logo
ResearchBunny Logo
Multi-channel portable odor delivery device for self-administered and rapid smell testing

Engineering and Technology

Multi-channel portable odor delivery device for self-administered and rapid smell testing

R. Hopper, D. Popa, et al.

Discover a groundbreaking portable device for efficient odor delivery, tailored for research and clinical applications. Created by Richard Hopper and his team, this innovative technology boasts 24 individual channels and precise digital control via a mobile app, paving the way for self-administered smell tests in various settings.

00:00
00:00
~3 min • Beginner • English
Introduction
Smell is a key sensory modality underpinning health and well-being. Smell dysfunction can impair quality of life, affect social confidence and safety (e.g., inability to detect hazards), and alter flavor perception with downstream effects on nutrition and immunity. It is also an early biomarker for neurological diseases such as Alzheimer’s and Parkinson’s. With aging populations, there is a growing need for early diagnosis, prevention, care closer to home, and continuous monitoring, motivating accessible smell testing technologies. Olfactory testing is challenging due to the high dimensionality of odor space and the broad dynamic range of human receptors, often requiring many odorants and dilutions. Tests must be rapid, easy to administer, and still provide clinically valuable metrics like thresholds. Current clinical threshold tests (e.g., Sniffin’ Sticks) rely on multiple dilution pens administered by trained staff using complex, time-consuming procedures, so they are seldom performed. There is a need for compact, flexible systems enabling efficient delivery of many odors with precise timing and minimal contamination. In this context, the authors developed a portable, affordable, multi-channel odor delivery device enabling rapid, flexible sequencing of odors and supporting self-administered threshold testing via a mobile app.
Literature Review
Olfactometers automate temporally precise odor delivery using filtered air from pumps or gas cylinders, odor sources, delivery channels, and user interfaces like masks or nasal attachments. Odorized air can be generated from liquid bottles, sample bags, syringes, or by thermal evaporation. Concentration control is typically via gaseous dilution with proportional valves and mixing chambers, which adds cost/complexity and risks cross-contamination unless rigorously cleaned. An alternative is multiple channels with serially diluted odorants controlled by solenoid valves, reducing cross-contamination and cost. Traditional olfactometer systems, comprising pumps, temperature-controlled reservoirs, valves, tubing, and mass flow controllers, are usually bulky, expensive, and lab-bound. Miniaturized devices (e.g., Aromajoin piezoelectric air pumps; OVR Tech VR-mounted atomizers) exist but target entertainment use and often compromise on flow, intensity control, directivity, temporal resolution, cross-contamination, and flexibility (e.g., proprietary cartridges). Therefore, a compact, flexible, multi-channel solution enabling rapid, contamination-free delivery of many odors for time-efficient testing remains needed.
Methodology
Device description: The device is a 24-channel portable olfactometer drawing ambient air via a diaphragm pump (Parker BTX Connect, up to 6 L/min), filtering it (activated carbon filter, Festo MS4/D-MINI-LFX), and distributing via an aluminum manifold to 24 solenoid valves (Zanty SDF-0626L). Flow is regulated (Festo GRLA-M5-QS-4-D) between 2–6 L/min. Each valve directs air to a separate odor reservoir in a removable aluminum cartridge holding sponge-absorbed liquid odorants; small headspace enables rapid saturation. Individual Teflon outlet pipes prevent cross-contamination; a resin-printed outlet adapter focuses the odor stream at ~10 cm. Sealing uses O-rings; cartridges are cleanable (baking soda solution, oven). Electronics: An Adafruit Feather nRF52840 (Raytac MDBT50Q-1MV2, BLE, ARM Cortex-M4) controls the system. Digital outputs drive shift registers and 12 V MOSFET drivers to actuate valves and pump (PWM). Up to three channels can be active simultaneously. Sensors: pressure (Freescale MPX 53GP), temperature/humidity (Sensirion SHT21S). Power rails (3.3/5/12 V) via Murata switching regulators. Control and data collection: Android mobile app (JavaScript) over Bluetooth allows odor triggering and collects user perceptual ratings; data stored in a cloud database. Specifications: 24 channels; max flow 6 L/min; 4-mm outlet; reservoir 46×14×16 mm; noise ~60 dB; power 15 W; USB/Bluetooth; 305×220×75 mm; 3.9 kg. Cost at current prices ~$2,140 plus per-odor disposables; expected to drop with volume. Device characterization protocols: Odorant (SMELL-RS/SMELL-S mixture) 300 µL applied to cellulose sponge in reservoir. Airflow measured with Flusso FLS-110; mean and SD from 50 cycles. Odor intensity measured with a photoionization detector (Ion Science MiniPID): sensor at 25 mm from 4-mm outlet, 25 °C, 3 L/min flow; means/SDs from 10 cycles. Short-pulse repeatability measured indoors over 1 h with activations every 300 s, varying pulse durations (1–6 s). Thermal stability measured in an environmental oven (Thermotron S-1.2 3800), with odorized air piped to an external PID; 30 min stabilization per temperature point. Spatial distribution measured by mounting PID on a motorized stage (Thorlabs LTS300) with baffles to reduce air currents; measurements parallel and transverse to flow at defined distances. Smell test study design: Ethics-approved test–retest reliability and accuracy study at Geneva University Hospital with healthy subjects (n=37) and patients with smell loss (n=31). Participants (≥18 years) attended two visits ~1 week apart. Visit 1 included Sniffin’ Sticks and SMELL-S using the device; completion times recorded. Order randomized; tests repeated on visit 2 for validity (results not shown). Statistics: Shapiro–Wilk for normality; due to non-normality, Mann–Whitney to compare completion times. Sniffin’ Sticks threshold subtest: phenylethyl alcohol with 16 geometric dilutions; triangle 3-AFC with two solvent pens and one target; single-staircase starting at most difficult (level 16), reversals after two successive correct; threshold = mean of last four of seven reversals. SMELL-S threshold subtest: self-administered via app; complex odor mixture with 10 geometric dilutions; starts at medium difficulty (level 5); otherwise same staircase/reversal rules as Sniffin’ Sticks.
