This is a project for DECO3500 - Social & Mobile Computing. It aims at sending timely rescue calls to volunteers when the elderly fall and facilitating volunteers to be instructed by medical staff remotely before they arrive.
Presented by Hyperdash.
You can check our promotional poster on this Miro board
There are two parts of our final prototype, app UI implemented by Figma and userflow simulation implemented by Unity.
Figma prototype shows the full volunteer side UI. Elderly and medical side UI, core features userflow are shown in the Unity scene.
- Experience our prototype here
- Experience our userflow simulation in a Unity scene here
This is a WebGL build hosted on GitHub page. Use Google Chrome or Firefox for the best experience. Fullscreen mode is not recommended since there are UI and performance issue
And we are open for anyone to build upon our project:
For Unity project, these platforms have the best compatibility:
- Unity Hub version: 3.0.0-beta.5
- Unity Editor version: 2020.3.15f2
- Visual Studio for Mac: 8.10.11 (Open on Windows or Linux may generate warnings, but they can be easily fixed)
To work on the Unity project:
- After the download, select "Add project from disk" in Unity Hub
- Select the project folder
- Click "open"
- You are good to go!
-
Elderly living alone may not get a rescue in time when accidents happen.
-
Increasing in the number of elderly people living alone.
- Falls are the leading cause of fatal injuries.
-
It is difficult to detect accidents in elderly people who live alone.
-
Elderly people living in rural areas often spend more time waiting for help.
-
As the rescue time grows, the probability of the elderly having an accident also increases dramatically.
Establish a network between the elderly, the volunteers and the medical staff, so that the volunteers can act as an extension of the medical staff's capacity to provide the elderly with the most timely and professional assistance.
A device is needed to monitor the physical and mental health of elderly people living alone in real time and to minimize rescue time.
It is necessary to observe the changes in volunteers' mindset before they choose to join the volunteer team, during their work in the volunteer team, and when they choose to leave the volunteer team. In addition, the sensitivity of volunteers to money is also important to explore.
Before providing assistance, it is important to find out if the elderly need assistance. What kind of assistance they need and who they would prefer to be helped by.
Although there are now a variety of fall detection devices on the market, but there are still a range of problems such as limited detection range, insensitive edge detection or errors in the detection results due to the difference between the production material and human skin tissue.
Our design is set to provide assistance for first aid volunteers in order to provide professional and in-time rescue to the elderly living alone.
The specific Litrature Research & Concept Design is available on the wiki page. here
The elderly prefer to choose family and friends as first responders. The elderly have the need to be rescued.
Most of the survey participants are willing to give help to the elderly.
The specific Survey is available on the wiki page. here
Our low-fidelity model is focused on facilitating the rescue process and daily first aid learning for volunteers.
Users will provide comments on our low-fidelity model based on 10 areas of questions.
We invited 6 evaluators and gained 18 valuable revised opinions.
-
The process of entering the home is not smooth enough.
-
Lack of return key.
-
Lack of constraint requirement on the username and password setting.
- Lack of ability for seniors to select rescue priority.
The specific Survey is available on the wiki page. here
We made plenty of changes based on the feedback of the Low Fedility Prototype evaluations, some changes include adding “go back” buttons, adding a constraint requirement on the username and password setting. And the design of the elderly side, the elderly or elderly family members can set the priority of rescue.
The evaluation process includes templates for the three evaluation methods, the test results and the conclusions obtained by the evaluation process.
We used the Time-on-Task evaluation method to observe the time users spent on each page of our mid-fidelity prototype, and the Think aloud evaluation method to record user comments during the evaluation process.
-
Since each evaluator has a different focus, they spend different amounts of time in different interfaces.
-
We collected 8 usability recommendations through the Think Aloud evaluation method.
- Define inputs to the walkthrough
- Convene the walkthrough
- Walkthrough the action sequences for each task
- Record critical information
- Revise the interface to fix the problems
18 evaluators were involved in the entire evaluation process. 10 valuable revised opinions were come up with.
We have listed the changes we are going to make in our high-fidelity prototype.
- User interface interaction is not smooth enough.
- Lack of address voice entry function.
- Lack of specific functions for the medical side.
The specific Survey is available on the wiki page. here
We fixed the volunteer interface interaction and the Lack of address voice entry on the elderly side in the high fidelity prototype. In addition, we also designed the medical side. Previously, the medical side would only receive video calls from volunteers, but now the medical side has an exclusive interface to receive rescue alerts, check vitals and get passwords.
To give the user a better understanding of the prototype, Unity scenes were created so that the user could experience the same rescue process in three different capacities, showing interactions that Figma could not achieve, such as verifying the location of the volunteer/medical staff or the automatic page jump when the door is unlocked.
The specific Survey is available on the wiki page. here
- Interface aesthetics problem.
- Lack of classification of the severity level of falls.
- Design a more friendly or silently fall detection system instead of smart watches.
| Member Name | Role |
|---|---|
| Jiacheng(Alex) Ni | Leader (Hi-Fi), Designer, Developer (Unity) |
| Zewei(Eric) Zhu | Designer |
| He(Gavin) Zhu | Designer, Developer (Unity) |
| Yingxin(Evelyn) Liu | Leader (Med), Designer |
| Zhenyu(Evan) Ding | Leader (Lo-Fi), Designer |
| Kai(Daniel) Wang | Minute-taker, Designer |