Abstract
Early detection of oral cancer is essential for improving survival rates and diagnostic precision. Raman cyto-spectroscopy is a useful method, but its interpretive complexity prevents frequent clinical use. Prediction algorithms have been used to evaluate RS data and create CAD systems. An ensemble of MLAs and dimensionality reduction techniques has been demonstrated to effectively analyse the complex RS. In present study we utilised two MLA ensembles—(1) Linear Discriminant Analysis with Decision Tree Classifier (2) LDA with Support Vector Machine. Results show that RS of oral epithelial cells, collected from different study groups (viz. Normal-healthy volunteers; histo-pathologically confirmed- potentially cancerable- leucoplakia and cancer patients-OSCC) when classified using the ensembles had better overall accuracy of (88 ± 8%) as compared to either DTC (48 ± 19%) or SVM (58 ± 15%) when applied individually. Class-wise evaluation showed improved performance of LDA-SVM compared to LDA-DTC, and blind test was performed with RS of habitual cigarette smokers (CS) as susceptible group and healthy volunteers as control group to assess the classifier ensembles' performance in cancer risk prediction. The LDA-SVM ensemble showed the highest classification accuracy, indicating that it can be used to accurately detect oral cancer risk and screen susceptible patients.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
All data generated or analysed during this study are included in this published article.
References
Ariyawardana A (2020) Malignant transformation of oral potentially malignant disorders. In: Warnakulasuriya S, Greenspan J (eds) Textbook of oral cancer. Textbooks in contemporary dentistry. Springer, Cham. https://doi.org/10.1007/978-3-030-32316-5_13
At A, Sm E-M (2012) Decision tree classifiers for automated medical diagnosis. Neural Comput Appl 23:2387–2403
Auner GW et al (2018) Applications of Raman spectroscopy in cancer diagnosis. Cancer Metastasis Rev 37(4):691–717. https://doi.org/10.1007/s10555-018-9770-9
Banaei N et al (2019) Machine learning algorithms enhance the specificity of cancer biomarker detection using SERS-based immunoassays in microfluidic chips. RSC Adv 9(4):1859–1868. https://doi.org/10.1039/C8RA08930B
Burda Z et al (2004) ‘Signal and noise in correlation matrix’. Phys A Stat Mech Appl 343(C):295–310. Available at: https://econpapers.repec.org/RePEc:eee:phsmap:v:343:y:2004:i:c:p:295-310
Chakraborty A, Ghosh A, Barui A (2019) Advances in surface-enhanced Raman spectroscopy for cancer diagnosis and staging. J Raman Spectrosc 51:7–36. https://doi.org/10.1002/jrs.5726
Chen P-H et al (2016) Automatic and objective oral cancer diagnosis by Raman spectroscopic detection of keratin with multivariate curve resolution analysis. Sci Rep 6(1):20097. https://doi.org/10.1038/srep20097
Chicco D, Jurman G (2020) The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genom 21(1):6. https://doi.org/10.1186/s12864-019-6413-7
Connolly JM et al (2016) ‘Non-invasive and label-free detection of oral squamous cell carcinoma using saliva surface-enhanced Raman spectroscopy and multivariate analysis. Nanomed Nanotechnol Biol Med 12(6):1593–1601. https://doi.org/10.1016/J.NANO.2016.02.021
Elumalai B et al (2015) Raman spectroscopic characterization of urine of normal and oral cancer subjects. J Raman Spectrosc 46(1):84–93. https://doi.org/10.1002/jrs.4601
Evgeniou T, Pontil M (2001) ‘Support vector machines: theory and applications BT. In: Paliouras G, Karkaletsis V, Spyropoulos CD (eds) Machine learning and its applications: advanced lectures. Springer, Berlin Heidelberg, pp 249–257
