About

Dr. Noah Goldfarb is an Associate Professor at the University of Minnesota, serving in the Departments of Medicine and Dermatology. He graduated from SUNY Stony Brook School of Medicine in 2009 and completed a combined residency in internal medicine and dermatology in 2014. His clinical interests include autoimmune skin diseases, hidradenitis suppurativa (HS), and complex medical dermatology. He runs an HS clinic and a combined rheum-derm clinic at the University of Minnesota site. Dr. Goldfarb is the director of inpatient dermatology consultation for the dermatology service at the Minneapolis VA Health Care System and continues to work as a hospitalist at the VA, engaging with medical students and residents. He is passionate about education, serving as the medical student director at the VA, the dermatology interest group advisor at the University of Minnesota, and the Residency Program Co-Director for the Combined Internal Medicine/Dermatology program.

Research topics

• Medicine
• Political Science
• Computer Science
• Medical education
• Family medicine
• Dermatology
• Artificial Intelligence
• Internal medicine
• Demography
• Medical physics
• Psychology
• Psychiatry
• Environmental health
• Pathology
• Social psychology
• Linguistics
• Intensive care medicine

Selected publications

• Implications of underrepresented ancestral populations in dermatologic genomic databases

  Journal of the American Academy of Dermatology · 2026-01-21

  articleSenior author
  PublisherDOI

• Harmonizing Hidradenitis Suppurativa: an Updated Glossary of Lesions with Clinical Presentations

  Current Dermatology Reports · 2026-03-21

  article
  PublisherDOI

• A Genome-Wide Association Study of Hidradenitis Suppurativa from the VA’s Million Veteran Program

  Dermatology · 2026-02-03

  articleOpen accessSenior author

  INTRODUCTION: Data from family and twin studies as well as prior genome-wide association meta-analyses suggest that hidradenitis suppurativa (HS) has a hereditary component. This GWAS identifies genetic variants associated with HS. METHODS: Individuals with a diagnosis of HS (defined as at least one instance of ICD9 705.83 or ICD10 L73.2) were identified within the Million Veteran Program. Multi-population and population-specific (African, European, and Hispanic ancestries) case-control genome-wide association studies (GWAS) were performed. Lead single-nucleotide polymorphisms were investigated in external resources providing phenome-wide associations (PheWAS), including UK Biobank, HugeAMP, and FinnGen. Demographic and clinical data for cases and controls were taken from the Corporate Data Warehouse and differences between the cases and control group were analyzed. RESULTS: A total of 4,959 participants with HS were identified among 597,819 MVP participants. The multi-population GWAS identified two significant (p < 5 × 10-8) loci associated with HS, including a novel HS-related variant on chromosome 6 near HLA-DRB1 (lead variant rs679242), and confirmed a previously identified locus on chromosome 17 near SOX9 (rs55811634). The following previously identified loci associated with HS achieved suggestive evidence for association (p < 1 x 10-3): rs17090189 (13q22.1; near KLF5), rs10816701 (9q31.3; near KLF4), rs121908120 (2q35; WNT10A), and rs17103088 (14q24.3; near TMED10). CONCLUSION: The analysis of the MVP resource for HS identified a novel signal on chromosome 6 near HLA-DRB1 and identified significant evidence and suggestive evidence for several previously reported signals for HS.

  PublisherOA PDFDOI

• ChatGPT, Gemini, and Claude in clinical and dermoscopic image analysis of basal cell carcinoma and its common mimickers: A comparative performance analysis

  JID Innovations · 2026-02-28 · 1 citations

  articleOpen access

  Basal cell carcinoma (BCC) is the most common skin cancer. Off-the-shelf multimodal large language models are widely accessible, yet their performance for BCC remains unclear. The aim of this study was to assess BCC detection (BCC vs non-BCC) and BCC subtype classification from clinical and dermoscopic images using 3 web-based large language models (ChatGPT-5, Gemini 2.5 Flash, Claude Sonnet 4). We evaluated 772 images: 402 from 290 histopathology-confirmed BCCs (290 clinical, 112 dermoscopic) and 370 from an independent BCC-mimicker cohort (250 clinical, 120 dermoscopic). Standardized prompts were used. Primary outcome was BCC detection accuracy; secondary outcomes were subtype-classification accuracy and performance by lesion features. For clinical images, ChatGPT-5 achieved the highest detection accuracy (75%), followed by Claude (64.3%) and Gemini (50.7%). For dermoscopy, Claude performed best (69.8%), compared with ChatGPT-5 (55.2%) and Gemini (50.9%). Accuracy was lower in crusted and flat lesions and higher in exophytic lesions; pigmentation effects were model dependent. Subtype-classification accuracy was modest across models. Images were primarily from European centers with limited skin-type diversity; several subgroups were small. Current web-based large language models are not clinically suitable for BCC detection or subtyping. Dermatology-specific training, transparent reporting, and rigorous prospective validation are required before any clinical use.

