Neuro Synca

landing
Model's accuracy
Distribution of data set
User chat history storage
Chat with ADHD Specialist
Chat Selection (Which is easily updatable if the app were to expand to neurosurgeons)
Showing Chat selection options can versatile in health care.
Cloud Panel
Recommended Testing Center
Testing Type and Location

Inspiration

After experiencing homelessness and struggling academically, I often wish someone had hinted toward the neurodiverse challenges I was facing. Being unaware of my ADHD, autism, and dyslexia made it difficult to understand why I felt so out of place. Navigating society, building friendships, and tackling schoolwork was an uphill battle, as I simply processed things differently. If I had someone to talk to—a guide to help me approach situations from another perspective—it could have greatly improved my mental wellbeing.

It wasn’t until I turned 24 that I could finally afford a diagnosis. Growing up poor in Baltimore City with multiple siblings, our focus was on survival: "What are we going to eat tomorrow?" or "Can we make it to school safely without being kidnapped or robbed on the MTA public bus?" Mental health wasn’t a priority in our household. An accessible chat companion tailored to neurodiverse needs could have saved me years of self-doubt and confusion.

Neurohealth in Underserved Communities:
The impact of undiagnosed neurodiverse conditions in underserved communities is staggering. According to the Centers for Disease Control and Prevention (CDC), nearly 1 in 6 children have a developmental disability, yet access to healthcare, particularly for neurodiverse diagnoses, is often lacking in marginalized communities. Studies show that Black and Latino children are diagnosed with autism later than their white counterparts, and nearly 50% of children with mental health issues do not receive the necessary treatment due to a lack of resources. This disparity leaves many individuals navigating neurodiversity without adequate support, which can lead to academic struggles, social isolation, and long-term mental health consequences.

Had there been a low-cost or free, AI-powered chatbot for mental health support available to me, it could have made a tremendous difference. That's why I created Neuro Synca.

What It Does

Neuro Synca is designed to provide neurodiverse individuals, particularly from underserved Black and Latino communities, with an accessible platform for diagnosis and support. The platform enables users to:

Find affordable testing centers for ADHD, autism, and mental health services based on location.
Use a chatbot companion powered by OpenAI, designed to offer guidance and information on neurodiverse conditions.
The chatbot can act as an educational companion, offering advice and resources based on user input.

In the future, I plan to enhance the chatbot to act as an unauthorized virtual doctor by pulling ADHD and autism data from trusted web sources and providing a machine learning-powered recommendation system. This would offer accessible, preliminary healthcare advice to users in need, particularly those struggling to access formal healthcare systems.

How I Built It

APIs Used:
- Google Maps API: Used to help users find testing centers based on their geographic location.
- OpenAI API: The chatbot companion is powered by OpenAI’s GPT model, enabling it to offer intelligent and context-aware responses to user questions.
Random Forest Regressor Model:
To predict the affordability of testing centers, I used a Random Forest Regressor model. The model predicts the cost of testing services based on various input features such as location, center type, and more. However, due to the lack of real-world data, I generated test numbers to simulate this dataset and build the model. This allowed the system to offer a preliminary recommendation for affordable centers.

   from sklearn.ensemble import RandomForestRegressor
   import numpy as np

   # Load the pre-trained model
   model = RandomForestRegressor(n_estimators=100, random_state=42)

   # Train the model on generated test data
   model.fit(X_train, y_train)

   # Predict affordable testing centers
   predictions = model.predict(input_data)

Challenges I Ran Into

One of the primary challenges I faced was sourcing reliable data on testing center prices. Accurate, up-to-date information on mental health testing, especially for ADHD and autism, is hard to find, particularly for underserved communities. To overcome this, I had to generate test data to create the model as there was no collective effort or database available that listed mental health services and their associated costs. This challenge highlighted the need for better resources and data transparency for neurodiverse diagnoses. Also, being a back-end developer, I had a lot of struggles with the UI and front-end aspect of the web app.

Accomplishments That I'm Proud Of

I’m proud of being able to pull together a functional platform that provides essential resources to communities in need. Neuro Synca is more than just a project—it is the manifestation of a lifelong struggle and the desire to create something that could help others who may be experiencing the same challenges I once faced. From integrating APIs to developing the Random Forest Regressor, building this platform felt like the first step toward making neurodiverse diagnosis more accessible.

What I Learned

Data Accessibility Issues: Mental health service information is not easily available or standardized across various locations. Creating a platform that provides users with localized testing services showed me the complexity of health data collection and the need for improvement in this space.
Modeling with Limited Data: Building a Random Forest model based on generated test data allowed me to understand the intricacies of training models with limited or incomplete data and how to adjust the model for better predictions.

What's Next for Neuro Synca

Looking ahead, I envision expanding the platform by creating a collective effort to build a more extensive and accurate database of testing centers and their associated costs. This would help close the gap in access to neurodiverse diagnosis and mental health resources for underserved communities.

Also, I would love for any front-end developer to improve the website UI and design.

Furthermore, I aim to improve the chatbot companion by enhancing its capabilities. Using AI, I want it to act as an unauthorized doctor, where it can pull ADHD and autism-related data from trusted web sources and build a machine learning recommendation system. This system would provide healthcare advice in an accessible format, offering preliminary guidance and directing individuals to appropriate resources based on their needs.