Analysis of Cryotherapy Treatment of Verruca by Machine Learning

Abstract

There are several patient- and verruca-specific factors that determine treatment response to cryotherapy. A comprehensive analysis of these factors necessitates the use of a systematic and rational approach. The present study uses machine learning algorithms to analyze the clinical patient- and verruca-specific factors that affect the success of cryotherapy treatment. Machine learning algorithms were applied to the cryotherapy dataset. The best results in the prediction of treatment response to cryotherapy were achieved through the C&R Tree classification method, with a 96% accuracy rate, followed by the C5.0 Tree, CHAID Tree and the adjusted J48 Decision Tree algorithms, respectively. The C&R Tree classification method revealed that the most significant factors that affected treatment response in verrucae, in the order of importance, were the time to the first session, the patient’s age, the type of verruca, the number of verrucae and the region of the verruca. We believe that by identifying factors that affect treatment success and investigating the relations between variables, machine learning approaches can guide clinical treatment decisions for the more effective management of verruca treatment, which represent an important social and economic burden in public health.

Keywords

• CRISP-DM,cryotherapy dataset,machine learning

Analysis of Cryotherapy Treatment of Verruca by Machine Learning

Abstract

There are several patient- and verruca-specific factors that determine treatment response to cryotherapy. A comprehensive analysis of these factors necessitates the use of a systematic and rational approach. The present study uses machine learning algorithms to analyze the clinical patient- and verruca-specific factors that affect the success of cryotherapy treatment. Machine learning algorithms were applied to the cryotherapy dataset. The best results in the prediction of treatment response to cryotherapy were achieved through the C&R Tree classification method, with a 96% accuracy rate, followed by the C5.0 Tree, CHAID Tree and the adjusted J48 Decision Tree algorithms, respectively. The C&R Tree classification method revealed that the most significant factors that affected treatment response in verrucae, in the order of importance, were the time to the first session, the patient’s age, the type of verruca, the number of verrucae and the region of the verruca. We believe that by identifying factors that affect treatment success and investigating the relations between variables, machine learning approaches can guide clinical treatment decisions for the more effective management of verruca treatment, which represent an important social and economic burden in public health.

Keywords

• CRISP-DM,cryotherapy dataset,machine learning,wart treatment

References

1. [1] S. C. Bruggink, M. N. de Koning, J. Gussekloo, P. F. Egberts, J. ter Schegget, M. C. Feltkamp, J. N. Bavinck, W. G. Quint, W. J. Assendelft, J. A. Eekhof, “Cutaneous wart-associated HPV types: prevalence and relation with patient characteristics”, Journal of Clinical Virology, vol. 55, pp. 250-255, 2012.
2. [2] M. D. Lynch, J. Cliffe, R. Morris-Jones, “Management of cutaneous viral warts”, Bmj, 348, g3339, 2014.
3. [3] F. M. Van Haalen, S. C. Bruggink, J. Gussekloo, W. J. J. Assendelft and J. A. H. Eekhof, “Warts in primary schoolchildren: prevalence and relation with environmental factors”, The British Journal of Dermatology, 161, pp. 148-152, 2009.
4. [4] R. J. Hay, N. E. Johns, H. C. Williams, I. W. Boliger, R. P. Dellavale, D. J. Margolis, R. Marks, L. Naldi, M. A. Weinstock, S. K. Wulf, C. Michaud et al., “The global burden of skin disease in 2010: an analysis of the prevalence and impact of skin conditions”, Journal of Investigative Dermatology, 134, pp. 1527-1534, 2014.
5. [5] J. C. Sterling, S. Gibbs, S. S. Haque Hussain, M. F. Mohd Mustapa, S. E. Handfield-Jones, “British Association of Dermatologists' guidelines for the management of cutaneous warts” The British Journal of Dermatology, 171, pp. 696-712, 2014.
6. [6] G. K. Hogendoorn, S. C. Bruggink, M. N. C. de Koning, J. A. H. Eekhof, K. E. Hermans,R. Rissmann, J. Burggraaf, R. Wolterbeek, K. D. Quint, S. T. P. Kouwenhoven et al., “Morphological characteristics and human papillomavirus genotype predict the treatment response in cutaneous warts”, The British Journal of Dermatology, 178, pp. 253-260, 2017.
7. [7] G. Doğan, S. Şaşmaz, “Identification of the factors affecting the cryotherapy on warts (article in Turkish with an abstract in English)”. Journal Of Turgut Ozal Medical Center, 13, pp. 163-166, 2006.
8. [8] P. L. Bencini, S. Guida, S. Cazzaniga, G. Pellacani, M. G. Galimberti, M. Bencini and L. Naldi, “Risk factors for recurrence after successful treatment of warts: the role of smoking habits”, The Journal of the European Academy of Dermatology and Venereology, 31, pp. 712-716, 2017.

