Contrastive and attention-based multiple instance learning for the prediction of sentinel lymph node status from histopathologies of primary melanoma tumours

Sentinel lymph node status is a crucial prognosis factor for melanomas; nonetheless, the invasive surgery required to obtain it always puts the patient at risk. In this study, we develop a Deep Learning-based approach to predict lymph node metastasis from Whole Slide Images of primary tumours. Albei...

ver descrição completa

Detalhes bibliográficos
Autores: Hernández Pérez, Carlos|||0000-0002-0054-8573, Combalia Escudero, Marc, Puig Sardá, Susana, Malvehy Guilera, Josep, Vilaplana Besler, Verónica|||0000-0001-6924-9961
Tipo de documento: capítulo de livro
Data de publicação:2022
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositório:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglês
OAI Identifier:oai:upcommons.upc.edu:2117/384779
Acesso em linha:https://hdl.handle.net/2117/384779
https://dx.doi.org/10.1007/978-3-031-17979-2_6
Access Level:Acceso aberto
Palavra-chave:Deep learning
Melanoma
Artificial intelligence -- Medical applications
Whole slide image
Contrastive learning
Attention-based multiple instance learning
Early detection
Aprenentatge profund
Intel·ligència artificial -- Aplicacions a la medicina
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Descrição
Resumo:Sentinel lymph node status is a crucial prognosis factor for melanomas; nonetheless, the invasive surgery required to obtain it always puts the patient at risk. In this study, we develop a Deep Learning-based approach to predict lymph node metastasis from Whole Slide Images of primary tumours. Albeit very informative, these images come with complexities that hamper their use in machine learning applications, namely their large size and limited datasets. We propose a pre-training strategy based on self-supervised contrastive learning to extract better image feature representations and an attention-based Multiple Instance Learning approach to enhance the model’s performance. With this work, we quantitatively demonstrate that combining both methods improves various classification metrics and qualitatively show that contrastive learning encourages the network to output higher attention scores to tumour tissue and lower scores to image artifacts.