ML Based Mental Health Diagnosis & Consulting App
MindCrafter is a mental health diagnosis application that uses data from android smartphones, smartwatches that automatically measure patterns of behavior. Real-time smart devices data measurement provides a complete picture of behavioral functioning. This helps MindCrafter to provide Dynamic AI tailored diagnosis and consulting to each person.
I am primarily focusing on the five most important mental issues. 2 out of 5 adults have these issues. I'll be using Smart devices with Android OS to collect behavioral data like sleep times & heart rate of the user.
| Physical Symtoms | Data Collection Methods |
|---|---|
| Pelpetations (Sudden Heart Rate Fluctuations) | Heart Rate Monitoring using SmartWatch or Facial Recognition |
| Sudden Fluctuations in Body Tempreture | Body Temperature Monitoring using SmartWatch or device's built-in temperature sensor |
| Involuntary Tremors in body | Symptom Analysis Questioning using Application Notifications |
| Voluntary Tremors | Voluntary Leg or Hand Shaking using Motion sensors |
| Issues with Stomach (Digestion) | Symptom Analysis Questioning using Application Notifications |
| Physical Symtoms | Data Collection Methods |
|---|---|
| Loose of Interest in Social & Entertainment Activities | Relaxation Time Time Spend on Social & Entertainment Media |
| Excessive Guilt Feelings | Therapy Session Analysis using Therapy Questioning in Application |
| Sudden Low Physical work & Excersize | Meditation, Yoga & Physical Activity Tracker using Application & SmartWatch |
| Sleep Abnormalities (Early Wakeups & Disturbed Sleep Patterns) | Sleep Tracking using SmartWatch & Device usage |
| Poor Concentration | Symptom Analysis Questioning using Application Notifications |
This type of disorder can be analyzed with mania & Depression together. Because of this mental issue. The user always switch between Mania Phase & Depression Phase for 7 t0 15 days for each Phase.
| Physical Symtoms | Data Collection Methods |
|---|---|
| Sudden Excessive Interest in Social & Entertainment Activities | Activity Time Time Spend on Social & Entertainment Media |
| Excessive Happiness Feelings | Therapy Session Analysis using Therapy Questioning in Application |
| Sudden increase in Energy Levels, Physical work & Excersize | Meditation, Yoga & Physical Activity Tracker using Application & SmartWatch |
| Very fewer Sleep Patterns with high energy levels | Sleep Tracking using SmartWatch & Device usage |
| Grandeur Feeling | Symptom Analysis Questioning using Application Notifications |
Each user generates more than 100+ behavioral data points each day. Which can help MindCrafter to analyze the situation of the user. Today users are digitizing more personal data than ever before.
| Product | Features |
|---|---|
| Android Application | Consulting ChatBot |
| Medidation Guide (Full) | |
| Daily Yoga Scheduling | |
| Emotional State Monitoring | |
| Google Assistant (Service) | Consulting ChatBot |
| Guided Meditation Session(Quick) | |
| Daily Yoga Scheduling |
| Frameworks | Purpose |
|---|---|
| Flutter | Android Application Development |
| Firebase | Performance & Monitoring |
| Tensorflow Lite | On Device tailored ML Model |
| Dialogflow | Consulting Bot |
| Google API's | Youtube Videos & Music |
- I've got a very limited understanding of feature extraction from behavioral data for ML model training that is different for each user. Google could help me with a better understanding of the Android OS & Data Collection methods that I can use.
- Google could help with Google API's integration like Youtube API & Music API.
- This application could impact thousands of people but with Google, it can impact millions of people who don't even know that they need help.
On-device APIs can process your data quickly and work even when there’s no network connection. Cloud-based APIs, on the other hand, leverage the power of Google Cloud Platform's machine learning technology to give you an even higher level of accuracy.
| On Device ML Kit API's | Use Cases (examples) |
|---|---|
| Text recognition | Language Processing & Conversation Engine |
| Face detection | Facial Emotion & Heart rate Tracking |
| Object detection & tracking | AR Based Meditation Guide |
| Smart Reply | ChatBot Reply Helper |
| AutoML model inference | On-device ML inference |
| Custom model inference | On-device ML inference |
| Start Date | End date | Milestones |
|---|---|---|
| 2 April | 29 April | Study about digital diagnosis |
| 1 May | 18 May | Planning about ML integration |
| 19 May | 2 June | Analysis Modelling |
| 3 June | 23 June | Development (Templates & Operations) |
| 24 June | 3 July | Tensorflow lite Integration |
| 4 July | 20 July | Machine Learning Training & Debugging |
| 23 July | 3 August | Functional Testing |
| 4 August | 20 August | User Interface Designing |
| 21 August | 31 August | User Experience Testing |
| Final Deployment 1 September | Deployment & Testing |
1 in 5 U.S. adults experience mental illness each year
1 in 25 U.S. adults experience serious mental illness each year
1 in 6 U.S. youth aged 6-17 experience a mental health disorder each year
Suicide is the 2nd leading cause of death among people aged 10-34
There is only one psychiatrist per 100 000 people
- More Than 100 Behavioural Datapoints per user/day
- Diagnosis Test Results
- Consulting
- Tailored Machine Learning Model
Statistics References
https://www.nami.org/learn-more/mental-health-by-the-numbers
https://www.who.int/whr/2001/media_centre/press_release/en/