Active Learning to Classify Macromolecular Structures in situ for Less Supervision in Cryo-Electron Tomography

TL;DR

This paper introduces a Hybrid Active Learning framework for cryo-electron tomography that reduces the need for extensive manual labeling by efficiently selecting the most informative subtomograms for annotation, maintaining high classification accuracy.

Contribution

The paper presents a novel active learning approach combining uncertainty sampling, a discriminator, and diversity strategies to improve subtomogram classification with limited labels.

Findings

Achieves only 3% accuracy drop with less than 30% labeled data.

Demonstrates effectiveness on both simulated and real cryo-ET data.

Reduces labeling effort significantly while maintaining performance.

Abstract

Motivation: Cryo-Electron Tomography (cryo-ET) is a 3D bioimaging tool that visualizes the structural and spatial organization of macromolecules at a near-native state in single cells, which has broad applications in life science. However, the systematic structural recognition and recovery of macromolecules captured by cryo-ET are difficult due to high structural complexity and imaging limits. Deep learning based subtomogram classification have played critical roles for such tasks. As supervised approaches, however, their performance relies on sufficient and laborious annotation on a large training dataset. Results: To alleviate this major labeling burden, we proposed a Hybrid Active Learning (HAL) framework for querying subtomograms for labelling from a large unlabeled subtomogram pool. Firstly, HAL adopts uncertainty sampling to select the subtomograms that have the most uncertain predictions. Moreover, to mitigate the sampling bias caused by such strategy, a discriminator is introduced to judge if a certain subtomogram is labeled or unlabeled and subsequently the model queries the subtomogram that have higher probabilities to be unlabeled. Additionally, HAL introduces a subset sampling strategy to improve the diversity of the query set, so that the information overlap is decreased between the queried batches and the algorithmic efficiency is improved. Our experiments on subtomogram classification tasks using both simulated and real data demonstrate that we can achieve comparable testing performance (on average only 3% accuracy drop) by using less than 30% of the labeled subtomograms, which shows a very promising result for subtomogram classification task with limited labeling resources.