OpenIDEO is an open innovation platform. Join our global community to solve big challenges for social good. Sign Up / Login or Learn more

Hazel: preserving meaningful conversations

An unobtrusive remote home monitoring system that can track and remind the care-recipients of their daily tasks.

Photo of Lillian Jia
4 4

Written by

Who is your idea designed for and how does it better support family caregivers as they care for a loved one with dementia?

Caregivers constantly visit their loved ones to make sure that they are safe and healthy. Their loved ones often feel interrogated and tend to lash out. This puts stress on their familial bond. With the help of Hazel, caregivers can check in on their loved ones anytime, anywhere, via a mobile app, granting them more personal time. It also softens the tension between both parties by allowing for better conversation during visits; something beyond menial questions like “did you take your pills”.

Hazel is a remote home monitoring system that allows the caregiver to see what the care-recipient is doing around their homes on their smartphones. Smart-home sensors are retrofitted throughout the care-recipient’s home. The smart-home sensors utilizes machine learning with gesture recognition to infer the care-recipient’s activities. Our MVP targets activities such as: their medicine habits, fall detection, and door activities.

The second part of the solution includes a smart-home dashboard, which lives on a tablet placed inside the care-recipient’s home. The smart-home dashboard can receive notifications from the sensors and alerts set by the caregiver application. When it comes time, it will verbally remind the care-recipient of their daily routine or tasks (eg: “It’s time to take your afternoon medicine”).

What early, lightweight experiment might you try out in your own community to find out if the idea will meet your expectations?

We have tested our sensors in a laboratory setting, but we need to retrofit our sensors in a few care-recipients’ homes and monitor them for a period of time to fully validate our methods. During these trials, we’d like to conduct periodic interviews to understand where we can improve our product design to better suit our users needs and preferences. We’d also like to conduct more discovery interviews with caregivers and care-recipients to better understand their daily struggles.

What skills, input, or guidance from the OpenIDEO community would be most helpful in building out or refining your idea?

We would really benefit from hearing from the family caregivers and their care-recipients, and in general anyone with experience in this industry. The more we know about your daily struggles and pain-points, the better we can tailor Hazel’s experience to helping you solve your problems. We’d welcome any feedback, suggestions, and most importantly, if you’re willing to be interviewed by us, or participate in our field trials.

How long has your idea existed?

  • Over 1 year

This idea emerged from

  • A group brainstorm

Tell us about your work experience:

We are a team consisting of: a biomedical engineering researcher who studies tech to objectively characterize Alzheimer’s disease; an electrical engineering researcher who studies IC design and power management; and a product designer who specializes in the IoT field.


Join the conversation:

Photo of Kate Rushton

Hi Lillian and team,

What are the costs for set up? How can this be accessible for low and middle-income families?

Photo of Lillian Jia

Hi Kate,

Hazel has three parts in terms of hardware:
1. the sensors, we're able to sell them for a nominal fee because they are cheap to manufacture;
2. the smartphone, which many caregivers will already have;
3. the tablet, which can easily be purchased for less than $100.
Finally, we are thinking of charging a nominal monthly fee for Hazel's services. In order to determine the number for that recurring subscription fee, we'd like to further interview users to understand what they would pay for, and how much they can afford. It is our goal to make Hazel affordable and accessible to all families, especially low to mid-income families.

Lastly, to give you a sense of how low we can charge for our custom sensors, the sensors we are using cost between $2-7 usd, depending on which functionality. To compensate for the basic hardware, we developed gesture recognition algorithms to detect specific actions that the users will perform. The algorithms are based on machine learning which can further specify the activity the user performs (e.g., taking a pill versus moving the pill box).


View all comments