Key Findings
Device performance: - Airflow uniformity: Mean outflow 3.0 L/min (regulated), maximum deviation from mean 5% across 24 channels (n=50 cycles). - Short-pulse repeatability: For 1–6 s activations over 1 h, relative standard deviation (RSD) <4.2% for all durations; PID shows sharp pulse with decay to baseline ~10 s (detector transient ~3 s). - Continuous activation: Peak odor intensity drops by ~10% after 86 s, indicating reservoir depletion; short pulses with recharge intervals recommended for stability. - Channel-to-channel odor intensity: RSD 4.7%, max deviation 11.4% for 3 s pulses across 24 channels. - Spatial distribution: Odor intensity decreases rapidly with distance; at 30 mm from outlet, intensity ~10% of near-source peak. Transverse distribution at 100 mm shows strong positional dependence, necessitating accurate user alignment (e.g., headrest) for repeatability. - Thermal response: Odor intensity temperature coefficient ~5% per °C over 15–40 °C. Application to threshold testing: - Completion time: Median time Sniffin’ Sticks threshold = 14 min (IQR 9) vs SMELL-S with device = 6 min (IQR 3). Mann–Whitney two-tailed p<0.0001 (α=0.05); U=371.5; sums of ranks 4732 (Sniffin’ Sticks) and 2650 (SMELL-S). - Example staircase performance illustrates lower thresholds for normosmic subject vs patient with smell loss; threshold score defined as mean of last four reversals. Overall, the device delivers temporally precise, repeatable odor pulses with limited cross-channel variability and enables significantly faster self-administered threshold testing compared with the current standard.
Discussion
Findings demonstrate that a compact, 24-channel, digitally controlled odor delivery device can provide reliable, temporally precise odor stimuli without cross-contamination, supporting rapid olfactory threshold testing. The system’s consistent airflow, low pulse-to-pulse variability, and manageable channel-to-channel differences indicate suitability for research and clinical settings. Spatial and thermal characterizations highlight operational considerations: ensuring consistent user positioning and maintaining quasi-stable temperatures. In application, SMELL-S administered via the device significantly reduced testing time relative to Sniffin’ Sticks, largely by eliminating manual tasks (capping/uncapping, verbal interactions, manual data recording). Digital integration with an app and cloud database streamlines data collection, potentially improving data quality and scalability. Although the technical performance is promising, clinical validation of SMELL-RS/SMELL-S with the device is ongoing; thus claims of clinical utility are preliminary. The device’s flexibility (up to 24 stimuli, three simultaneous channels) and cartridge-based design enable customizable protocols, positioning it as a practical tool for self-administered smell testing and broader olfactory research.
Conclusion
A portable, affordable, 24-channel odor delivery device was developed and characterized, offering high temporal precision, minimized cross-contamination via dedicated channels, and easy cartridge exchange. Characterization confirmed stable, repeatable odor delivery at short pulse durations, with manageable channel variability. In a comparative study, the device enabled substantially faster self-administered threshold testing (SMELL-S) than the standard Sniffin’ Sticks. Digital control via a mobile app and cloud ecosystem facilitates efficient data collection and broader deployment. The device’s design allows administration of any test using up to 24 stimuli, supporting adaptable research and potential clinical applications. Future work includes full clinical validation of SMELL-RS/SMELL-S with the device, potential enhancement of thermal stability (e.g., temperature-controlled reservoirs), and refinement of user alignment aids to further improve repeatability, enabling routine self-monitoring of smell health.
Limitations
- Thermal sensitivity: Odor intensity varies with ambient temperature (~5% per °C); mitigating this would require temperature-controlled reservoirs, increasing cost/complexity. - Continuous-use depletion: During continuous activation, intensity declines (~10% drop after ~86 s); stable delivery requires short pulses with recharge intervals. - Spatial sensitivity: Odor intensity depends strongly on user position; accurate alignment (e.g., headrest) is necessary to ensure repeatability. - Environmental effects: Air currents and minor temperature changes contribute to intensity variability; baffling and controlled environments help. - Channel variability: While modest (RSD ~4.7%), differences may arise from odorant distribution on sponges and airflow nuances. - Clinical validation: Although time savings are demonstrated, broader clinical utility and validation of SMELL-RS with this device are ongoing; second-visit validity results were not reported here. - Simultaneous activation limit: Up to three channels can be active simultaneously, which may constrain certain paradigms.
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny