TriDeNT: Triple Deep Network Training for Privileged Knowledge Distillation in Histopathology

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Farndale, Lucas, Insall, Robert, Yuan, Ke
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914894907965440
author Farndale, Lucas
Insall, Robert
Yuan, Ke
author_facet Farndale, Lucas
Insall, Robert
Yuan, Ke
contents Computational pathology models rarely utilise data that will not be available for inference. This means most models cannot learn from highly informative data such as additional immunohistochemical (IHC) stains and spatial transcriptomics. We present TriDeNT, a novel self-supervised method for utilising privileged data that is not available during inference to improve performance. We demonstrate the efficacy of this method for a range of different paired data including immunohistochemistry, spatial transcriptomics and expert nuclei annotations. In all settings, TriDeNT outperforms other state-of-the-art methods in downstream tasks, with observed improvements of up to 101%. Furthermore, we provide qualitative and quantitative measurements of the features learned by these models and how they differ from baselines. TriDeNT offers a novel method to distil knowledge from scarce or costly data during training, to create significantly better models for routine inputs.
format Preprint
id arxiv_https___arxiv_org_abs_2312_02111
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TriDeNT: Triple Deep Network Training for Privileged Knowledge Distillation in Histopathology
Farndale, Lucas
Insall, Robert
Yuan, Ke
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Tissues and Organs
Computational pathology models rarely utilise data that will not be available for inference. This means most models cannot learn from highly informative data such as additional immunohistochemical (IHC) stains and spatial transcriptomics. We present TriDeNT, a novel self-supervised method for utilising privileged data that is not available during inference to improve performance. We demonstrate the efficacy of this method for a range of different paired data including immunohistochemistry, spatial transcriptomics and expert nuclei annotations. In all settings, TriDeNT outperforms other state-of-the-art methods in downstream tasks, with observed improvements of up to 101%. Furthermore, we provide qualitative and quantitative measurements of the features learned by these models and how they differ from baselines. TriDeNT offers a novel method to distil knowledge from scarce or costly data during training, to create significantly better models for routine inputs.
title TriDeNT: Triple Deep Network Training for Privileged Knowledge Distillation in Histopathology
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Tissues and Organs
url https://arxiv.org/abs/2312.02111