AAD
Author: Willhemenia Loremipsum Special session: Lorem ipsum dolor sit amet Category: Lorem ipsum View all results

Please login or create an account. If you do not have access to this content, you will be shown a 30 second preview and licensing options.

  • Presentation

DataDerm: The First Decade and Goals for the Next

Description

The presentation discusses the use of artificial intelligence (AI) in identifying Generalized Psoriasis Pustulosa (GPP) patients more effectively, based on the first decade of DataDerm's data and looking forward to future advancements. The speaker emphasizes that an alarming number of GPP patients experience misdiagnosis or prolonged delays before receiving appropriate treatment, indicating significant gaps in current diagnostic practices. By leveraging extensive real-world data, including a large dataset from DataDerm, the team aims to create 'digital phenotypes' or fingerprints of known GPP patients to identify those at high risk of undiagnosed GPP. Utilizing AI allows for scalable analysis of patient histories, helping to distinguish between those with and without GPP. The mechanism, termed 'Patient Finder,' highlights undiagnosed patients by comparing their health data against established fingerprints. Performance metrics indicate that the AI tools show promising results, identifying high-risk individuals effectively. The speaker asserts the importance of ongoing research to translate these insights into clinician-focused education and clinical workflows, enabling proactive patient management. Future efforts will explore understanding the specific factors that contribute to the identification process, enhancing the practical utility of these AI applications in everyday clinical settings.

View more

Conclusions

  • AI and real-world data analysis can identify patients at high risk for undiagnosed generalized pustular psoriasis (GPP).
  • More than half of GPP patients experience misdiagnosis and face significant delays in reaching the proper diagnosis.
  • A distinction between GPP patients with a prior history of psoriasis and those without is crucial for improving the identification process.
  • The Patient Finder tool demonstrated strong analytical performance with AUC values of 0.79 and 0.82 for psoriasis positive and negative groups respectively.
  • Enrichment metrics showed that the Patient Finder can identify GPP patients effectively, significantly increasing the likelihood of encountering them in a screened population.
  • The findings also hold up across various demographic factors, such as age and geographic location.
  • Future research will focus on explaining the specific factors that contribute to the identification of GPP patients using Patient Finder.
  • Ly K, Beck KM, Smith MP, Thibodeaux Q, Bhutani T. Diagnosis and screening of patients with generalized pustular psoriasis. Psoriasis (Auckl). 2019 Jun 20;9:37-42. doi: 10.2147/PTT.S181808.
  • Ito T, Takahashi H, Kawada A, Iizuka H, Nakagawa H, Japanese Society For Psoriasis Research. Epidemiological survey from 2009 to 2012 of psoriatic patients in Japanese Society for Psoriasis Research. J. Dermatol. 2017 Nov 8;435(3):251-333. doi: 10.1111/1346-8138.14105.
  • Reisner DV, Johnsson FD, Kotowsky N, Brunette S, Valdecantos W, Eyerich K. Impact of Generalized Pustular Psoriasis from the Perspective of People Living with the Condition: Results of an Online Survey. Am J Clin Dermatol. 2022 Jan;23(Suppl1):65-71. doi: 10.1007/s40257-021-00663-y.
  • Löfvendahl S, Norlin JM, Schmitt-Egenolf M. Prevalence and incidence of generalized pustular psoriasis in Sweden: a population-based register study. Br J Dermatol. 2022 Jun 1;186(6):970-976. doi: 10.1111/bjd.20966.
  • Prinz JC, Choon SE, Griffiths CEM, Merola JF, Morita A, Ashcroft DM, et al. Prevalence, comorbidities and mortality of generalized pustular psoriasis: A literature review. J Eur Acad Dermatol Venereol. 2022 Nov 4;37(2):256-273. doi: 10.1111/jdv.18720.