Dorsal expansion-related manual disability in verrucae vulgaris patients treated with cryotherapy: A longitudinal study via big data

Abstract

Background: Common warts frequently occur over the hands and may overlie critical structures, including the extensor expansion of the digits, for which cryotherapy can lead to manual disability. Aims and Objectives: We aim to implement big data to infer knowledge on the interest of web users regarding the dorsal expansion-related manual disability in verrucae vulgaris patients managed with cryotherapy. Materials and Methods: We mapped the interest of users of the surface web in connection with five topics: cryosurgery, extensor digitorum muscle, human papillomavirus, manual disability, and verruca vulgaris. We retrieved longitudinal data [2004-2019] concerning the spatio-temporal variations of interest in these topics, using Google Trends. We are also reporting a case of interest while analyzing it using the Bradford Hill criteria. Results: Sixty-four nations contributed to the spatial (geographic) map, including ten countries from the Middle East and the north of Africa (15.63%). There was high temporal variability concerning cryosurgery 29.61 (+/- 0.94), extensor digitorum muscle 64.43 (+/- 0.86), human papillomavirus 0.01 (+/- 0.01), manual disability 2.89 (+/- 0.13), and verruca vulgaris 9.39 (+/- 0.20). Conjoint inference, via Post-Hoc testing and neural networks [machine learning], assigned the highest synaptic weight [effect size] to “cryosurgery” and “extensor digitorum muscle”, thus indicating that these topics are of prime interest for web users, including anatomists and dermatologists. Conclusions: We are conveying two messages: 1) Dermatologists are encouraged to collaborate with data scientists to realize the importance of big data for evidence-based dermatology. 2) Physicians should exercise extra caution in managing patients with verrucae vulgaris using cryotherapy.

Keywords

• cryotherapy
• disability studies
• extensor expansion
• human papillomavirus
• machine learning