9. [9] S. Gibbs, D. G. Altman, I. Harvey, J. Sterling and R. Stark, “Local treatments for cutaneous warts: systematic review”, Bmj, 325, pp. 461, 2002.
10. [10] İ. İçke, P. Y. Başak, “Cryotherapy in Dermatology (article in Turkish with an abstract in English)”, Turkiye Klinikleri Journal of Medical Sciences, 24, pp. 383-395, 2004.
11. [11] M. J. Godley, C. S. Bradbeer, M. Gellan, R. N. Thin, “Cryotherapy compared with trichloroacetic acid in treating genital warts”, Sexually Transmitted Infections, 63, pp. 390-392, 1987.
12. [12] A. Khaled, S. R. Ben, M. Kharfi, F. Zeglaoui, B. Fazaa, M. R. Kamoun, “Assessment of cryotherapy by liquid nitrogen in the treatment of hand and feet warts”, Tunis Med, 87, pp. 690-692, 2009.
13. [13] I. Kononenko, “Machine learning for medical diagnosis: history, state of the art and perspective”, Artificial Intelligence in Medicine, 23, pp. 89-109, 2001.
14. [14] K. R. Foster, R. Koprowski and J. D. Skufca, “Machine learning, medical diagnosis, and biomedical engineering research-commentary”, BioMedical Engineering OnLine, 13, pp. 94, 2014.
15. [15] F. Khozeimeh, R. Alizadehsani, M. Roshanzamir, A. Khosravi, P. Layegh, S. Nahavandi, “ An expert system for selecting wart treatment method”, Computers in Biology and Medicine, 81, pp. 167-175, 2017.
16. [16] U. Jamil, S. Khalid, M. U. Akram, A. Ahmad, S. Jabbar, “Melanocytic and nevus lesion detection from diseased dermoscopic images using fuzzy and wavelet techniques”, vol. 22, no. 5, pp. 1577-1593, 2018.
17. [17] V. K. Shrivastava, N. D. Londhe, R. S. Sonawane, J. S. Suri, “Exploring the color feature power for psoriasis risk stratification and classification: A data mining paradigm”, Computers in Biology and Medicine, 65, pp. 54-68, 2015.
18. [18] V. K. Shrivastava, N. D. Londhe, R. S. Sonawane, J. S. Suri, “Reliable and accurate psoriasis disease classification in dermatology images using comprehensive feature space in machine learning paradigm”, Expert Systems with Applications, 42, pp. 6184-6195, 2015.
19. [19] R. Sumithra, M. Suhil, D. S. Guru, “Segmentation and classification of skin lesions for disease diagnosis”. Procedia Computer Science, 45, pp. 76-85, 2015.
20. [20] K. Bunte, M. Biehl, M. F. Jonkman, N. Petkov, “Learning effective color features for content based image retrieval in dermatology”, Pattern Recognition, 44, pp. 1892-1902, 2011.
21. [21] M. A. AI Aboud and P. K. Nigam. “Wart (Plantar, Verruca Vulgaris, Verrucae)” [Updated 2017 Nov 27]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2017. [Available from: https://www.ncbi.nlm.nih.gov/books/NBK431047/]
22. [22] S. C. Bruggink, J. Gussekloo, M. N. de Koning, M. C. Feltkamp, J. N. Bavinck, W. G. Quint, W. J. Assendelft and J. A. Eekhof, “HPV type in plantar warts influences natural course and treatment response: secondary analysis of a randomised controlled trial”, Journal of Clinical Virology, 57, pp. 227-232, 2013.
23. [23] S. C. Bruggink, J. Gussekloo, M. Y. Berger, K. Zaaijer, W. J. Assendelft, M. W. de Waal, J. N. Bavinck, B. W. Koes and J. A. Eekhof, “Cryotherapy with liquid nitrogen versus topical salicylic acid application for cutaneous warts in primary care: randomized controlled trial”, Canadian Medical Association Journal, 182, pp. 1624-1630, 2010.
