Towards the in-vivo automated assessment of nasal cilia mobility

Abstract : Introduction. Cilia motility is an important diagnostic feature for various nasal diseases, both of acquired and genetic origin. So far this assessment has been performed ex-vivo in nasal samples. Sampling is invasive and may damage cilia, and ex-vivo measurements may not always reflect the in-vivo cilia function. In this work we investigated the possibility of assessing cilia motility in vivo in humans with a preliminary study using confocal micro-endoscopy. Materials and methods. We used the high-power laser confocal endoscope developed by Mauna Kea Technology (MKT). This device has a 1µm spatial resolution and a temporal resolution varying between 8 and 90 Hz. Ex-vivo pig trachea samples and human nasal biopsy samples were labelled with fluorescent marker Octadecyl Rhodamine B Chloride (R18). Beating cilia were easily identified allowing the acquisition of 40 videos ready for analysis. Using in-house software that estimates frequencies based on luminance variation in a small window and FFT analysis, we evaluated cilia and compared the beat frequency with ground truth. Results. We validated our estimations on all sequences acquired between 30 and 90Hz. Videos acquired at less than 30Hz did not offer sufficient temporal resolution We only observed occasional errors when the software identified a harmonic oscillation instead of the fundamental frequency.
Document type :
Conference papers
Complete list of metadatas

https://hal-upec-upem.archives-ouvertes.fr/hal-01333003
Contributor : Elodie Puybareau <>
Submitted on : Tuesday, September 27, 2016 - 4:29:56 PM
Last modification on : Friday, October 4, 2019 - 1:28:08 AM

Identifiers

  • HAL Id : hal-01333003, version 1

Citation

Élodie Puybareau, Emilie Bequignon, Mathieu Bottier, Gabriel Pelle, Bruno Louis, et al.. Towards the in-vivo automated assessment of nasal cilia mobility. Congress of the European Rhinologic Society, Jul 2016, Stockholm, Sweden. ⟨hal-01333003⟩

Share

Metrics

Record views

354

Files downloads

24