Skin Health Applications: Blessing or Misdiagnosis?
Anum Wasim1, Madiha Hassan Rizvi1, Ayisha Farooq Khan1
15th year Medical Student, Dow Medical College, Karachi, Pakistan
Corresponding Author: anum.wasim5@gmail.com
There has been a global rise in skin cancer over the last few years with the rising diagnosis of 2–3 million non-melanoma and 132,000 melanoma skin cancer cases each year1. This growing prevalence is not only attributed to the increasing ultraviolet radiation by ozone depletion but also by other major predisposing factors centred to an individual’s own responsibility controls like recreational sun exposure and sunburn1. However despite the escalating figures, low skin screening rates and awareness levels among the general population are highly disconcerting2.
An estimated five billion worldwide mobile phone subscribers have permitted the creation of custom built skin health applications on personal mobile phones for self-diagnosis and screening of different skin health conditions3,4. The easily accessible self-diagnostic/screening skin health applications such as Skin Scan, SpotCheck and Skin of Mine has allowed self-care dermatology to remote regions5.
In an eHealth review by Tyagi et al, skin 46 skin health applications were identified. These applications deploy one or all of the five basic principles, namely mobile teledermoscopy, algorithm analysis, video demonstration, teaching and education6. Telemedicine based skin health applications sanction utilization of a dermoscope attached to a smartphone camera to capture high resolution images of skin lesions to be sent to a team of dermatologists analysing and diagnosing the sent images within 24 hours6. However unlike telemedicine applications, algorithm based applications involves complete self-evaluation without a clinician7. While algorithms vary from program to program, the majority of algorithms use a list of low, medium, and high-risk skin lesions being diagnosed to date, along with a world map of where each case happened to help these applications define any advancement of the lesion to an uncharacteristic growth8.
Other skin health applications focus on providing informational material to the user to perform self-skin exams (SSE) in distinguishing the asymmetry, border irregularity, colour, diameter, and evolution of the skin lesions (ABCDE criteria)6. Providing diagrams and instructional text, teaching modules with quizzes to progress through tutorials or inclusion of videos about skin safety advice and interventions for regular performance of SSE are the techniques of these applications9. An archive log of each lesion’s growth along with a complete list of doctors tailored to personal locations for future appointments are some additional features of majority skin health applications8.
The first major study to highlight the efficacy of these applications by Wolf et al, analysed four skin health applications which are being utilized for melanoma tracking and risk assessment10. Three of the aforementioned applications which employed automated algorithms to assess melanoma were exposed as inaccurate in misdiagnosing more than 30% of melanomas as benign. The fourth and most sensitive application, was based on the principle of store and forward teledermatology wherein images of the lesion were sent to a board certified dermatologist for evaluation within a time frame of 24 hours10. While the physician based method proved to be superior in sensitivity, it was also the most expensive, costing $5 per use. The algorithm based applications had a price ranging from free to $4.99 for unlimited use10. Due to financial reasons, some people may use the cheaper applications, though the use of these potentially inaccurate applications may delay appropriate healthcare6.
While these skin health applications are accessible to the general public, further research and improvement is required to address inaccuracies. Physicians should also be educated on the different applications that are available to their patients, allowing them to counsel their patients on appropriate usage of these technologies. Whilst there are significant issues with the current generation of skin health applications, through further research and development, these applications may play a valuable role in the screening of suspicious skin lesions in the future.
Funding Sources
None Declared
Conflict of Interest
No conflict of interest
Acknowledgements
We would like to extend our sincere thanks to Abdul Nafey Kazi for his untiring support throughout the writing of this manuscript.
References
1. Who.int [internet]. WHO Skin cancers; [cited at 2013 August 29] Available from: http://www.who.int/uv/faq/skincancer/en/index1.html.
2. Lakhani NA, Shaw KM, Thompson T. Prevalence and predictors of total-body skin examination among US adults: 2005 National Health Interview Survey. Journal of the American Academy of Dermatology 2011;65(3):645–8. 
3. Awan Z. Do mobile phones cause brain tumors. J Pak Med Stud 2012;2(20):64–7.
4. Payne KFB, Wharrad H, Watts K. Smartphone and medical related App use among medical students and junior doctors in the United Kingdom (UK): a regional survey. BMC Med Inform Decis Mak 2012 Oct; 12: 121. PubMed PMID: 23110712; PubMed Central PMCID: PMC3504572
5. Farshidi D, Craft N, Ochoa MT. Mobile teledermatology: As doctors and patients are increasingly mobile, technology keeps us connected. Skinmed 2011 Jul–Aug; 9(4):231–8.
6. Abhilasha Tyagi, Kimberly Miller, and Myles Cockburn. e-Health Tools for Targeting and Improving Melanoma Screening: A Review. Journal of Skin Cancer 2012; 2012: 8. . PubMed PMID: 23304515; PubMed Central PMCID: PMC3530856
7. Iyatomi H, Oka H, Celebi ME, Hashimoto M, Hagiwara M, Tanaka M, Ogawa K. An improved Internet-based melanoma screening system with dermatologist-like tumor area extraction algorithm. Computerized Medical Imaging and Graphics 2008;32(7):566–79.
8. Reistad-Long S. Diagnosing skin cancer via iphone- the apps to know. The Atlantic [internet]. Cited 2012 Sep 13. Available from: http://www.theatlantic.com/health/archive/2012/09/diagnosing-skin-cancer-via-iphone-the-apps-to-know/262325/. [Accessed at 25/August 2013]
9. Skin Cancer Detection. Available from: http://azcc.arizona.edu/sci/about/videos/detection. [Accessed at 25/August 2013]
10. Wolf JA, Moreau J, Akilov O, et al. Diagnostic Inaccuracy of Smartphone Applications for Melanoma Detection. JAMA Dermatol 2013 April;149(4):422–6.. PubMed PMID: 23325302
