH, Tse GM, Badve S, Ellis IO (2010) Breast cancer prognostic classification in the molecular era: the role of histological grade. Breast Cancer Res 12:207. https://doi.org/10.1186/bcr2607
Ramadijanti N, Barakbah A, Husna FA (2019) Automatic breast tumor segmentation using hierarchical K-means on mammogram. Int Electron Symp Knowl Creat Intell Comput. https://doi.org/10.1109/KCIC.2018.8628467
Salsabili S, Mukherjee A, Ukwatta E, Chan ADC, Bainbridge S, Grynspan D (2019) Automated segmentation of villi in histopathology images of placenta. Comput Biol Med 113:103420. https://doi.org/10.1016/j.compbiomed.2019.103420
Sebai M, Wang T, Al-Fadhli SA (2020) PartMitosis: a partially supervised deep learning framework for mitosis detection in breast cancer histopathology images. IEEE Access 8:45133–45147. https://doi.org/10.1109/ACCESS.2020.2978754
Sebai M, Wang X, Wang T (2020) MaskMitosis: a deep learning framework for fully supervised, weakly supervised, and unsupervised mitosis detection in histopathology images. Med Biol Eng Comput 58:1603–1623. https://doi.org/10.1007/s11517-020-02175-z
Shen S, Sandham W, Granat M, Sterr A (2005) MRI fuzzy segmentation of brain tissue using neighborhood attraction with neural-network optimization. IEEE Trans Inf Technol Biomed 9:459–467. https://doi.org/10.1109/TITB.2005.847500
Sigirci IO, Albayrak A, Bilgin G (2021) Detection of mitotic cells in breast cancer histopathological images using deep versus handcrafted features. Mult Tools Appl. https://doi.org/10.1007/s11042-021-10539-2
Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F (2021) Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71(3):209–249. https://doi.org/10.3322/caac.21660
Tan XJ, Mustafa N, Mashor MY, Ab Rahman KS (2019) An improved initialization based histogram of K-mean clustering algorithm for hyperchromatic nucleus segmentation in breast carcinoma histopathological images. Lect Notes Electr Eng. https://doi.org/10.1007/978-981-13-6447-1_67
Tariq M, Iqbal S, Ayesha H, Abbas I, Ahmad KT, Niazi MFK (2021) Medical image based breast cancer diagnosis: state of the art and future directions. Expert Syst Appl 167:114095. https://doi.org/10.1016/j.eswa.2020.114095
Vo DM, Nguyen NQ, Lee SW (2019) Classification of breast cancer histology images using incremental boosting convolution networks. Inf Sci 482:123–138. https://doi.org/10.1016/j.ins.2018.12.089
Yu C, Chen H, Li Y, Peng Y, Li J, Yang F (2019) Breast cancer classification in pathological images based on hybrid features. Mult Tools Appl 78:21325–21345. https://doi.org/10.1007/s11042-019-7468-9
Acknowledgements
The authors gratefully acknowledge the financial support from the Fundamental Research Grant Scheme (FRGS) under a grant number of FRGS/1/2016/SKK06/UNIMAP/02/3 from the Ministry of Higher Education Malaysia.
Funding
This study was funded by the Ministry of Higher Education Malaysia under Fundamental Research Grant Scheme (FRGS) (FRGS/1/2016/SKK06/UNIMAP/02/3).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Ethical approval
The protocol of this study had been approved by Medical Research and Committee of National Medical Research Register (NMRR) Malaysia referring to the protocol number: NMRR-17-281-34236.
Consent for publication
Informed consent was obtained from all individual participants included in the study.
Conflicts of interest/competing interests
The authors declare that they have no conflict of interest.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Tan, X.J., Mustafa, N., Mashor, M.Y. et al. Spatial neighborhood intensity constraint (SNIC) and knowledge-based clustering framework for tumor region segmentation in breast histopathology images. Multimed Tools Appl 81, 18203–18222 (2022). https://doi.org/10.1007/s11042-022-12129-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-12129-2