  PublisherDOI

• Trends in Surgical Management of Hidradenitis Suppurativa.

  UNC Libraries · 2025-11-28

  articleOpen access

  Hidradenitis suppurativa (HS) is a chronic inflammatory disorder with proclivity for intertriginous anatomic sites. Management of HS is unique, with complementary roles for surgery combined with pharmacologic treatments. Several surgical options exist for HS with<br />reported high rates of satisfaction among patients. Surgical options include lesional and regional deroofing procedures and excisions. Notably, there is a dearth of research regarding the physician specialties performing these procedures, as well as how these have changed over time. With the evolution of treatment strategies, such as surgical deroofing, and the Federal and Drug Administration&ndash;approval of adalimumab and secukinumab&mdash;biologics commonly used within the field of dermatology&mdash;we hypothesized that dermatologists and dermatology surgeons may be assuming greater responsibility for performing these procedures.

  PublisherDOI

• 0171 Accuracy of administrative codes for hidradenitis suppurativa in the veterans affairs healthcare system

  Journal of Investigative Dermatology · 2025-07-21

  articleOpen accessSenior author
  PublisherOA PDFDOI

• Delphi Panel for the Development of Diagnostic Criteria for Metastatic Cutaneous Crohn Disease

  JAMA Dermatology · 2025-05-07 · 8 citations

  letter

  Importance: Metastatic cutaneous Crohn disease (MCD) is a rare condition for which there are currently no published diagnostic criteria. Formal MCD diagnostic criteria will expand clinical care and research by enabling future diagnostic code validation, crystallizing a more uniform disease entity for the purposes of translational research, and allowing the development of more formalized outcome measures aimed at treatment response. Objective: To define a set of criteria for the diagnosis of MCD. Evidence Review: In this consensus statement, a panel of experts on MCD and related conditions was assembled. Their opinions were gathered regarding the clinical and histologic findings that are sufficient, necessary, and highly suggestive of MCD through serial survey. Subsequent surveys were continued until consensus was achieved. Sufficient was defined as a feature that, if noted, would enable diagnosis of MCD without any other signs, symptoms, or testing. Necessary was defined as a feature that must be present to establish a diagnosis of MCD. Highly suggestive referred to any other features that were indicative of MCD but did not qualify as sufficient or necessary to diagnose MCD. Findings: Of 24 experts invited, 19 agreed to participate, with a 79% to 100% response rate across 5 rounds conducted. Five patients were also invited to participate, but none opted to participate, citing lack of clinical expertise and time concerns. The experts were all from the US; 5 (26%) have practiced medicine for 0 to 5 years, 5 (26%) practiced for 6 to 10 years, 6 (32%) practiced for 11 to 15 years, and 3 (16%) practiced for more than 20 years. A total of 18 experts (95%) worked in an academic practice setting. There was consensus agreement that MCD could present as oral disease, genital disease, or, rarely, in another location. Consensus was achieved on 2 major and 5 minor criteria for genital/other subtypes of MCD and 5 minor criteria for oral MCD. There was consensus that, while there are some pathological findings that are highly suggestive of MCD, histologic criteria are not necessary for diagnosis. Conclusions and Relevance: A panel of experts reached consensus on the diagnostic criteria for MCD described in this study. These criteria include major and minor clinical criteria, along with histopathological criteria for the diagnosis of genital and oral MCD. This consensus is a crucial step in the diagnosis of and future treatment development for MCD, a rare and clinically understudied condition.

  PublisherDOI

• Shaping the Future of Holistic Hidradenitis Suppurativa Care: The Power of Multimodal Continuing Medical Education

  SKIN The Journal of Cutaneous Medicine · 2025-07-23

  articleOpen access

  Introduction: Hidradenitis suppurativa (HS) profoundly impacts patients’ physical and psychosocial well-being. Combining approved therapies with surgical interventions in clinical practice is critical to effective, long-term management, and all members of the HS care team (advanced practitioners [APs] and physicians) should be well-equipped to provide optimal care for patients. To meet this educational need, CMEsquared® partnered with the HS Foundation to develop a multimodal, educational initiative focused on holistic HS management. Methods: Two 90-minute, in-person CME workshops on HS were delivered at the 2024 Winter Clinical Miami Dermatology Conference® and 2024 Fall Clinical Dermatology Conference® for PAs &amp; NPs, with a video recording endured on Dermsquared.org until July 2025. The workshops were led by a multidisciplinary expert panel that included physicians, a physician assistant, and a patient advocate with lived experience of HS. A variety of learning methods were used to maximize educational impact; these included: 1) didactic lecture on HS diagnosis and guidelines, clinical data for approved/emerging therapies, in-office surgical procedures, wound care and pain management, and patient perspectives, 2) pre-recorded video demonstration of surgical procedures (punch deroofing and deroofing for persistent nodules, plaques and tunnels), and 3) hands-on, simulation workshop on surgical procedures using food models of HS nodules and tunnels in the skin. Aggregate data on clinician perspectives and paired pre-/post-test questions across both live workshops and the endured activity were evaluated. Results: Through the 978 clinician learners (42% APs, 58% physicians), the education will impact 1,698 patients with HS/week and 88,296 patients with HS/year. More than half of learners were early-career clinicians (≤10 years). Over 95% of clinicians agreed that the teaching style was appropriate, and that content was relevant to practice. The education strongly impacted intent to change practice (APs, 88%; physicians, 84%); both groups ranked “treatment,” “diagnosis,” and “use of new medications” among their top 5 intended changes. There were observed improvements in knowledge on therapeutic mechanism of action (relative increase: APs, 40%; physicians, 16%) and competence in therapeutic selection (relative increase: APs, 5%; physicians, 10%). Conclusion: HS education that incorporated didactic instruction, tactile skill-building, and patient insights enhanced understanding of holistic HS management among members of the dermatology care team, especially for clinicians early in their careers. It also stimulated consideration of meaningful practice changes. This multimodal educational method could be applied to other therapeutic areas, where appropriate, to equip clinicians with the knowledge and skills needed to improve patient outcomes.

  PublisherOA PDFDOI

• Histiocytoid Sweet syndrome-like presentation of mature plasmacytoid dendritic cell proliferation

  Dermatology Online Journal · 2025-06-17

  articleOpen accessSenior author

  Mature plasmacytoid dendritic cell proliferation is a condition associated with myeloid neoplasms, most commonly chronic myelomonocytic leukemia. Plasmacytoid dendritic cells can resemble lymphocytes and histiocytes morphologically and immunophenotypically. Mature plasmacytoid dendritic cell proliferation may therefore go unrecognized if the diagnosis is not suspected and appropriate stains for plasmacytoid dendritic cells are not performed. Herein, we present a case of mature plasmacytoid dendritic cell proliferation masquerading clinically and histologically as histiocytoid Sweet syndrome. The patient, who had previously been diagnosed with mature plasmacytoid dendritic cell proliferation that presented as pink, edematous, pruritic papules and plaques, had initially resolved following induction chemotherapy for acute myelomonocytic leukemia. However, he presented later with indurated purpuric plaques on the trunk within weeks of receiving filgrastim for neutropenia. Biopsies demonstrated marked dermal edema, interstitial, superficial, and deep infiltrate with histiocytoid appearing cells concerning for histiocytoid Sweet syndrome. Further work-up demonstrated that the infiltrate was predominantly composed of CD3-, CD4+, CD34-, CD123+, CD56-, CD68-, myeloperoxidase negative mononuclear cells consistent with mature plasmacytoid dendritic cell proliferation. This case demonstrates that MPDCP should be considered in the differential diagnosis of eruptions that clinically and histologically look like histiocytoid Sweet syndrome but stain negatively for myeloperoxidase.

  PublisherOA PDFDOI

• Large language models in evaluating hidradenitis suppurativa from clinical images

  Journal of the European Academy of Dermatology and Venereology · 2025-07-16 · 24 citations

  letterOpen access

  The data that support the findings of this study are available from the corresponding author upon reasonable request.

  PublisherOA PDFDOI

Frequent coauthors

• Afsáneh Alavi

  WinnMed

  38 shared

• Katelyn Rypka

  Minneapolis VA Medical Center

  31 shared

• Michelle A. Lowes

  Rockefeller University

  29 shared

• Janewit Wongboonsin

  Siriraj Hospital

  28 shared

• Joslyn S. Kirby

  27 shared

• Zachary Wendland

  University of Minnesota

  18 shared

• Amit Garg

  Hofstra University

  17 shared

• Sonya Ahuja

  Mahidol University

  16 shared