Falamas A et al (2021) Rapid and noninvasive diagnosis of oral and oropharyngeal cancer based on micro-Raman and FT-IR spectra of saliva. Spectrochim Acta Part A Mol Biomol Spectrosc 252:119477. https://doi.org/10.1016/j.saa.2021.119477
Farrelly CM (2017) ‘KNN ensembles for tweedie regression: the power of multiscale neighborhoods. arXiv:1708.02122
Gautam R et al (2015) Review of multidimensional data processing approaches for Raman and infrared spectroscopy. EPJ Tech Instrum. https://doi.org/10.1140/epjti/s40485-015-0018-6
Ghaddar B, Naoum-Sawaya J (2018) ‘High dimensional data classification and feature selection using support vector machines. Eur J Oper Res 265(3):993–1004. https://doi.org/10.1016/j.ejor.2017.08.040
Ghosh A et al (2019) Chemometric analysis of integrated FTIR and Raman spectra obtained by non-invasive exfoliative cytology for the screening of oral cancer. Analyst 144(4):1309–1325. https://doi.org/10.1039/c8an02092b
Ghosh A et al (2022) Deep reinforced neural network model for cyto-spectroscopic analysis of epigenetic markers for automated oral cancer risk prediction. Chemom Intell Lab Syst 224:104548. https://doi.org/10.1016/j.chemolab.2022.104548
Gigliotti J, Madathil S, Makhoul N (2019) Delays in oral cavity cancer. Int J Oral Maxillofac Surg 48(9):1131–1137. https://doi.org/10.1016/j.ijom.2019.02.015
Groen EJ, Hudecek J, Mulder L, van Seijen M, Almekinders MM, Alexov S, Kovács A, Ryska A, Varga ZN, Andreu Navarro FJ (2020) ‘Prognostic value of histopathological DCIS features in a large-scale international interrater reliability study.’ Breast Cancer Res Treat 183:759–770
Hanna K, Krzoska E, Shaaban AM, Muirhead D, Abu-Eid R, Speirs V (2021) Raman spectroscopy: current applications in breast cancer diagnosis, challenges and future prospects. Br J Cancer 126:1125–1139. https://doi.org/10.1038/s41416-021-01659-5
Hawkes N (2019) ‘Cancer survival data emphasise importance of early diagnosis’. BMJ 364
Hole A et al (2018) Exploration of Raman exfoliated cytology for oral and cervical cancers. Vib Spectrosc 98:35–40. https://doi.org/10.1016/j.vibspec.2018.07.001
Jeng MJ et al (2019) ‘Spectroscopy analysis for optical diagnosis of oral cancer detection’. J Clin Med 27
Kallaway C, Almond LM, Barr H, Wood J, Hutchings J, Kendall C, Stone N (2013) Advances in the clinical application of Raman spectroscopy for cancer diagnostics. Photodiagn Photodyn Ther 10:207–219. https://doi.org/10.1016/j.pdpdt.2013.01.008
Li X, Claramunt C (2006) ‘A spatial entropy-based decision tree for classification of geographical information.’ Transactions in GIS John 10(3):451–467. https://doi.org/10.1111/j.1467-9671.2006.01006.x
Li B et al (2017) Evaluating oral epithelial dysplasia classification system by near-infrared Raman spectroscopy. Oncotarget. https://doi.org/10.18632/oncotarget.19343
Martinez A, Kak, A (2001) PCA versus LDA. IEEE Transactions on Pattern Analysis Machine Intelligence, 23:228–233
Matthies L et al (2021) Optical diagnosis of oral cavity lesions by label-free Raman spectroscopy. Biomed Opt Express 12(2):836–851. https://doi.org/10.1364/BOE.409456
Nikitidis S, Zafeiriou S, Pantic M (2014) Merging SVMs with linear discriminant analysis: a combined model. In: IEEE Conference on Computer Vision and Pattern Recognition
Oliveira JL et al (eds) (2005) Biological and medical data analysis. Springer, Berlin, Heidelberg
Panta P, Wong DTW (2019) Saliva-based point-of-care in oral cancer detection: current trend and future opportunities. Oral cancer detection. Springer International Publishing, Cham, pp 297–314
Paraskevaidi M et al (2018) Potential of mid-infrared spectroscopy as a non-invasive diagnostic test in urine for endometrial or ovarian cancer. Analyst 143(13):3156–3163. https://doi.org/10.1039/C8AN00027A
Ren X et al (2020) Scalable nanolaminated SERS multiwell cell culture assay. Microsyst Nanoeng 6(1):47. https://doi.org/10.1038/s41378-020-0145-3
