Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Advancing forward osmosis predictions: A deep learning-based surrogate modeling approach.

Hyeon Woo Park1,2, Woo-Ju Kim3,4

  • 1Department of Food Science and Biotechnology, Korea University, Sejong, Republic of Korea.

Journal of the Science of Food and Agriculture
|May 12, 2026
PubMed
Summary

Related Concept Videos

Osmosis00:47

Osmosis

Approximately 60% to 95% of the weight of living organisms is attributed to water. Therefore, maintaining appropriate water balance within cells is of paramount importance. Osmosis is the movement of water across a semipermeable membrane, such as a cell’s plasma membrane. In living organisms, water plays a crucial role as a solvent—a molecule that dissolves other molecules.
Osmosis and Osmotic Pressure of Solutions02:40

Osmosis and Osmotic Pressure of Solutions

A number of natural and synthetic materials exhibit selective permeation, meaning that only molecules or ions of a certain size, shape, polarity, charge, and so forth, are capable of passing through (permeating) the material. Biological cell membranes provide elegant examples of selective permeation in nature, while dialysis tubing used to remove metabolic wastes from blood is a more simplistic technological example. Regardless of how they may be fabricated, these materials are generally...

You might also read

Related Articles

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

Sort by
Same author

Application of machine learning for the prediction of Bacillus cereus growth: A study on the integration of mathematical models.

International journal of food microbiology·2026
Same author

Development of whey protein-derived aerogel microcarriers for cultivated meat production.

Food chemistry·2026
Same author

Author Correction: Magnon confinement in epitaxial antiferromagnetic oxide heterostructures.

Nature materials·2026
Same author

The cylindrical devices with tunable positive, infinite, and negative capacitance for dynamic random access memory.

Nature communications·2026
Same author

Optimization and validation of air impingement cleaning parameters for removing nonfat dry milk powder from stainless steel surfaces by machine learning.

Journal of dairy science·2026
Same author

Magnon confinement in epitaxial antiferromagnetic oxide heterostructures.

Nature materials·2026

A deep learning surrogate model accurately predicts forward osmosis (FO) performance, significantly reducing computational time for efficient process optimization in food manufacturing.

Area of Science:

  • Chemical Engineering
  • Data Science
  • Process Optimization

Background:

  • Assessing data-driven approaches for predicting forward osmosis (FO) performance.
  • Developing and comparing machine learning models: decision tree, random forest, support vector machine, and deep neural network (DNN).
  • Utilizing datasets of varying sizes to evaluate model applicability.

Purpose of the Study:

  • To develop a rapid and accurate deep learning-based surrogate model for FO performance prediction.
  • To compare the performance of different machine learning models under diverse operating conditions.
  • To enable efficient process optimization in food manufacturing.

Main Methods:

  • Systematic development and comparison of machine learning models.
Keywords:
deep neural networkforward osmosismachine learningsurrogate model

Related Experiment Videos

  • Training and evaluation using datasets of varying sizes.
  • Application of Shapley additive explanation (SHAP) for model interpretability.

    • Main Results:

    • Deep neural network (DNN) model achieved superior accuracy with larger datasets (>1563 data points).
    • Random forest model performed better with smaller datasets (<1563).
    • DNN surrogate predicted water flux with a normalized root mean square error (NRMSE) of 0.082, comparable to numerical simulations (0.083).

    Conclusions:

    • The proposed DNN-based surrogate model offers high predictive accuracy and significantly reduces computational time.
    • This approach enables rapid and efficient process optimization in food manufacturing.
    • The DNN model effectively captures physical relationships between input parameters and FO performance.