PROCESS
It shows how I come to the current design solution through this process. It started with a discovery stage to explore and scope down the problem space, then we came to the define stage to understand pain points, each stakeholder's needs and design constrains. With the pain points, I started to do ideation and design with constrains and came to the design features. Finally, after several failures, pivots in design features and several iterations, I came to the final design.
1/DISCOVERY
PROBLEM
Sudden Infant Death Syndrome (SIDS) is the sudden, unexpected death of an infant under 1 year of age that remains unexplained after a complete investigation.
In the United States, SIDS is the major cause of death in infants between 1 month and 1 year of age. SIDS is sudden and silent, occurring most often during sleep, with no signs of suffering. Annually, approximately 2,500 infants die of SIDS in the U.S., and as the third leading cause of infant mortality.
Though the precise cause(s) of SIDS is unknown, eliminating controllable risk factors is a recommended practice
Secondary
research
Sleeping
environment
Sleeping
posture
Ed
Mitchell
Use of a non-safety approved mattress or crib
Use of a drop side crib.
Use of wedges, positioners, or bumper pads in cribs.
Use of an overly soft mattress (should not conform to the shape of baby’s head or body)
Presence of soft or loose bedding (e.g., quilts, heavy blankets and bumper pads)
Use of a non-fitted sheet
Presence of other soft or extraneous items (e.g., pillows, stuffed toys)
Presence of other adults, children or animals in the same sleeping surface as the infant (including in beds or couches)
Presence of excessively high temperature (above 72°C) and smoke
Studies have found that the side sleep position is unstable and increases the chance that infants will roll onto their stomachs—the sleep position associated with the highest SIDS risk.
In 2005, the American Academy of Pediatrics (AAP) recommended that healthy babies should be placed on their backs to sleep as the main way to reduce SIDS risk.
Expert
Interview
1/ Sleeping on the tummy elevates risk.
2/ Sleeping on the sides is also risky.
3/ Sleeping on the back is the safest.
4/ Pillows, quilt and cushions are the major causes of death according to
the pictures of recovered baby sleeping scenes sent back from the
family whose infants died of SIDS.
5/ This project has a strong meaning for research and will help collect
data to understand the causes of SIDS
Summary
Realistic problem
Meaningful impact
Raise people's awareness
Help find cause
Two possible solutions:
1/ sleeping environment
2/ Baby’s sleep postures
Constrains
1/Not fully ascertained causes
2/ can only give suggestions
Design questions:
2/Define
Parent
Interview
1
User's needs
Reliability
1/ Reduce false alert / no alert
2/ Stable connections
3/ Give immediate alerts when danger happens
4/ Safe mounting mechanism
5/ Not aware of SIDS
6/ Professional and simple
7/ Multiple users
Protect privacy
1/ Less exposed baby's information.
2/ Know where the video go to.
Peace of mind
1/ Less unnecessary notifications.
2/ Show most important information.
3/ 2-way communication.
4/ Love the analysis, coach and advice.
1/ Having a baby at home is a nightmare, especially for those having a busy working
schedule
2/ I always hear my baby crying and suddenly wake up in the middle of the night to
check if my baby is safe.
3/ I want a product to tell me whether my baby is safe or not, but I just feel that I don't
want to expose my baby's images or video to these baby monitor companies. I need
to know where these video are stored and it's better when I can delete them.
4/ I want a baby monitor that can give me an accurate alerts. Sometimes, it sends me
useless alert or even false alerts. I am annoyed at the amount of the notifications it
sends, it sometimes sends me the sound from downstairs which is useless.
5/ I don't want to put the monitor in the crib or hang it on the wall.
6/ I don't want the baby monitor to fully replace my supervision so I only use them
when my wife and I are both not at home.
7/ I have a baby monitor at home, but the app looks silly and not professional at all.The
wifi connection is so bad....
8/ I don't really feel my baby cam make a huge difference. You can't tell anything wrong
from the video at night, you have to zoom in the video with your finger to see clear
whether your baby's alive. My wife still wakes up several times at midnight.
9/ I love the way it tracks and analyses my baby's sleeping habits and coach me on how to improve them.
10/ Love the 2- way communication!
11/ Love the breathing sensor accessories.
12/ I want the audio be on sleeping mode and push audio notification just when baby cry
Current
Solutions
PROS
1/Voice communication
2/ Fix the monitor on the crib
3/ Fisheye camera - better view.
4/ Baby album
1/non-contact sleep and respiration tracking
2/ Fit into the context of the child’s environment
CONS
1/No full view of the crib.
2/ Most babies will want to pull, tug and put their mouth on the gadget
1/It's bad at determining baby is sleeping/awake/not in crib
2/ Don't work well in low light situations
INSIGHT
1/Fisheye camera
2/ Do not fix the monitor on the crib, but hang it on the wall.
3/ Take photos of your infants.
1/Using machine learning model to have higher detection accuracy.
2/ Fit into the context of the child’s environment
2
1/Crying detection
2/ A tree limb that you mount to your wall and then you wrap the Lollipop camera arm around it.
1/Terrible connections.
2/ Some detections not work.
3/ Horrible App.
1/Cute appearance and smart mounting strategies
2/ A user-friendly app without making too much sounds
1/Real-time breathing data
2/See the movement of your baby with vision mode
3/Instant alerts - baby is crying or beginning to wake.
1/Bad detection accuracy
2/ The video is too large to upload
3/ The app is so buggy.
1/Good mounting system.
2/ Compressed the video.
3/ Give instant alerts or messages.
4/ Track the movement of the kids.
Current solution on Privacy Issue
1/ Not saying where is the video goes to. you can only see the live stream video.
2/ Users can take photos of their babies from the video
3/ Store on a website but hard to download and delete.
4/ Store in a SD card but only has limited storage
Stakeholder
Analysis
3
As a product, we have plenty of other stakeholders other than parents and kids. These different stakeholders have different expectations for us and they also bring constrains to us.
Design requirements
Summary
1/Open-source: cheap material and easy to DIY
2/ Non-profit projects
1/Accurate results -- safe or not
2/Track history -- do coaching
3/Air quality and temperature
4/Detect baby's sleeping environment
5/Give instant alerts
Business
Technology
User Experience
1/Give most important alerts
2/Less unnecessary notifications
3/Privacy protection
4/Kid-friendly enclosure
5/Profession and simple
Constrains
1/Open-source: limited processing power
2/ No real baby data
3/Ideation
Features
STORYBOARD
1/Get BGB baby monitor.
2/Hold in on the crib.
3/Connect the phone and the monitor.
4/Take photos of baby's sleeping environment.
5/Upload photos into the cloud.
6/Download machine learning model.
7/Set up personal preference.
8/Log in dashboard and watch real-time video in office.
9/Get alert when risk factors detected in the crib.
10/Parents can turn off the alert after going back home.
11/Have a good rest at night with BGB monitor at your back.
Touch points
analysis
4/DESIGN
Prototype
Brainstorming
Attractive
Animals
Toy
Child-friend enclosure
Round corners
Educational
Hide the camera
Compact
Candies
Cartoon
Blinking LED
Camera
Toy camera
Track history
User-friendly Dashboard
Color
DIY - Change-able enclosure
Open-source
Customized level of risks
data visualization
Mood Board
Sketch
Iterations
In order to reduce cost and get rid of digital screens, we add LED indicators and cute round circle on the top of enclosure to indicate risk
The infrared light source is too bright at night, so we cut a round translucent acrylic to cover it during the night.
Battery powered -- to make it wireless users only need to charge during the day.
Final Prototype
Problem:
1/ We can't hold it on the wall because the raspberry pi is too heavy and it is not safe enough for the kids.
Front-End Design
I did the front end development using flask to connect data sent from cloud and sensor with the dashboard.
We upload images instead of video with a 0.5s frame rate. We show more graphs and user can turn off all the alerts they don't want.
Problem:
At that time, we don't have enough time to do UI design.
5/improvement
Likes
Usability test
"I love the design and the cute shape!!"
"Actually I love the tripod, I can put it on anything!"
"I love the magnet chargers, it is convenient and make me feel safe for me kids"
"It sends my instant alert messages to my phone when high risk factors detected in the crib"
For UI interface
1/ "There is too much information in the dashboard, I don't really need to know every graphs and every reading."
Problems
2/ "The dashboard is beautiful, but it looks kind of unprofessional, not looks like a standard baby monitor app"
3/"I want to know more about what should I do instead of just telling
me there are risk factors existing in the crib"
4/"I have twin babies, can I have only one camera?"
5/"I don't want my wife to be seen in the camera, especially when she is breast-feeding my baby, I want to stop the video at the time."
For prototype
1/ "The indicators LEDs are too bright in the night for kids. Even they are coved with translucent acrylics."
2/ "It's better to make it rotatable or I have to change the angles and direction of the tripod manually"
3/ "It's kind of heavy as a baby monitor. I am afraid to mount it on the wall."
UI Design
In order to provide more user-friendly dashboard. I translated users' insight into design features.
1/ "There is too much information in the dashboard, I don't really need to know every graphs and every reading."
-Separate real-time information and history reading
-Give scores on each day's safety level and users
can understand the data in a more intuitive way.
2/ "The dashboard is beautiful, but it looks kind of unprofessional, not looks like a standard baby monitor app"
- Change the color.
- Show information in a more organized way.
3/"I want to know more about what should I do instead of just telling me there are risk factors existing in the crib"
- Give tips and recommend people to read more articles and buy some certain products.
- Show weekly progress (in associate with scores)
4/"I have twin babies, can I have only one camera?"
-Add two or more devices.
-Add two or more users
5/"I don't want my wife to be seen in the camera, especially when she is breast-feeding my baby, I want to stop the video at the time."
-Customization options to stop videos, audio, alerts
-Customization options for sensitivity, camera access
"Protect privacy."
"Open-source"
-Users upload the images of their babies and their
sleeping environment into the cloud to train on their own machine learning model
-Less information and in a neat block layout.
-Audio, video, led alert, text alert options.
-Only show average readings
-Show overall scores -- Haven't been tested yet- I think it will help users know better how safe their babies sleeping environment
-Tips on how to improve
-Have some business possibility?
-Add one or more devices.
-Upload your images into the cloud. read the instructions.
-Customize privacy issue and risk level.
Usability test#2
In order to further improve the user interface, I conducted a second usability test. I recorded the mistakes they made in a sequence of tasks and asked them their opinion about these pages.
Insights
Homepage:
1/ Change the position of text to the left side of the switch.
2/ Show real-time sensor reading on a more obvious way.
3/ Show the video by default
History:
1/ Show current date in a more obvious way
2/ Tell me more what I need to do to improve it instead of only showing the data and scores.
3/ Empathize the overall scores more in this page, that’s the most comprehensive and important information
Account:
1/ Changing system setting is a less frequent action for users. So, if changing the sensitivity level is a frequent action, it’s better to put it in the homepage just near the real-time video. However, what I need to consider here is that should users change the sensitivity level themselves, should users take responsibility of fault tolerance. How much freedom should I give them?
So, in the end I still put it in the setting page, I don’t encourage users to take responsibility of choosing the right sensitivity level, but I give user opportunities to customize their choices based on their preference.
2/ The relation between the information card and cloud card are very confusing to the users. At first, I put them in two separate cards, so they thought these two were irrelevant to each other. So in order to tell users information means the basic information of the cloud, I re-organized the layout of the page.
