This is a repository for the hackathon, MedHacks 2020. Read more at https://devpost.com/software/independence_app.
Inspiration
One of the 3 tracks for this year's MedHacks was Aging in Place with Resilience and Resources. At first, our group overlooked this track as we preferred the idea of the other 2. However, after listening to health care professionals at MedHacks and experiencing workshops focussed on aging resiliency, we realized the difficulties dementia patients and their families have to go through. Along with the known need for loved ones to check up on relatives and elderly individuals wanting independence, we agreed to tackle this issue.
What It Does
Through an app, our project, Independence, assesses your loved ones' mental and physical status along with offering reminders for important daily events such as taking medication. Due to possible visual and motor impairments, the application will both read the questions aloud and convert speeches to text. This information will then be added to a health record that can be checked by medical practitioners and be used for diagnosis and tracking of neurocognitive disorders, mood disorders, and injuries. With the creation of a health record, the information is also sent to the primary contact of the elderly patient. This allows loved ones to check up on their family members and have peace of mind without removing the feeling of independence that many people desire. Should the checkup not be completed that day, a message is sent to the primary contact alerting them that something could possibly be wrong.
What We Larned
As the main technical focus was the app, we all learn a great deal about app development. Collaboratively, we learned that rather than constantly brainstorming ideas until we find the perfect idea, it can be more effective to select one good idea and fully commit to it.
How We Built It
To make the app, we employed the software development kit, Flutter, and Dart as our primary language. With Flutter's diversity, we were able to utilize many plugins to create specifications such as text-to-speech and speech-to-text features. To manage the database, we used Firebase, and machine learning was programmed with Python.
Challenges We Ran Into
As the majority of the group members had minimal app development experience, we struggled with the steep learning curve that came with Flutter. Incorporating machine learning was another challenge we faced, as it was difficult to work around natural language machine learning.
Accomplishments that We are Proud of
We are proud of creating a product that will help elderly patients seeking independence. By incorporating a speech-to-text feature, elderly people will have a much easier time using the product if they have certain physical limitations.
What's Next for Independence
We hope to add smarter natural language machine learning so improve the quality and efficiency of our speech-to-text feature. We also aim to lessen the dexterity requirements from our clients for our product usage as many elders suffer from a lack of dexterity control.