top of page

Alzheimer’s Diagnostic Breathalyzer Device Design

The Problem and Opportunity

There currently exists no non-invasive test for Alzheimer's Disease. A commercial diagnostic product would enable better treatment options and life outcomes for millions of patients.

The Advanced Materials and Microsystems Lab developed an electrochemical gas sensor technology that is sensitive (0.0001 ppt) and selective enough to detect clinically relevant Alzheimer's biomarkers (BHT, pivalic acid, and 2,3-DMH) from exhaled breath. They needed a partner to develop a manufacturable breath collection system with a tolerable user experience for older adult patients.


Northeastern University Advanced Materials and Microsystems Lab


product research, solutioning and prototyping, project management 

The Solution

My bioengineering team of 4 built functional user-facing device form factor to support state-of-the-art gas-particle detection sensor technology. I prioritized and modular function and accessibility through hardware bound user experience design. 

Our advanced design features that met these challenges as opportunities. (1) Designed a device to expose gas sensor to a sample of the patient's exhaled alveolar air. Our transport and fluid dynamics informed internal architecture also incorporated rigid safety and environmental standards. (2) The sensor technology is scalable to other clinically-relevant template molecules. Advocated for modular functionality via cartridge form factor that would allow a single breathalyzer device to support a platform of sensor chips. (3) Carefully considered user experience from the perspective of patients and point-of-care providers, payers, and other end-users. Our device design is minimal, durable, and supports self and assisted administration. 

My Process

(1) Understand how the user will experience the product and physical constraints to perform intuitive and reproducible breath testing

  • Expose sensor to sufficient biomarkers from exhaled breath to generate a successful test result.

  • 90% of users will be able to collect their breath sample and receive a successful diagnosis indication with minimal device orientation.

  • Support adoption via integration with existing health infrastructure and use in a variety of environments (versatility).

  • Device form factor supports sensor-chip re-use and replacement and sanitation practices.

(2) Define and specify patient and device requirements

  • The system shall be able to get a reading within 3 deep breaths at vital lung capacity.

  • The cost of goods shall be less than $100 per unit and the design compatible with injection molding manufacturing.

  • Our device shall use industry standard communication (copy and iconography) and use limited commands and alerts (that meet inclusive design standards).

(3) Ideate solutions that solve the user's needs

  • Our team formalized a number of design decisions including rapid test vs batching and mass-processing systems; procedures to capture necessary deep-lung alveolar air breath samples (quantified exhale force requirements); and transport dynamics at sensor surface to facilitate biomarker diffusion. These possible solution form factors were evaluated for user experience for 98% percentile of patients.

(4) Prototype and demonstrate functionality 

  • The position of the sensor and the orientation of the breath flow in the device was a paramount design considerations with great impact on architecture requirements to create the environment for a successful test. Our final prototype utilized impinging jet configuration (sensor oriented perpendicular to the flow). Breath passes over the sensor in the internal sensor chamber, before being channeled out of the device.

  • The breath nozzle's mouthpiece integrates with industry-standardized disposable mouthpiece brands. 

  • The closure of the front and back panel (part b, below) creates seals the sensor chamber. Opening the device's panel 'resets' the sensor and allows modular chip functionality, optimizing platform system cost.

(5) Evaluate the solutions you designed against the user's needs 

  • Our calculated time constant to biomarker saturation plateaued between 1.21-1.82 seconds. Concentration data indicated that the time constant was dependent on flow rate, not biomarker concentration; validating efficacy of our (standardized) diagnostic test parameters. Validated that target compounds are being successfully delivered to the sensor (our design did not impede detection)

  • Our prototype solution paired with a health app interface, greatly reducing the necessary complexity of hardware indicators of earlier versions. Our digital orientation would coach the patient through three progressively deeper breaths, followed by a hold, and finally a long exhale into the mouthpiece. The device hardware features an LED that blinks in sync with the indications on the app screen, signaling to continue exhaling, and flashes rapidly when the test is done. The indicator turns green if the diagnostic test was completed successfully and red if the test encountered error. 


The purpose of this Capstone was to integrate the Advanced Materials and Microsystem Lab's gas biomarker detection sensor into a prototype breathalyzer for Alzheimer's Disease diagnosis. All our design decisions were quantified and weighted for (1) ability to generate a repeatable and successful breath test (system must adequately expose breath sample to sensor for a sufficient duration and intensity to pass/fail a test). (2) To create a positive and intuitive user experience based on principals of universal design empathy for all end-users, especially older adults and point of care providers. And (3) that the cost of materials and manufacturing would be less than $100/unit.

We used flow modeling in 3D CAD to verify that jet impinged design allowed for the best tradeoff between necessary force of exhale (high force would be difficult for older adults) and gas-chemical binding interactions at the sensor surface (that the air movement is slower than the binding time constant.

We tested our hardware design and digital/physical interface with 5 adults over the age of 55. Further testing to be continued.

Finally both the costs of our 3D printing fabrication process and manufacturers-estimate of mass-produced injection modeling scalability were under our targeted allowable costs.

bottom of page