Research

Cutaneous lymphomas are a heterogeneous group of rare malignancies that arise in the skin. Their diagnosis can be challenging because clinical and histopathological features frequently overlap with inflammatory skin diseases, sometimes resulting in prolonged diagnostic uncertainty. Following diagnosis, predicting disease progression and selecting the most appropriate treatment can be equally difficult because of considerable biological and clinical heterogeneity.

Our research aims to improve diagnosis, risk stratification and treatment of cutaneous lymphomas by better understanding the clinical and biological characteristics of these diseases and translating these insights into improved patient care. We combine clinical and translational research with advanced computational approaches, including digital pathology and artificial intelligence, and study the tumour microenvironment and molecular characteristics of cutaneous lymphomas.

Within our AID-CLYM programme [link naar Stichting Hanarth Fonds - Anne-Roos Schrader], we investigate how computational pathology and AI can support diagnosis and identify morphological biomarkers associated with disease progression and clinical outcome. We are extending this work towards multimodal and longitudinal approaches that integrate pathology with clinical photographs, longitudinal clinical information, treatment history and molecular profiling.

Because progress in rare diseases requires collaboration, we established the Cutaneous Lymphoma International Digital Pathology (CLIDIPA) network [link naar https://clidipa.org] to facilitate international research and external validation. CLIDIPA is evolving towards a FAIR multimodal research infrastructure supporting collaborative clinical, translational and computational research across European cutaneous lymphoma centres.

Key publications

1. Doeleman T, Brussee S, Valkema PA, Kempf W, Vermeer MH, Kers J, Wynaendts LCD, Kerckhoffs KGP, de Jonge MM, Nguyen AH, Peters EEM, Wobser M, Rosenwald A, Stadler R, Jansen PM, Battistella M, Roccuzzo G, Quaglino P, Schrader AMR. Histological triage of early-stage mycosis fungoides using a weakly supervised deep learning-based model: a multicentre, external validation, and clinical utility study. Under review (2026).
   
   • Preprint: https://medrxiv.org/cgi/content/short/2026.07.27.26359009v1
     
   • Code available: https://github.com/LUMCPathAI/MIMIC
     
2. Doeleman T, Beljaards ESM, Ottevanger R, Jansen P, Vermeer MH, Quint KD, Willemze R, Schrader AMR. Added value of pathology consultations in cutaneous lymphomas: a 2-year review from the Dutch national referral and expertise centre. Br J Dermatol. 2025 Aug 18;193(3):514-520. PMID: 40367149.
3. Doeleman T, Brussee S, Hondelink LM, Westerbeek DWF, Sequeira AM, Valkema PA, Jansen PM, He J, Vermeer MH, Quint KD, van Dijk MR, Verbeek FJ, Kers J, Schrader AMR. Deep Learning-Based Classification of Early-Stage Mycosis Fungoides and Benign Inflammatory Dermatoses on H&E-Stained Whole-Slide Images: A Retrospective, Proof-of-Concept Study. J Invest Dermatol. 2025 May;145(5):1127-1134.e8. PMID: 39306030.
   
   • Code available: https://github.com/Sbrussee/MF_BID_MIL
     
4. Brussee S, Buzzanca G, Schrader AMR, Kers J. Graph neural networks in histopathology: Emerging trends and future directions. Med Image Anal. 2025 Apr;101:103444. PMID: 39793218.
   
5. Doeleman T, Hondelink LM, Vermeer MH, van Dijk MR, Schrader AMR. Artificial intelligence in digital pathology of cutaneous lymphomas: A review of the current state and future perspectives. Semin Cancer Biol. 2023 Sep;94:81-88. PMID: 37331571.​​​​​​​