Machine Learning-Aided Microdroplets Breakup Characteristic Prediction in Flow-Focusing Microdevices by Incorporating Variations of Cross-Flow Tilt Angles
Bibliographic record
Abstract
Controlling droplet breakup characteristics such as size, frequency, regime, and droplet quality within flow-focusing microfluidic devices is critical for different biomedical applications of droplet microfluidics such as drug delivery, biosensing, and nanomaterial preparation. The development of a prediction platform capable of forecasting droplet breakup characteristics can significantly improve the iterative design and fabrication processes required for achieving desired performance. The present study aims to develop a multipurpose platform capable of predicting the working conditions of user-specific droplet size and frequency and reporting the quality of the generated droplets, regime, and hydrodynamical breakup characteristics in flow-focusing microdevices with different cross-junction tilt angles. Four different neural network-based prediction platforms were compared to accurately estimate capsule size, generation rate, uniformity, and circle metric. The trained capsule size and frequency networks were optimized using the heuristic optimization approach for establishing the Pareto optimal solution plot. To investigate the transition of the droplet generation regime (i.e., squeezing, dripping, and jetting), two different classification models (LDA and MLP) were developed and compared in terms of their prediction accuracy. The MLP model outperformed the LDA model with a cross-validation measure evaluated as 97.85%, demonstrating that the droplet quality and regime prediction models can provide an engineering judgment for the decision maker to choose between the suggested solutions on the Pareto front. The study followed a comprehensive hydrodynamical analysis of the junction angle effect on the dispersed thread formation, pressure, and velocity domains in the orifice.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".