An Adversarial Approach to Register Extreme Resolution Tissue Cleared 3D Brain Images

TL;DR

This paper introduces InvGAN, a patch-based generative network that effectively registers extremely high-resolution 3D tissue images from tissue clearing, significantly outperforming traditional methods in speed and accuracy.

Contribution

The paper presents InvGAN, a novel deep learning model specifically designed for high-resolution 3D tissue image registration, addressing limitations of existing methods in speed and effectiveness.

Findings

InvGAN achieves comparable accuracy to traditional methods at lower resolution.

At full resolution, InvGAN outperforms traditional registration tools in speed, reducing registration time from 28 hours to 10 minutes.

InvGAN maintains high registration quality across different resolutions.

Abstract

We developed a generative patch based 3D image registration model that can register very high resolution images obtained from a biochemical process name tissue clearing. Tissue clearing process removes lipids and fats from the tissue and make the tissue transparent. When cleared tissues are imaged with Light-sheet fluorescent microscopy, the resulting images give a clear window to the cellular activities and dynamics inside the tissue.Thus the images obtained are very rich with cellular information and hence their resolution is extremely high (eg .2560x2160x676). Analyzing images with such high resolution is a difficult task for any image analysis pipeline.Image registration is a common step in image analysis pipeline when comparison between images are required. Traditional image registration methods fail to register images with such extant. In this paper we addressed this very high resolution image registration issue by proposing a patch-based generative network named InvGAN. Our proposed network can register very high resolution tissue cleared images. The tissue cleared dataset used in this paper are obtained from a tissue clearing protocol named CUBIC. We compared our method both with traditional and deep-learning based registration methods.Two different versions of CUBIC dataset are used, representing two different resolutions 25% and 100% respectively. Experiments on two different resolutions clearly show the impact of resolution on the registration quality. At 25% resolution, our method achieves comparable registration accuracy with very short time (7 minutes approximately). At 100% resolution, most of the traditional registration methods fail except Elastix registration tool.Elastix takes 28 hours to register where proposed InvGAN takes only 10 minutes.