Indexat a

Llicència i ús

Citacions

Altmetrics

Grant support

The work of N. P. Garcia de la Puente was supported by the grant PID2022-140189OB-C21 funded by MICIU/AEI/10.13039/501100011033 ERDF/UE and FSE+. The work of M. Lopez-Perez was supported by the grant JDC2022-048318-I funded by MICIU/AEI/10.13039/501100011033 and the "European Union NextGenerationEU/PRTR". This work was also supported by the project PID2022-140189OB-C21 (ASSIST) funded by MICIU/AEI/10.13039/501100011033 and by "FEDER, EU".

Anàlisi d'autories institucional

Compartir

Domain Adaptation for Unsupervised Cancer Detection: An Application for Skin Whole Slides Images from an Interhospital Dataset

Publicat a:Lecture Notes In Computer Science. 15004 58-68 - 2024-01-01 15004(), DOI: 10.1007/978-3-031-72083-3_6

Autors: Garcia-de-la-Puente, Natalia P; Lopez-Perez, Miguel; Launet, Laetitia; Naranjo, Valery

Afiliacions

Resum

Skin cancer diagnosis relies on assessing the histopathological appearance of skin cells and the patterns of epithelial skin tissue architecture. Despite recent advancements in deep learning for automating skin cancer detection, two main challenges persist for their clinical deployment. (1) Deep learning models only recognize the classes trained on, giving arbitrary predictions for rare or unknown diseases. (2) The generalization across healthcare institutions, as variations arising from diverse scanners and staining procedures, increase the task complexity. We propose a novel Domain Adaptation method for Unsupervised cancer Detection (DAUD) using whole slide images to address these concerns. Our method consists of an autoencoder-based model with stochastic latent variables that reflect each institution's features. We have validated DAUD in a real-world dataset from two different hospitals. In addition, we utilized an external dataset to evaluate the capability for out-of-distribution detection. DAUD demonstrates comparable or superior performance to the state-of-the-art methods for anomaly detection https://github.com/cvblab/DAUD-MICCAI2024.

Paraules clau