Chemical and process engineering offers scientific tools for solving problems in the biomedical field, including drug delivery systems. This paper presents examples of analyzing the dynamics of dispersed systems (aerosols) in medical inhalers to establish a better relationship between the test evaluation results of these devices and the actual delivery of drugs to the lungs. This relationship is referred to as in vitro-in vivo correlation (IVIVC). It has been shown that in dry powder inhalers (DPls), the aerosolization process and drug release times are determined by the inhalation profile produced by the patient. It has also been shown that inspiratory flow affects the size distribution of aerosols generated in other inhalation devices (vibrating mesh nebulizers, VMNs), which is due to the evaporation of droplets after the aerosol is mixed witha dditional air taken in by the patient. The effects demonstrated in this work are overlooked in standard inhaler testing methods, leading to inaccurate information about the health benefits of aerosol therapy, thus limiting the development of improved drug delivery systems.

The aim of this research is to evaluate the performance of four UAV image processing software for the automatic estimation of volumes based on estimated volume accuracy, spatial accuracy, and execution time, with and without Ground Control Points (GCPs). A total of 52 images of a building were captured using a DJI Mavic Air UAV at 60m altitude and 80% forward and side overlap. The dataset was processed with and without GCPs using Pix4DMapper, Agisoft Metashape Pro, Reality Capture, and 3DF Zephyr. The UAV-based estimated volume generated from the software was compared with the true volume of the building generated from its as-built 3D building information modeled in Revit 2018 environment. The resulting percentage difference was computed. The average volumes estimated from the four software with the use of GCPs were 4757.448 m3 (3.87%), 4728.1 m3 (2.54%), 4291.561 m3 (11.5%), and 4154.938 m3 (14.35%), respectively. Similarly, when GCPs were not used for the image processing, average volumes of 4631.385 m3 (4.52%), 4773.025 m3 (1.6%), 4617.899 m3 (4.89%), and 4420.403 m3 (8.92%) were obtained in the same order. In addition to the volume estimation analysis, other parameters, including execution time, positional RMSE, and spatial resolution, were evaluated. Based on these parameters, Agisoft Metashape Pro proved to be more accurate, time-efficient, and reliable for volumetric estimations from UAV images compared to the other investigated software. The findings of this study can guide decision-making in selecting the appropriate software for UAV-based volume estimation in different applications.