SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.13650 · CVPR · 2024

Tyche: Stochastic In-Context Learning for Medical Image Segmentation

John Guttag, Mosaicml Databricks, Adrian Dalca, Marianne Rakic, Jose Gonzalez Ortiz, Csail Mgh, Beth Mit, Cimini

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Existing learning-based solutions to medical image segmentation have two important shortcomings. First, for most new segmentation tasks, a new model has to be trained or fine-tuned. This requires extensive resources and machinelearning expertise, and is therefore often infeasible for medical researchers and clinicians. Second, most existing segmentation methods produce a single deterministic segmentation mask for a given image. In practice however, there is often considerable uncertainty about what constitutes the correct segmentation, and different expert annotators will often segment the same image differently. We tackle both of these problems with Tyche, a framework that uses a context set to generate stochastic predictions for previously unseen tasks without the need to retrain. Tyche differs from other in-context segmentation methods in two important ways. (1) We introduce a novel convolution block architecture that enables interactions among predictions. (2) We introduce incontext test-time augmentation, a new mechanism to provide prediction stochasticity. When combined with appropriate model design and loss functions, Tyche can predict a set of plausible diverse segmentation candidates for new or unseen medical images and segmentation tasks without the need to retrain.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2401.13650")
get_code_for_paper("2401.13650")
have("2401.13650")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2401.13650.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2401.13650.svg)](https://syntology.ai/paper/2401.13650)

Connect an agent — have() is free.