Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Radical Cascades on Seawater Microdroplets Drive Atmospheric Mercury Oxidation.

Journal of the American Chemical Society·2026

Same author

PFAS in Fast Foods with Hot-Contact to Food-Contact Materials: Occurrence, Packaging Influence, and Risk Priority.

Environmental science & technology·2026

Same author

Weather-Specific Individual Exposure to Organic Contaminants in Fine Particulate Matter by Nontarget Screening and Risk Prioritization.

Environment & health (Washington, D.C.)·2026

Same author

Core-shell Cu<sub>2</sub>O@UiO-66 enabled pre-concentration electrochemical sensor for rapid, sensitive detection of ammonia nitrogen in groundwater.

Talanta·2026

Same author

Legacy and Emerging Per- and Polyfluoroalkyl Substances (PFAS) in Glacial Meltwater from Mt. Everest.

Environmental science & technology·2026

Same author

Bisphenol TMC Impaired Mouse Oocyte Maturation by Disrupting Spindle Assembly via NEK3-Mediated Microtubule Acetylation.

Environmental science & technology·2026

Search research articles

Related Experiment Video

Updated: Sep 5, 2025

Integrating Remote Sensing with Species Distribution Models; Mapping Tamarisk Invasions Using the Software for Assisted Habitat Modeling SAHM

Integrating Remote Sensing with Species Distribution Models; Mapping Tamarisk Invasions Using the Software for Assisted Habitat Modeling SAHM

Published on: October 11, 2016

Siamese Network-Based Transfer Learning Model to Predict Geogenic Contaminated Groundwaters.

Hailong Cao¹, Xianjun Xie¹, Jianbo Shi¹

¹School of Environmental Studies & State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan 430074, China.

Environmental Science & Technology

|July 11, 2022

Summary

A new deep learning method, Siamese network-based transfer learning (SNTL), effectively predicts geogenic contaminated groundwaters (GCGs) using limited, imbalanced data. This approach improves public health protection where traditional methods fail due to poor data quality.

Keywords:

Siamese network class-imbalanced data groundwater prediction transfer learning

More Related Videos

Measuring Carbon-based Contaminant Mineralization Using Combined CO2 Flux and Radiocarbon Analyses

Measuring Carbon-based Contaminant Mineralization Using Combined CO2 Flux and Radiocarbon Analyses

Published on: October 21, 2016

Continuous Instream Monitoring of Nutrients and Sediment in Agricultural Watersheds

Continuous Instream Monitoring of Nutrients and Sediment in Agricultural Watersheds

Published on: September 26, 2017

Related Experiment Videos

Last Updated: Sep 5, 2025

Integrating Remote Sensing with Species Distribution Models; Mapping Tamarisk Invasions Using the Software for Assisted Habitat Modeling SAHM

Integrating Remote Sensing with Species Distribution Models; Mapping Tamarisk Invasions Using the Software for Assisted Habitat Modeling SAHM

Published on: October 11, 2016

Measuring Carbon-based Contaminant Mineralization Using Combined CO2 Flux and Radiocarbon Analyses

Measuring Carbon-based Contaminant Mineralization Using Combined CO2 Flux and Radiocarbon Analyses

Published on: October 21, 2016

Continuous Instream Monitoring of Nutrients and Sediment in Agricultural Watersheds

Continuous Instream Monitoring of Nutrients and Sediment in Agricultural Watersheds

Published on: September 26, 2017

Area of Science:

Environmental Science
Public Health
Machine Learning

Background:

Geogenic contaminated groundwaters (GCGs) pose significant public health risks.
Existing machine learning models struggle with insufficient and imbalanced groundwater quality data.
Accurate prediction of GCGs is crucial for effective risk management.

Purpose of the Study:

To develop a novel deep learning method for predicting GCGs.
To address challenges posed by limited and imbalanced groundwater quality datasets.
To improve the accuracy and reliability of GCG prediction models.

Main Methods:

A Siamese network-based transfer learning (SNTL) approach was employed.
The method was tested on limited and class-imbalanced groundwater quality data.
Performance was benchmarked against traditional Random Forest models.

Main Results:

SNTL significantly reduced the need for extensive training data.
The method mitigated the negative impacts of class-imbalanced data on model performance.
SNTL models demonstrated higher (approx. 80%) and more balanced sensitivity and specificity for predicting GCGs (arsenic, fluoride, iodine) compared to Random Forest.

Conclusions:

SNTL offers a robust solution for predicting GCGs even with scarce and imbalanced data.
This method enhances the ability to protect populations from GCG exposure in data-limited regions.
SNTL provides a valuable tool for environmental health risk assessment and management.