An expert-driven data generation pipeline for histological images

TL;DR

This paper introduces a novel expert-driven pipeline for generating synthetic histological images to augment limited annotated datasets, improving deep learning model training in biomedical cell segmentation.

Contribution

The proposed pipeline allows experts to incorporate domain knowledge, producing realistic synthetic images from minimal annotations to enhance deep learning training datasets.

Findings

Generated datasets improve segmentation model performance

Synthetic images closely resemble real histological data

Method reduces need for extensive manual annotations

Abstract

Deep Learning (DL) models have been successfully applied to many applications including biomedical cell segmentation and classification in histological images. These models require large amounts of annotated data which might not always be available, especially in the medical field where annotations are scarce and expensive. To overcome this limitation, we propose a novel pipeline for generating synthetic datasets for cell segmentation. Given only a handful of annotated images, our method generates a large dataset of images which can be used to effectively train DL instance segmentation models. Our solution is designed to generate cells of realistic shapes and placement by allowing experts to incorporate domain knowledge during the generation of the dataset.