Innovative microfossil (radiolarian) analysis using a system for automated image collection and AI-based classification of species

Takuya Itaki,Minoru Ikehara,Tatsuhiko Hoshino,Hitoshi Saito,Naoki Kuwamori,Yosuke Taira

doi:10.1038/s41598-020-77812-6

Abstract

Microfossils are a powerful tool in earth sciences, and they have been widely used for the determination of geological age and in paleoenvironmental studies. However, the identification of fossil species requires considerable time and labor by experts with extensive knowledge and experience. In this study, we successfully automated the acquisition of microfossil data using an artificial intelligence system that employs a computer-controlled microscope and deep learning methods. The system was used to calculate changes in the relative abundance (%) of Cycladophora davisiana, a siliceous microfossil species (Radiolaria) that is widely used as a stratigraphic tool in studies on Pleistocene sediments in the Southern Ocean. The estimates obtained using this system were consistent with the results obtained by a human expert (< ± 3.2%). In terms of efficiency, the developed system was capable of performing the classification tasks approximately three times faster than a human expert performing the same task.

Highlights

Microfossils that have been preserved in sediments and rocks from geological strata have been used extensively over the last 70 years for determining geological ages and in paleoenvironmental studies
The aim of this study is to demonstrate the usefulness of the miCRAD system for revealing a microfossil assemblage using image collection and classification units of the system
The C. davisiana% has been used as a paleoceanographic indicator of intermediate water formation (e.g.,9,16). This experiment was composed of following three steps: (1) collection of images of individual objects for the training dataset using the miCRAD system, (2) construction and test of the classification model based on deep learning method, and (3) estimation of the particle composition based on classification results

Summary

Introduction

Microfossils that have been preserved in sediments and rocks from geological strata have been used extensively over the last 70 years for determining geological ages and in paleoenvironmental studies. Marchant et al.[13] reported that changes in the relative abundance of benthic foraminiferal assemblages estimated using a CNN-based classification showed good agreement with manual counts performed by humans. These recent studies have shown the effectiveness of deep learning as a method for microfossils classification. The automated system, which has a rapid image acquisition function combined with an accurate classification model, enables non-experts to efficiently identify large numbers of microfossils and is expected to be applied to the analysis of microfossil assemblages. The C. davisiana% has been used as a paleoceanographic indicator of intermediate water formation (e.g.,9,16)

Objectives

Methods

Results

Conclusion