Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Challenges in developing a digital pathology consultation network: insights from the MELCAYA experience
1
Zitationen
20
Autoren
2025
Jahr
Abstract
<h2>Abstract</h2><h3>Objectives</h3> The increasing adoption of Whole Slide Image (WSI) technology has created new opportunities for collaboration in diagnostic pathology. This study aims to describe the development, implementation, and challenges of establishing a second-opinion platform for pathology revision, focusing on improving the diagnostic assessment of melanoma in children, adolescents, and young adults (CAYA). <h3>Methods</h3> Two-hundred-eight melanomas and atypical melanocytic lesions from CAYA patients were retrospectively collected, resulting in 311 associated WSIs. Recognizing the rarity of these cases, patient enrollment was distributed across 11 centers spanning 6 European countries. All slides were digitalized at each participating institution and uploaded into a centralized digital platform to be reviewed. <h3>Results</h3> A total of 144 WSIs (46.3%) exhibited technical defects, classified as critical, major, or minor based on their impact on WSI quality and visualization. Of these, 101 (70.1%) showed analytical defects, while the remaining 43 (29.9%) presented pre-analytical issues. Significant inter-center variability in file sizes was observed, reflecting differences in scanning protocols and tissue processing. Twenty-four cases (11.5%) exhibited major discordances on melanoma vs atypical melanocytic lesions classification, underscoring the importance of centralized pathological review. <h3>Discussion</h3> This study highlights the challenges of obtaining standardized, high-quality WSIs in real-world settings, particularly due to the rarity of the disease and the limited availability of tissue. These findings reinforce the importance of establishing standardized protocols across institutions to provide pathologists with a high-quality repository and improve patient outcome through effective second-opinion reviews.
Ähnliche Arbeiten
A survey on deep learning in medical image analysis
2017 · 13.591 Zit.
Dermatologist-level classification of skin cancer with deep neural networks
2017 · 13.195 Zit.
A survey on Image Data Augmentation for Deep Learning
2019 · 11.813 Zit.
QuPath: Open source software for digital pathology image analysis
2017 · 8.198 Zit.
Radiomics: Images Are More than Pictures, They Are Data
2015 · 8.024 Zit.
Autoren
- Dario Di Gangi
- Roberta Gugliotta
- Filippo Ugolini
- Anna Szumera‐Ciećkiewicz
- Llúcia Alós
- Natalia Castrejón de Anta
- Stephan Forchhammer
- Barbara Valeri
- Nicolas Macagno
- Şule Öztürk Sarı
- Özge Hürdoğan
- Sokol Sina
- Sylvie Fraitag
- Valerio Gaetano Vellone
- Chiara Trambaiolo
- Rita Alaggio
- Sabrina Rossi
- Giovanni Arcuri
- Daniela Massi
- Claudio J. Conti
Institutionen
- University of Florence(IT)
- Agostino Gemelli University Polyclinic(IT)
- Azienda Ospedaliero-Universitaria Careggi(IT)
- The Maria Sklodowska-Curie National Research Institute of Oncology(PL)
- Universitat de Barcelona(ES)
- University of Tübingen(DE)
- Fondazione IRCCS Istituto Nazionale dei Tumori(IT)
- Inserm(FR)
- Aix-Marseille Université(FR)
- Istanbul University(TR)
- University of Verona(IT)
- Assistance Publique – Hôpitaux de Paris(FR)
- Hôpital Necker-Enfants Malades(FR)
- Istituti di Ricovero e Cura a Carattere Scientifico(IT)
- Istituto Giannina Gaslini(IT)
- Bambino Gesù Children's Hospital(IT)