Loading...

quantusSKIN

Analysis and Classification of dermoscopic images for the determination os Skin lesion malignancy risk.

Non-invasive: quantusSKIN is a non-invasive test to predict the risk of malignancy of different skin lesions from a photograph or dermatoscopic image.

Fast: quantusSKIN generates accurate results in just a few minutes.

QuantusSKIN reliability table

  Sensitivity Specificity PPV * NPV *
quantusSKIN 89.6% 85.2% 52.6% 97,8%
* PPV and NPV (Positive Predictive Value and Negative Predictive Value)

WHY DOES quantusSKIN WORK?

An automated support tool is defined as one that requires minimal or no physician intervention to obtain a result. In recent years, research has focused on automatic algorithms to improve current clinical diagnosis based on images. The rise of Artificial Intelligence techniques, and especially Deep Learning techniques, has increased the number of studies using these types of algorithms in diagnostic dermatology.

Several recently published studies provide evidence that detection of malignant dermatologic lesions using trained Deep Learning models can achieve high accuracy in diverse populations and provide quantitative comparisons of how model performance may vary across datasets consisting of glaucoma of different disease severity and ethnicity.

quantusSKIN is presented as a novel Artificial Intelligence method, based on state-of-the-art Deep Learning. Several studies have proven the correlation between the quantitative analysis method proposed by quantusSKIN. The technology is based on performing a quantitative analysis of the texture of the skin Nevus image obtained by means of a smartphone, reflex camera or dermatoscope. This analysis makes it possible to identify patterns associated with specific pathologies and to determine the risk of malignancy of the lesion. According to the literature, the different tests and tools used by the dermatologist give an individual sensitivity of 75-84% (see reference 9 ); while quantusSKIN has obtained a sensitivity of 85.6% in its tests (see reference 16).

WHEN TO USE quantusSKIN?

quantusSKIN has been designed with a clear focus on the general population, and is intended to be a tool for the detection of malignant skin lesions (melanoma, basal cell carcinoma or squamous cell carcinoma), being of great help in the screening of patients with risk factors and prioritization of waiting lists.

REFERENCES:

  1. [1] U. Leiter, T. Eigentler, and C. Garbe, "Epidemiology of Skin Cancer BT - Sunlight, Vitamin D and Skin Cancer," in Advances in experimental medicine and biology, vol. 810, J. Reichrath, Ed. Springer New York, 2014, pp. 120–140.
  2. [2] C. Garbe and U. Leiter, "Melanoma epidemiology and trends," Clin. Dermatol., vol. 27, no. 1, pp. 3–9, Jan. 2009, doi: 10.1016/j.clindermatol.2008.09.001.
  3. [3] G. P. Guy, S. R. Machlin, D. U. Ekwueme, and K. R. Yabro , "Prevalence and costs of skin cancer treatment in the U.S., 2002-2006 and 2007- 2011," Am. J. Prev. Med., vol. 48, no. 2, pp. 183–187, Feb. 2015, doi: 10.1016/j.amepre.2014.08.036.
  4. [4] H. W. Rogers, M. A. Weinstock, S. R. Feldman, and B. M. Coldiron, "Incidence estimate of nonmelanoma skin cancer (kera nocyte carcinomas) in the US population, 2012," JAMA Dermatology, vol. 151, no. 10, pp. 1081–1086, Oct. 2015, doi: 10.1001/jamadermatol.2015.1187. [5] R. L. Siegel, K. D. Miller, and A. Jemal, "Cancer statistics, 2017," CA. Cancer J. Clin., vol. 67, no. 1, pp. 7–30, Jan. 2017, doi: 10.3322/caac.21387.
  5. [6] "Melanoma Warning Signs and Images - The Skin Cancer Foundation." https://www.skincancer.org/skin-cancer-information/melanoma/melanoma-warning-signs-and-images/ (accessed Sep. 29, 2020)
  6. [7] H. A. Haenssle et al., “Associa on of pa ent risk factors and frequency of nevus-associated cutaneous melanomas,” JAMA Dermatology, vol. 152, no. 3, pp. 291–298, Mar. 2016, doi: 10.1001/jamadermatol.2015.3775.
  7. [8] P. Tschandl and P. Doz Philipp Tschandl, “Sequential digital dermatoscopic imaging of patients with multiple atypical nevi,” Rev. | Dermatol Pr. Concept, vol. 8, no. 3, pp. 231–237, 2018, doi: 10.5826/dpc.0803a16.
  8. [9] M. E. Vestergaard, P. Macaskill, P. E. Holt, and S. W. Menzies, “Dermoscopy compared with naked eye examination for the diagnosis of primary melanoma: A meta-analysis of studies performed in a clinical setting,” Br. J. Dermatol., vol. 159, no. 3, pp. 669–676, Sep. 2008, doi: 10.1111/j.1365-2133.2008.08713.x.
  9. [10] H. Ki ler, H. Pehamberger, K. Wol , and M. Binder, “Diagnostic accuracy of dermoscopy,” Lancet Oncology, vol. 3, no. 3. Lancet Publishing Group, pp. 159–165, Mar. 01, 2002, doi: 10.1016/S1470-2045(02)00679-4.
  10. [11] A. C. Geller, S. M. Swe er, K. Brooks, M. F. Demierre, and A. L. Yaroch, “Screening, early detection, and trends for melanoma: Current status (2000-2006) and future directions,” Journal of the American Academy of Dermatology, vol. 57, no. 4. Mosby, pp. 555–572, Oct. 01, 2007, doi: 10.1016/j.jaad.2007.06.032.
  11. [12] A. Rosenberg and J. H. Meyerle, “Total-body photography in skin cancer screening: The clinical use of standardized imaging,” Cu s, vol. 99, no. 5, pp. 312–316, May 2017, Accessed: Sep. 29, 2020. [Online]. Available: https://europepmc.org/ar cle/med/28632800
  12. [13] N. C. F. Codella et al., “Skin lesion analysis toward melanoma detection: A challenge at the 2017 International symposium on biomedical imaging (ISBI), hosted by the international skin imaging collaboration (ISIC),” in Proceedings - International Symposium on Biomedical Imaging, May 2018, vol. 2018-April, pp. 168–172, doi: 10.1109/ISBI.2018.8363547.
  13. [14] P. Tschandl, C. Rosendahl, and H. Ki ler, “Data descriptor: The HAM10000 dataset, a large collec on of mul -source dermatoscopic images of common pigmented skin lesions,” Sci. Data, vol. 5, no. 1, pp. 1–9, Aug. 2018, doi: 10.1038/sdata.2018.161
  14. [15] M. Combalia et al., “BCN20000: Dermoscopic Lesions in the Wild,” Aug. 2019, Accessed: Jul. 01, 2020. [Online]. Available: http://arxiv.org/abs/1908
  15. [16]Coronado-Gu érrez, D., López, C., & Burgos-Ar zzu, X. (2021). Skin cancer high-risk patient screening from dermoscopic images via Artificial Intelligence: an online study. doi: 10.1101/2021.02.04.21251132