“Pixie” asthma alert wearable

I worked with @Xiaoyi and @Wanqiao on this project for Philips Healthcare Hackathon and landed 3rd prize.

We were given the challenge to leverage data collected by wearables to help individuals with respiratory diseases self-manage. We identified key metrics of asthma prediction and identification with clinical evidence, conducted field research and leveraged data from multiple fields to achieve first-stage prediction accuracy.

We started from the journeys of the patients and identified two major opportunities:

  • How might we provide preventive alerts to patients to help them avoid asthma attacks?  
  • How might we create safer environments for asthma patients to live in and reduce the risk for patients at an aggregate level?

We identified the stakeholders and explored what role our product could play. We aimed to help alert family of the patient’s asthma attacks real-time, improve communication between the patient and medical workers, and involve insurance companies to expand usage of our product.

Narrowing of the respiration tract is the first thing that happens in an asthma attack, which consequently leads to breathing difficulties. Breathing generates air flow in the trachea, and when the trachea narrows, the frequency of breathing sound changes.

We designed a headset with detectors adhering to the neck that mimic a stethoscope chestpiece, embedded with microphones to collect the sound waves generated in the trachea during breathing. Several key components are located at the back of the headset: a bluetooth component connects the device to a caregiver’s phone, an LTE component communicates data collected from the detector to the cloud for analysis, battery is for continuous use, and a vibration component to alert the patient when an asthma attack is about to occur.

The appearance of headsets can also be customized to look fashionable!

The basic function of our product includes location-based potential hazard alert, and soundwave-based asthma attack alert. The patient or his/her caregiver will use a mobile phone app which is paired to the headset. The app will have access to a map with known asthma-related information, for example pollen content, and when the mobile device comes into an area with high asthma attack risk, the app will send an alert to both the mobile device and the headset to alert both the caregiver and patient.

When the headset detects a change in the breathing sound that indicates trachea narrowing, the vibration module will vibrate to alert the patient, and the linked app will simultaneously generate a notification to the caregiver. Both the patient and the caregiver will have a better chance to take early action for treatment.

Users also have an option of whether an emergency call should be generated if an alert is not responded within a certain period of time.

With accumulated data, Pixie is able to generate predictions with higher accuracy.

Pixie collects information from users to continuously update its database. Users can log the time, location, identified cause and outcome of an asthma attack, to both keep track of his/her own medical history and help Pixie provide more accurate prediction and alerts for other users. Pixie can also identify high-risk locations with unknown causes with this method. With pooled information input from Pixie, cooperation with civil engineering departments is also feasible to improve the general environment for asthma patients.

As Pixie accumulates breathing soundwave recordings and the users logs the occurrence of attacks, recordings before an attack can be extracted to look for signs that appear immediately before an attack. Pixie can then use these signs to give alerts even earlier and buy the user more time in emergency treatment, which results in better outcome.

With the geometric and attack rate data altogether, cooperation with civil engineering departments is also feasible to improve the general environment for asthma patients.

We developed a use case for such patients:

Pixie can not only alert patients and their family to better respond at the time of an attack, but also help them plan ahead to minimize risks and feel secure. This helps them lice a more care-free life overall.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s