Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model

作者： Veerayuth Kittichai1, Morakot Kaewthamasorn2, Suchansa Thanee2, Thanyathep Sasisaowapak3, Kaung Myat Naing3, Rangsan Jomtarak4, Teerawat Tongloy3, Santhad Chuwongin3, Siridech Boonsang5 1Faculty of Medicine,King Mongkut’s Institute of Technology Ladkrabang, 2Veterinary Parasitology Research Unit, Faculty of Veterinary Science,Chulalongkorn University, 3College of Advanced Manufacturing Innovation,King Mongkut’s Institute of Technology Ladkrabang, 4Faculty of Science and Technology,Suan Dusit University, 5Department of Electrical Engineering, School of Engineering,King Mongkut’s Institute of Technology Ladkrabang

简介：

Overview

This study focuses on developing an AI model utilizing a hybrid deep learning technique for identifying trypanosome species from microscopic images. The approach aims to enhance the diagnosis of blood parasites through automated screening methods.

Key Study Components

Area of Science

• Neuroscience
• Biology
• Artificial Intelligence

Background

• Blood parasites pose significant health risks worldwide.
• Traditional diagnostic methods can be time-consuming and require expert knowledge.
• AI and deep learning techniques offer potential solutions for rapid diagnosis.
• The study focuses on trypanosome species, including T. cruzi, T. brucei, and T. evansi.

Purpose of Study

• To develop an AI model for rapid identification of trypanosome species.
• To improve diagnostic accuracy in remote areas.
• To automate the screening process for blood parasites.

Methods Used

• Hybrid deep learning model combining object detection and classification.
• Microscopic imaging of blood samples.
• Low-code AI platform (CiRA CORE) for ease of use.
• Data labeling and training using the DeepTrain function.

Main Results

• Successful identification of trypanosome species from microscopic images.
• Improved accuracy in diagnosing mixed and immature infections.
• Automated standard taxonomy for species classification.
• Potential for local staff to utilize the AI model in remote areas.

Conclusions

• The AI model can revolutionize the surveillance of blood parasites.
• It provides a rapid and accurate screening method.
• Collaboration of active monitoring and AI enhances disease control efforts.

Frequently Asked Questions