High-accuracy prediction of 5-year progression-free melanoma survival using spatial proteomics and few-shot learning on primary tissue
Preprint
Abstract
SUMMARY: Emerging techniques in imaging-based spatial proteomics (ISP) enable in-depth insights into the architecture and protein abundance of tissue(s). Explainable machine learning (xML) models promise to yield substantial advances in ISP data-based diagnosis and prognosis. However, a clinical application of these new possibilities predicting the course of a tumor has not been suggested yet. Here, we use a few-shot learning workflow on histological multi-antigen images to predict 5-year progression-free survival (PFS) in melanoma. We address the problem of a relatively small cohort ( n = 22), by utilizing a pre-trained convolutional neural network (CNN) model, which we further pre-train on a proxy task for which more samples were available ( n = 39) before fine-tuning for PFS prediction. Our approach yielded a model achieving an accuracy of more than 90%, outperforming baseline models trained on clinical data by around 10%. Using an xML technique, we identified immune infiltration and proteins associated with tumor progression as crucial predictors. This indicated that our models’ PFS predictions are not only highly accurate but also grounded in a relevant biological background.Keywords
This publication has 58 references indexed in Scilit:
- A historical perspective of biomedical explainable AI researchPatterns, 2023
- A survey of uncertainty in deep neural networksArtificial Intelligence Review, 2023
- A distinct stimulatory cDC1 subpopulation amplifies CD8+ T cell responses in tumors for protective anti-cancer immunityCancer Cell, 2023
- Neoantigen-specific CD4+ T cells in human melanoma have diverse differentiation states and correlate with CD8+ T cell, macrophage, and B cell functionCancer Cell, 2022
- Self supervised contrastive learning for digital histopathologyMachine Learning with Applications, 2022
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based LocalizationInternational Journal of Computer Vision, 2019
- The (Un)reliability of Saliency Methods2019
- Cell Detection with Star-Convex Polygons2018
- Toponome Imaging System (TIS): imaging the proteome with functional resolutionNature Methods, 2007
- Analyzing proteome topology and function by automated multidimensional fluorescence microscopyNature Biotechnology, 2006