livelink to matlab  (MathWorks Inc)

Journal: iScience

Article Title: Deep learning-augmented T-junction droplet generation

doi: 10.1016/j.isci.2024.109326

Figure Lengend Snippet: Dynamic process of the study based on the T-junction droplet simulations (A) The COMSOL simulation resulted in a phase description of the final generated droplets, stemming from the settings initialized for interaction between oil and water, which was then conducted using the Livelink MATLAB transcript function of the software to create and store images of phase-defined generated droplets for various input parameters. (B) Images resulting from FEA were then subjected to an image analysis process in which a binary format of those images was used to enhance the performance of measuring parameters visualized in the droplet formation. (C) According to measured variables in binary images, two important output parameters were extracted that emphasize the goal of the current research: Regime and Droplet Length. (D) In each scenario of image creation, four main inputs resulted in two numerical outputs, which were then ordered in a table to create a dataset of 8020 data points. (E) The resulting dataset was then trained with ML and DL methods, including classification models to train the droplet generation regime and regression models with the purpose of training-droplet length. (F) Finally, the trained models were used to estimate the main outputs for the proposed T-junction droplet generation setup.

Article Snippet: Through the Livelink MATLAB transcript function, images of the generated droplets were captured and stored ( A).

Techniques: Generated, Software

Journal: iScience

Article Title: Deep learning-augmented T-junction droplet generation

doi: 10.1016/j.isci.2024.109326

Figure Lengend Snippet:

Article Snippet: Through the Livelink MATLAB transcript function, images of the generated droplets were captured and stored ( A).

Techniques: Software

Journal: iScience

Article Title: Deep learning-augmented T-junction droplet generation

doi: 10.1016/j.isci.2024.109326

Figure Lengend Snippet: Dynamic process of the study based on the T-junction droplet simulations (A) The COMSOL simulation resulted in a phase description of the final generated droplets, stemming from the settings initialized for interaction between oil and water, which was then conducted using the Livelink MATLAB transcript function of the software to create and store images of phase-defined generated droplets for various input parameters. (B) Images resulting from FEA were then subjected to an image analysis process in which a binary format of those images was used to enhance the performance of measuring parameters visualized in the droplet formation. (C) According to measured variables in binary images, two important output parameters were extracted that emphasize the goal of the current research: Regime and Droplet Length. (D) In each scenario of image creation, four main inputs resulted in two numerical outputs, which were then ordered in a table to create a dataset of 8020 data points. (E) The resulting dataset was then trained with ML and DL methods, including classification models to train the droplet generation regime and regression models with the purpose of training-droplet length. (F) Finally, the trained models were used to estimate the main outputs for the proposed T-junction droplet generation setup.

Article Snippet: Utilizing Livelink MATLAB, phase-defined droplet images were generated and stored for various parameters.

Techniques: Generated, Software

Journal: iScience

Article Title: Deep learning-augmented T-junction droplet generation

doi: 10.1016/j.isci.2024.109326

Figure Lengend Snippet:

Article Snippet: Utilizing Livelink MATLAB, phase-defined droplet images were generated and stored for various parameters.

Techniques: Software