Can Machine Learning Predict Particle Deposition at Specific Intranasal Regions Based on Computational Fluid Dynamics Inputs and Nasal Geometry Measurements - study

Aug 5, 2026

Upload by Jiven

Machine learning can be used to estimate where inhaled particles deposit in the upper airway, using computational fluid dynamics (CFD) outputs and patient nasal geometry measurements. A protocol was proposed to map nasal-spray formulation parameters to device settings for effective drug delivery, addressing the limited availability of patient data across diverse demographics. Eleven anatomical parameters were measured from 10 CT-based nasal geometries, while 160 CFD simulations modeled drug delivery under varied breathing scenarios and pressure gradients. Eighteen machine-learning models were trained to predict deposition across upper-airway regions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages1

Reading Time3 min

AccessFree to download

Machine-learning prediction for intranasal deposition

Study rationale and goal

Anatomical measurements from CT geometries

CFD simulations and deposition evaluation

Model training and regional prediction

FAQs

What problem does the study address for intranasal drug delivery?

It targets improved, patient-aligned deposition in upper airway regions by linking nasal spray design parameters and device-relevant settings to predicted drug placement.

How were nasal geometry measurements obtained and used?

Eleven anatomical parameters were measured from 10 CT-based nasal geometries representing the population to parameterize the delivery scenario.

What data and modeling approach produced deposition predictions?

The study ran 160 CFD simulations across breathing situations and pressure gradients, then trained 18 machine-learning models to estimate targeted deposition in different upper-airway regions.

Aug 5, 2026

Upload by Jiven