A Point-of-Care, Real-Time Artificial Intelligence System to Support Clinician Diagnosis of a Wide Range of Skin Diseases

  • Author Footnotes
    5 These authors contributed equally to this work.
    Brittany Dulmage
    Footnotes
    5 These authors contributed equally to this work.
    Affiliations
    Department of Dermatology, The Ohio State University Wexner Medical Center, Columbus, Ohio, USA
    Search for articles by this author
  • Author Footnotes
    5 These authors contributed equally to this work.
    Kyle Tegtmeyer
    Footnotes
    5 These authors contributed equally to this work.
    Affiliations
    Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
    Search for articles by this author
  • Michael Z. Zhang
    Affiliations
    Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA

    Vanderbilt University School of Medicine, Nashville, Tennessee, USA
    Search for articles by this author
  • Maria Colavincenzo
    Affiliations
    Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
    Search for articles by this author
  • Shuai Xu
    Correspondence
    Correspondence: Shuai Xu, 676 N. St. Clair Street, Suite 1600, Chicago, Illinois 60611, USA.
    Affiliations
    Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA

    Querrey Institute for Bioelectronics, Northwestern University, Evanston, Illinois, USA
    Search for articles by this author
  • Author Footnotes
    5 These authors contributed equally to this work.
Published:October 13, 2020DOI:https://doi.org/10.1016/j.jid.2020.08.027
      Dermatological diagnosis remains challenging for nonspecialists because the morphologies of primary skin lesions widely vary from patient to patient. Although previous studies have used artificial intelligence (AI) to classify lesions as benign or malignant, there have not been extensive studies examining the use of AI on identifying and categorizing a primary skin lesion's morphology. In this study, we evaluate the performance of a standalone AI tool to correctly categorize a skin lesion's morphology from a test bank of images. To provide a marker of performance, we evaluate the accuracy of primary care physicians to categorize skin lesion morphology in the same test bank of images without any aids and then with the aid of a simple visual guide. The AI system achieved an accuracy of 68% in determining the single most likely morphology from the test image bank. When the AI’s top prediction was broadened to its top three most likely predictions, accuracy improved to 80%. In comparison, the diagnostic accuracy of primary care physicians was 36% without any aids and 68% with the visual guide ( P < 0.001). The AI was subsequently tested on an additional set of 222 heterogeneous images of varying Fitzpatrick skin types and achieved an overall accuracy of 70% in the Fitzpatrick I–III skin type group and 68% in the Fitzpatrick IV–VI skin type group ( P = 0.79). An AI is a powerful tool to assist physicians in the diagnosis of skin lesions while still requiring the user to critically consider other possible diagnoses.

      Abbreviation:

      AI ( artificial intelligence)
      To read this article in full you will need to make a payment

      References

        • Adamson A.S.
        • Smith A.
        Machine learning and health care disparities in dermatology.
        JAMA Dermatol. 2018; 154: 1247-1248
        • Ahiarah A.
        • Fox C.
        • Servoss T.
        Brief intervention to improve diagnosis and treatment knowledge of skin disorders by family medicine residents.
        Fam Med. 2007; 39: 720-723
        • Alther M.
        • Reddy C.K.
        Clinical decision support systems.
        in: Reddy C.K. Aggarwal C.C. Healthcare data analytics. CRC Press, Boca Raton, FL2015: 625-656
        • American Academy of Dermatology
        Position statement on augmented intelligence (AuI).
        (accessed 3 February 2020)
        • Arnold R.
        Role of pilot lack of manual control proficiency in air transport aircraft accidents.
        Procedia Manuf. 2015; 3: 3142-3146
        • Breitbart E.W.
        • Choudhury K.
        • Andersen A.D.
        • Bunde H.
        • Breitbart M.
        • Sideri A.M.
        • et al.
        Improved patient satisfaction and diagnostic accuracy in skin diseases with a Visual Clinical Decision Support System - a feasibility study with general practitioners.
        PLoS One. 2020; 15e0235410
        • Bukhari I.
        • AlAkloby O.
        Evaluation of diagnostic skills of interns electively rotating at the dermatology department of King Fahad Hospital of the University in Alkhobar, Saudi Arabia.
        Internet J Dermatol. 2006; 5
        • Burgin S.
        Guidebook to dermatologic diagnosis.
        1st ed. McGraw-Hill Education, New York, NY2019
        • Char D.S.
        • Shah N.H.
        • Magnus D.
        Implementing machine learning in health care - addressing ethical challenges.
        N Engl J Med. 2018; 378: 981-983
        • Esteva A.
        • Kuprel B.
        • Novoa R.A.
        • Ko J.
        • Swetter S.M.
        • Blau H.M.
        • et al.
        Dermatologist-level classification of skin cancer with deep neural networks.
        Nature. 2017; 542: 115-118
        • Huang G.
        • Liu Z.
        • Pleiss G.
        • Van Der Maaten L.
        • Weinberger K.
        Convolutional networks with dense connectivity [e-pub ahead of print].
        IEEE Trans Pattern Anal Mach Intell. 2019; (accessed 3 January 2020)https://doi.org/10.1109/TPAMI.2019.2918284
        • Jones A.
        • Walling H.
        Retiform purpura in plaques: a morphological approach to diagnosis.
        Clin Exp Dermatol. 2007; 32: 596-602
        • Lim H.W.
        • Collins S.A.B.
        • Resneck Jr., J.S.
        • Bolognia J.L.
        • Hodge J.A.
        • Rohrer T.A.
        • et al.
        The burden of skin disease in the United States.
        J Am Acad Dermatol. 2017; 76: 958-972.e2
        • Marka A.
        • Carter J.B.
        • Toto E.
        • Hassanpour S.
        Automated detection of nonmelanoma skin cancer using digital images: a systematic review.
        BMC Med Imaging. 2019; 19: 21
        • McCleskey P.E.
        • Gilson R.T.
        • DeVillez R.L.
        Medical student core curriculum in dermatology survey.
        J Am Acad Dermatol. 2009; 61: 30-35.e4
        • Monheit G.
        • Cognetta A.B.
        • Ferris L.
        • Rabinovitz H.
        • Gross K.
        • Martini M.
        • et al.
        The performance of MelaFind: a prospective multicenter study.
        Arch Dermatol. 2011; 147: 188-194
        • Papier A.
        Decision support in dermatology and medicine: history and recent developments.
        Semin Cutan Med Surg. 2012; 31: 153-159
        • Rajkomar A.
        • Hardt M.
        • Howell M.D.
        • Corrado G.
        • Chin M.H.
        Ensuring fairness in machine learning to advance health equity.
        Ann Intern Med. 2018; 169: 866-872
        • Rajpara S.M.
        • Botello A.P.
        • Townend J.
        • Ormerod A.D.
        Systematic review of dermoscopy and digital dermoscopy/artificial intelligence for the diagnosis of melanoma.
        Br J Dermatol. 2009; 161: 591-604
        • Santistevan J.
        • Long B.
        • Koyfman A.
        Rash decisions: an approach to dangerous rashes based on morphology.
        J Emerg Med. 2017; 52: 457-471
        • Sherertz E.F.
        Learning dermatology on a dermatology elective.
        Int J Dermatol. 1990; 29: 345-348
      1. Zoph B, Vasudevan V, Shlens J, Le Q. Learning transferable architectures for scalable image recognition. Paper presented at: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 18–23 June 2018; Salt Lake City, UT