HealthCast
Our project encompasses two distinct models: one predicts diseases based on symptoms, while the other forecasts diabetes using insulin, glucose levels, and blood pressure.
Created on 31st March 2024
•
HealthCast
Our project encompasses two distinct models: one predicts diseases based on symptoms, while the other forecasts diabetes using insulin, glucose levels, and blood pressure.
The problem HealthCast solves
Users can leverage our project for:
Enhanced Disease Diagnosis: Healthcare professionals can utilize the symptom-based disease prediction model to improve the accuracy and speed of disease diagnosis, leading to timely interventions and better patient outcomes.
Diabetes Management: Individuals with diabetes can benefit from the diabetes prediction model, which uses insulin, glucose, and blood pressure data to provide personalized insights for better management of the condition.
Early Intervention: By providing early warnings and insights, our project enables healthcare providers to intervene sooner, potentially preventing complications and improving overall health outcomes.
Tailored Healthcare: The dual-model approach offers tailored insights for each individual, allowing for personalized healthcare management based on specific symptoms and health data.
Challenges we ran into
During the development of our project, we encountered several challenges:
Overfitting of Decision Tree Model: Our initial decision tree model for disease prediction exhibited overfitting, leading to poor generalization on unseen data. To address this, we implemented pruning techniques, which helped reduce the complexity of the tree and improve its performance on test data.
Backend Unprocessable Entity: At times, the backend encountered issues with processing incoming data due to malformed or unexpected requests. We enhanced our data validation and error handling mechanisms on the backend to gracefully handle such cases and provide informative responses to the frontend.
Tracks Applied (1)
AIML
Heva AI