Related Experiment Video
Updated: Jan 9, 2026
12:11
ELIME Enzyme Linked Immuno Magnetic Electrochemical Method for Mycotoxin Detection
Published on: October 23, 2009
14.7K
Machine learning approach using electrochemical immunosensor data for precise classification of Opisthorchis
Nang Noon Shean Aye1,2, Sakda Daduang3, Patcharaporn Tippayawat2
1Biomedical Sciences Program, Graduate School, Khon Kaen University, Khon Kaen, 40002, Thailand.
Scientific Reports
|December 3, 2025
Summary
A new machine learning model accurately detects Opisthorchis viverrini (OV) infection using electrochemical biosensor data and patient information. This breakthrough aids early diagnosis and public health management of this foodborne parasitic disease.
Area of Science:
- Parasitology
- Infectious Diseases
- Biomedical Engineering
Background:
- Opisthorchiasis, caused by Opisthorchis viverrini (OV), is a significant foodborne parasitic zoonotic disease in Southeast Asia.
- Accurate classification of OV infection is crucial for effective public health interventions and disease management.
Purpose of the Study:
- To develop and validate a reliable machine learning (ML) classification model for Opisthorchis viverrini (OV) infection.
- To integrate electrochemical immunosensor data with patient condition features for intuitive OV infection classification.
Main Methods:
- Analysis of 531 urine samples from OV-positive and OV-negative individuals using an electrochemical immunosensor.
- Evaluation of six different ML models, including decision tree, AdaBoost, and neural networks, through cross-validation.
- Utilizing peak current data from the immunosensor and patient condition features for model training.
Main Results:
- Decision tree and AdaBoost classifiers achieved the highest accuracy of 90.65% in classifying OV infection.
- The decision tree model showed 91% F1 score, 95% sensitivity, and 83% specificity.
- The AdaBoost model achieved a 90% F1 score, 94% sensitivity, and 86% specificity, demonstrating robust performance.
Conclusions:
- The proposed ML model, integrating electrochemical biosensor data, offers a reliable method for accurate OV infection classification.
- This approach facilitates early diagnosis and decision-making, even without expert personnel, aiding disease surveillance and control in endemic regions.
- The study highlights the potential of combining biosensor technology with ML for improved management of parasitic zoonotic diseases.
