(C-105) Studying physiological flow dynamics in the upper respiratory system using a fabricated anatomically precise nasal simulator

The structure and function of the human upper respiratory system is complex, and the diversity of flow modes within the nasal cavity are vital for proper exchange of heat, moisture, and mass.  Few bioengineered systems have been developed to replicate the physiological flow dynamics within the nasal cavity. Here, we detail the process for the three-dimensional printing of a nasal simulator based on a computed tomography (CT) scan of adult nasal cavities that can generate physiologically relevant nasal airflow patterns. The simulator is composed of nasal “slices,” to measure and monitor air distribution through different cavity regions. Breathing is stimulated via an Arduino- and GUI-controlled fan which mimics bidirectional airflow through input waveform signals. With the fabrication of this nasal simulator, more accurate airflow simulations can be achieved for the conduction of biological and environmental research related to air quality screening.

In order to recreate the precise anatomical geometry of the nasal cavity and inner turbinates, various modeling programs such as AD Inventor were used to convert a CT scan into the face/nasopharynx object as well as associated parts. Further, to accurately print such organic shapes we utilized the Anycubic Photon Mono M3 resin printer. An Ultimaker 3D printer was used to print the hardware, such as the sliding rack which holds the nasal slices, and the fan casing which includes channels for bidirectional airflow. This nasopharynx base sits on a custom laser-cut acrylic box housing the electronic components. Finally, an Arduino Uno microcontroller generates waveforms to mimic respiratory patterns. 

The nasal simulator holds anatomically correct features of the nasal cavity in slices placed on a sliding track system for easy access. Anemometers are inserted in sampling regions to assess air velocity or particle distribution, taking 1 reading/100 milliseconds. Two channels of outer casing allow for bidirectional airflow. The control panel and motors generate waveforms to mimic typical airflow patterns such as inhalation, exhalation, forced breathing, sniffing, etc. 

Results:

Two main experiments were conducted using the fabricated robotic nasal simulator. The results and conclusions are listed below for each.

Experiment 1: Proof of Concept

The first experiment was to validate the precision of the setup and fabrication strategies used to create the model. A constant flow rate was set through the Arduino and the maximum flow rate at different locations was measured using the anemometer. The results of this were generally found to be comparable to the results from a computational fluid dynamics simulation done on the upper respiratory system in 2017 (Li et al.).

Experiment 2: Flow Rate by Location

The purpose of this experiment was to measure how the flow rate changes as air passes through regions of the nose. We simulated quiet breathing using the Arduino, generating standard inhalation and exhalation waveforms. We then measured with anemometers how the flow rate varies at each probing location as a function of airflow velocity for a full minute. We found the flow velocity during inhalation to be highest at the front of the nose whereas the flow velocity during exhalation was highest at the back of the nasal cavity.

Reynold’s Analysis:

One of our objectives was to see how air flow classification changes by region in the nose. In particular, we looked at turbulent air flow, vital to air humidification and cleaning. Thus, the Reynold’s number was also plotted using the collected data. These classifications are extremely important because they aid in the further study of the functionality of the nose, as well as a more accurate understanding of particle distribution and deposition, perhaps as it pertains to nasal drug delivery.

Discussion:

In the future, we aim to use differently powered motors to change flow rates to stimulate processes like sniffing and forced breathing. In order to better plot and understand regional flow changes, we may scale up the model in the future to take more accurate measurements. One proposed extension of this study is to incorporate biological components such as ciliated tissue cultures. These developments will allow us to effectively replicate and understand the physiology and flow dynamics of the nose.

I would like to thank my mentor Dr. Dixon for her invaluable guidance and for pushing me consistently, as well as Isira Vithanage for his leadership in this project. Thank you also to the Think[box] student makerspace staff for their help during fabrication.

 1.Rygg, A., Hindle, M., & Longest, P. W. (2015). Absorption and Clearance of Pharmaceutical Aerosols in the Human Nose: Effects of Nasal Spray Suspension Particle Size and Properties. Pharmaceutical Research, 33(4), 909-921.

2. Li, C., Jiang, J., Dong, H., & Zhao, K. (2017). Computational modeling and validation of human nasal airflow under various breathing conditions. Journal of Biomechanics, 64, 59-68.