On the validation of generational lung deposition computer models using planar scintigraphic images: the case of mimetikos preludium
Olsson, B; Kassinos, SC
HERO ID
8803992
Reference Type
Journal Article
Year
2021
Language
English
PMID
| HERO ID | 8803992 |
|---|---|
| In Press | No |
| Year | 2021 |
| Title | On the validation of generational lung deposition computer models using planar scintigraphic images: the case of mimetikos preludium |
| Authors | Olsson, B; Kassinos, SC |
| Journal | Journal of Aerosol Medicine and Pulmonary Drug Delivery |
| Volume | 34 |
| Issue | 2 |
| Abstract | Background: Mechanistic computer models for calculation of total and regional deposition of aerosols in the lungs are important tools for predicting or understanding clinical studies and for facilitating development of pharmaceutical inhalation products. Validation of such models must be indirect since generational in vivo data are lacking. Planar scintigraphy is probably the most common method addressing regional lung deposition in humans. Scintigraphic regions of interest (ROI) contain mixtures of airway generations and can therefore not be directly compared to model results. We propose a method to translate computed deposition per generation to deposition in scintigraphic ROI to be able to compare computed results with corresponding results obtained in humans. Methods: The total and regional lung deposition computed by the one-dimensional algebraic typical-path software Mimetikos Preludium was compared for 18 study legs in 14 published deposition studies involving 9 dry powder inhaler brands to the activity in planar scintigraphic ROIs (oropharyngeal, central [C], intermediate, and peripheral [P]) using for the computed regional lung distribution a generic mapping of the contribution of each airway generation to the ROIs. Results: The computed oropharyngeal and total lung deposition correlated with high significance (p < 0.0001) to the scintigraphic results with a near one-to-one relationship. For the regional lung distribution, computed C, P, and P/C results correlated with high significance (p < 0.01) to the corresponding scintigraphic measures. The computed C (P) deposition was on average about 28% lower (8% higher) than the mean scintigraphic results. The computed P/C ratio was on average 29% higher than the mean scintigraphic ratio. Conclusions: The results indicate that both the computational deposition model and the mapping algorithm are valid. The small underprediction of the C region merits further investigations. We believe that this method may prove useful also for the validation of computational fluid particle dynamic lung deposition models. |
| Doi | 10.1089/jamp.2020.1620 |
| Pmid | 32790531 |
| Wosid | WOS:000570547400001 |
| Is Certified Translation | No |
| Dupe Override | No |
| Is Public | Yes |
| Language Text | English |
| Keyword | computer models; lung deposition; scintigraphy; validation |