A High-Throughput Automated Microfossil Classification System Using Deep Learning
Year: 2025
Type: conference-abstract
Abstract: The rapid advancement of computational power has facilitated the widespread adoption of deep learning, a subset of artificial intelligence (AI), in various fields. Automated microfossil classification using AI is increasingly explored as a solution to reduce labor and address the declining availability of skilled personnel. However, practical implementation in research remains limited due to challenges such as the need for extensive training datasets and the lack of advanced equipment like automated microscopes. To address these issues, we implemented deep learning as a function to automatically classify microfossils on a virtual slide scanner that can process up to 360 microscope slides continuously. This study applied the system to sediment core DCR-1PC from the Indian Ocean sector of the Southern Ocean to obtain high-resolution records of the radiolarian analysis.
Authors: Takuya Itaki, Ayumu Miyakawa, Kazuhide Mimura, Minoru Ikehara
Language: English
Subfield: Biophysics
Domain: Life Sciences
Open Access status: gold
OpenAlex ID: W6948748376
Cites:
Cited by:
Related to: