Divcap: Dive Into Sleep
Introduction
Insomnia is a common sleep problem for adults. To solve this, I developed Divcap, a wearable device that can track people's data throughout the day and then treat people with sleep disorders through auditory and olfactory feedback based on people's daytime condition. After learning about shape-changing fabrics in the Center for Digital Innovation at Tongji University as an intern, I have a vision for the future of this product in the form of fabric.
My initial interest in sleep started in December 2019. This project was completed in June 2020. In July 2021, when I was an intern at BrainCo, I have done the preliminary product design about sleep. In January 2022, I also designed an inflatable pajama that can change a user's dreams through tactile feedback for MIT Media Lab.
Open Source
Daytime condition tracking program: https://github.com/w1200508/Facial-Expression-Recognition
Self-Driven Research Project
Supervisor
Zhanling Feng (graduated from RCA IDE)
Award
1. China National Innovation and Entrepreneurship Project
2. 2020 China-US Young Maker Competition Finalist
Publication
Divcap: A Smart Nightcap that Promotes Sleep through the Five Senses
My Skills
Sleep science & Shape-changing material research
Tangible prototype making
Electronic design (Arduino)
Background
Sleep Research
The most fascinating aspect of sleep is that it’s highly connected to almost every aspect of our waking lives and everyone’s sleep needs, patterns, and influences are highly individualized. This matters because it means that the best way to better rest and recover through sleep science is an awareness of how sleep impacts our moods and feelings and health and how our lives in return - impact the way we sleep.
However, insomnia is a common problem for adults.
30% of the general population complains of sleep disruption.
10% have associated symptoms of daytime functional impairment consistent with the diagnosis of insomnia.
The National Institutes of Health estimates.
Facts influence people's daily routine
Sleep & Dream Projects
COCOON aimed to see and shape the dream.
From this, I have a deeper understanding of the relationship between dreams and sleep.
DORMIO built future dream interactions. From this, I learned some techniques for detecting sleep.
ESSENCE project was about releasing the corresponding odor based on human data. From this, I know the importance of olfactory perception.
All three projects come from the Fluid Interfaces group at MIT Media Lab. The initial research on these projects laid the foundation for the inflatable pajama I designed for them in 2023 on the theme of dreams.
Sleep is an interesting area of science. There are two basic types of sleep: Rapid Eye Movement (REM) Sleep and Non-REM Sleep (which has three different stages: Falling Asleep Stage, Light Sleep Stage, and Deep Sleep Stage.)While there is still no agreement on “why” we sleep - there is plenty of evidence that sleep, as incredibly complex as it is is important for the proper function of the human body. So, the problem of sleep is worth exploring and solving. In the spirit of humanism, I want to design a program that can intervene and regulate sleep.
User Understanding
To better understand the specific characteristics of people with sleep problems and their own experiences as well as to understand their expectations for solving sleep problems, I chose a group aged from 15-47 to investigate their reasons for insomnia. There were 35 interviewees, 23 of whom suffered from insomnia for a long time. Here below are three typical cases.
Conclusion
Among the insomniac people interviewed, they all have taken certain measures to treat: taking sleeping pills, using sleep monitoring apps, microcurrent stimulation treatment devices, etc. However, these products have not been customized for the user's own condition. The treatment effect is not as expected, and they are not proactive intervention products.
Ideation
After research and brainstorming, I try to explore a product that can be worn by people to intervene and regulate people's sleep. In the sketch, I considered the human-computer relationship and I was trying to design a product that can help sleep.
I want to develop a product that can treat people with sleep disorders through auditory and olfactory dimensions, combined with expression recognition technology and brain wave detection technology which can get user data to customize the feedback.
Sleep Experiments
Purpose: Analyze whether music and smell can promote sleep by comparing the amplitude of brainwaves when people sleep in the ordinary state with the variation of brain waves when people are accompanied by music or smell.
Process: I did four tests. All are starting at 0:00 AM and each lasts one hour.
Result Data: Waveforms in the right picture are the values of attention and mediation detected by the TGAM module, ranging from 0-100. The red line is attention and the blue line is mediation. The bottom four graphs ranged from 100s to 2600s. (excluding the unstable data at the beginning of 100s) After the experiment, I compared the four groups of data and found that music can help to enter shallow sleep early 450s and aroma can help to enter shallow sleep early 900s.
Promoting Sleep Factors Verification
Determine Odor & Music Based on Mood
Purpose: Use Facial Expression Recognition to define five emotions that a person might have in a day, and then determine certain songs and smells for each emotion status.
Why this tech: The main symptoms of insomnia are the effects on daytime performance, such as feeling tired, irritable, emotional disorders, and inattention. It is vital to know a person’s all-day mood rather than just detect his/her brainwaves at night. I hope to collect a full day's data on people and analyze the data in two aspects to get a treatment plan.
Five Emotions: Missing, Anger, Excitement, Sorrow, Fright.
Music Example: Hundred Birds Phoenix, Autum Moon. Music can improve the stability of the mental state and affect the personality of the individual, so as to reduce the body's sensitivity to adverse social environment stimuli, thereby improving the quality of sleep.
Scent Example: Agarwood, Real Lavender, Rosa Fragrant.
Special thanks to Han Ding, my classmate at Shanghai University majoring in computer science, for his help in the program.
Design
Design Concept
Good sleep is not only a necessary condition for physical health but also a prerequisite for a happy life. Divcap aims to help people with sleep problems have a better sleep experience. Technically speaking, it combines AI and big data analytics to connect human emotions with sleep subtly. It can monitor brainwave data in real-time and provide feedback, which meets the personalized needs of different users and the intelligent adjustment enhances the user experience.
Concept Model
Product Operation Process
Prototype
Shape-Changing Material Research (Prototype IV)
I made three generations of prototypes.
Prototype I: I pasted electronic components directly onto a nightcap. For the first time, my main goal is to verify whether the function of the product can be realized and whether the effect of the product is effective.
Prototype II: For the second prototype, I begin to explore the product's appearance. I bought a resin 3d printer myself and I printed it out.
Prototype III: For the third prototype, I focused more on the comfort of the product. So I chose to mold a product "container" with silicone. Due to the limited resources at that time, and my hardware skills limitation at that time, the shape of the third generation of products is still relatively large.
The fourth iteration of the design was inspired by the knowledge I learned about smart fabrics from my internship at the Center for Digital Innovation at Tongji University in June 2020. Olfactory feedback wearable devices are currently quite thick and the most common is to use an atomizer plate including my previous three prototypes, which require liquids and is not easy to carry. To solve this, I develop DivcapTex, a smart fabric with a scented structure that can release olfactory feedback through fabric deformation after being energized.
DivcapTex Structure
Experiemtns
Future Potential
DivcapTex can be connected to a brain-computer interface, and according to the different states of a person’s EEG, it can release aroma for interaction. Because of how thin and light DivcapTex is, it opens up a lot of possibilities in designing scenes. However, it still takes a long time for research and development. I believe it illustrates an entirely new form of olfactory interaction and it has the potential to redefine the way we interact.
For the sleep scenario, it no longer needs a complex and massive EEG sensor and it can be directly attached to the face. It can be heated and deformed after receiving certain EEG data to release sleep-promoting fragrances.
Divcap App Design