Riva M et al (2021) ‘Glioma biopsies classification using raman spectroscopy and machine learning models on fresh tissue samples’. Cancers 13
Sahoo GR et al (2018) Improving diagnosis of cervical pre-cancer: combination of PCA and SVM applied on fluorescence lifetime images. Photonics. https://doi.org/10.3390/photonics5040057
Sahu AK et al (2015) Oral cancer screening: serum Raman spectroscopic approach. J Biomed Opt 20(11):115006. https://doi.org/10.1117/1.JBO.20.11.115006
Sahu A, Tawde S et al (2015a) ‘Raman spectroscopy and cytopathology of oral exfoliated cells for oral cancer diagnosis.’ Anal Methods 7(18):7548–7559. https://doi.org/10.1039/C5AY00954E
Sahu A, Sawant S et al (2015b) Raman spectroscopy of serum: a study on oral cancers. Biomed Spectrosc Imaging 4(2):171–187. https://doi.org/10.3233/BSI-150107
Sahu A et al (2019) Raman exfoliative cytology for prognosis prediction in oral cancers: a proof of concept study. J Biophotonics. https://doi.org/10.1002/jbio.201800334
Santos IP, Barroso EM, Schut TCB, van CaspersLanschot PJCG et al (2017) Raman spectroscopy for cancer detection and cancer surgery guidance: translation to the clinics. Analyst 142:3025–3047
Schie IW et al (2016) Rapid acquisition of mean Raman spectra of eukaryotic cells for a robust single cell classification. Analyst 141(23):6387–6395. https://doi.org/10.1039/C6AN01018K
Shetty S, Rao YS (2016) ‘SVM based machine learning approach to identify Parkinson’s disease using gait analysis’. In: 2016 international conference on inventive computation technologies (ICICT), pp 1–5. https://doi.org/10.1109/INVENTIVE.2016.7824836
Suarez-Alvarez MM et al (2012) Statistical approach to normalization of feature vectors and clustering of mixed datasets. Proc R Soc Math Phys Eng Sci 468(2145):2630–2651. https://doi.org/10.1098/rspa.2011.0704
Trabelsi A, Elouedi Z, Lefevre E (2019) Decision tree classifiers for evidential attribute values and class labels. Fuzzy Sets Syst 366:46–62. https://doi.org/10.1016/j.fss.2018.11.006
Yan H et al (2019) Tongue squamous cell carcinoma discrimination with Raman spectroscopy and convolution neural networks. Vib Spectrosc 103(1):102938. https://doi.org/10.1016/j.vibspec.2019.102938
Ye J, Ji S (2009) Discriminant analysis for dimensionality reduction: an overview of recent developments. Biometrics. John Wiley & Sons Inc., Hoboken, NJ, USA, pp 1–19
Yu M et al (2019) Deep convolutional neural networks for tongue squamous cell carcinoma classification using Raman spectroscopy. Photodiagn Photodyn Ther 26:430–435. https://doi.org/10.1016/j.pdpdt.2019.05.008
Zhang X et al (2020) Deep learning based analysis of breast cancer using advanced ensemble classifier and linear discriminant analysis. IEEE Access 8:120208–120217. https://doi.org/10.1109/ACCESS.2020.3005228
Funding
The authors would like to acknowledge Indian Council of Medical Research (ICMR, India) for providing an ICMR-Senior Research Fellowship [IRIS ID -2017-4029] to AG.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no financial or commercial conflicts of interests regarding the publication of this paper.
Ethical approval
All experiments were performed in compliance with the Indian Institute of Engineering, Science and Technology’s ethical committee policy.
Informed consent
All samples were collected from patients under their informed written consent.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Chaudhuri, D., Ghosh, A., Raha, S. et al. Machine learning algorithm ensembles for early oral cancer risk assessment using Raman cyto-spectroscopy. Soft Comput 27, 13861–13875 (2023). https://doi.org/10.1007/s00500-023-08995-z
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-023-08995-z