24. [24] C. K. Varnavides, C. A. Henderson and W. J. Cunliffe, “Intralesional interferon: ineffective in common viral warts”, Journal of Dermatological Treatment, 8, pp. 169-172, 1997.
25. [25] M. H. Bunney, M. W. Nolan and D. A. Williams, “An assessment of methods of treating viral warts by comparative treatment trials based on a standard design”. The British Journal of Dermatology, 94, pp. 667-679, 1976.
26. [26] J. Berth-Jones and P. E. Hutchinson, “Modern treatment of warts: cure rates at 3 and 6 months”, The British Journal of Dermatology, 127, pp. 262-265, 1992.
27. [27] P. Larsen and G. Laurberg, “Cryotherapy of viral warts”, Journal of dermatological treatment, 7, pp. 29-31, 1996.
28. [28] A. M. Kuwabara, B. M. Rainer, H. Basdag, B. A. Cohen, “Children with warts: a retrospective study in an outpatient setting”, Pediatric dermatology, vol.32, no.5, pp. 679-683, 2015.
29. [29] I. Ahmed, S. Agarwal, A. Ilchyshyn, S. Charles-Holmes, J. Berth-Jones “Liquid nitrogen cryotherapy of common warts: cryo‐spray vs. cotton wool bud”, British Journal of Dermatology, vol. 144, no. 5, pp. 1006-1009,2001.
30. [30] P. Kirby, Human papillomavirus infection. In: Moschella SL, Hurley HJ, editors.Dermatology. 3rd ed. Philadelphia (PA): WB Saunders, 1992.
31. [31] E. G. Kuflik, “Cryosurgery updated”, Journal of the American Academy of Dermatology, vol.31, pp. 925-944, 1994.
32. [32] Z. Erbağcı, N. Kırtak and O. Özgöztaşı, “The effect of cryotherapy in verruca vulgaris and plantaris (article in Turkish with an abstract in English)”, Turkiye Klinikleri Journal of Dermatology, vol. 6, pp. 18-20,1996.
33. [33] E. Alpsoy, E. Yilmaz, L. Çetin and E. Başaran, “Effectiveness of cryotherapy in different types of verrucae (article in Turkish with an abstract in English)”, Turkiye Klinikleri Journal of Dermatology, vol. 4, pp. 160-162, 1994.
34. [34] M. Özpoyraz, S. Uzun, M. A. Acar and H. R. Memişoğlu, “Verrukalarda kriyoterapi”, XIV. Ulusal Dermatoloji Kongresi (XIV. National Congress of Dermatology), 1992, pp. 1-4.
35. [35] Z. Erbağcı, N. Kırtak and O. Özgöztaşı, “The effect of cryotherapy in verruca vulgaris and plantaris (article in Turkish with an abstract in English)”, Turkiye Klinikleri Journal of Dermatology, vol. 6, pp. 18-20,1996.
36. [36] E. Alpsoy, E. Yilmaz, L. Çetin and E. Başaran, “Effectiveness of cryotherapy in different types of verrucae (article in Turkish with an abstract in English)”, Turkiye Klinikleri Journal of Dermatology, vol. 4, pp. 160-162, 1994.
37. [37] M. Özpoyraz, S. Uzun, M. A. Acar and H. R. Memişoğlu, “Verrukalarda kriyoterapi”, XIV. Ulusal Dermatoloji Kongresi (XIV. National Congress of Dermatology), 1992, pp. 1-4.
38. [38] E. Göçmen, O. Derse, “Forecasting of Electricity Generation Shares by Fossil Fuels Using Artificial Neural Network and Regression Analysis in Turkey”. International Scientific and Vocational Studies Journal, vol.2, no.2, pp.20-30, 2018.