TSEML: A task-specific embedding-based method for few-shot classification of cancer molecular subtypes
Authors:
Ran Su,
Rui Shi,
Hui Cui,
Ping Xuan,
Chengyan Fang,
Xikang Feng,
Qiangguo Jin
Abstract:
Molecular subtyping of cancer is recognized as a critical and challenging upstream task for personalized therapy. Existing deep learning methods have achieved significant performance in this domain when abundant data samples are available. However, the acquisition of densely labeled samples for cancer molecular subtypes remains a significant challenge for conventional data-intensive deep learning…
▽ More
Molecular subtyping of cancer is recognized as a critical and challenging upstream task for personalized therapy. Existing deep learning methods have achieved significant performance in this domain when abundant data samples are available. However, the acquisition of densely labeled samples for cancer molecular subtypes remains a significant challenge for conventional data-intensive deep learning approaches. In this work, we focus on the few-shot molecular subtype prediction problem in heterogeneous and small cancer datasets, aiming to enhance precise diagnosis and personalized treatment. We first construct a new few-shot dataset for cancer molecular subtype classification and auxiliary cancer classification, named TCGA Few-Shot, from existing publicly available datasets. To effectively leverage the relevant knowledge from both tasks, we introduce a task-specific embedding-based meta-learning framework (TSEML). TSEML leverages the synergistic strengths of a model-agnostic meta-learning (MAML) approach and a prototypical network (ProtoNet) to capture diverse and fine-grained features. Comparative experiments conducted on the TCGA Few-Shot dataset demonstrate that our TSEML framework achieves superior performance in addressing the problem of few-shot molecular subtype classification.
△ Less
Submitted 13 January, 2025; v1 submitted 17 December, 2024;
originally announced December 2024.
