Ahmed Elmalla
Ahmed Elmalla
Expert Python, Java, and VB.NET Programming Guidance IB Computer Science Tutor AP Computer Science A Tutor Python Programming Tutor Java Programming Tutor Experienced VB.NET Coding Tutor
Ahmed Elmalla

Portfolio

Smart Health Monitoring: AI-Powered Disease Prediction and Remote Patient Care

Healthcare

Smart Health Monitoring: AI-Powered Disease Prediction and Remote Patient Care

Smart Health Monitoring: AI-Driven Disease Prediction and Remote Patient Care

Our mission is to save lives by using AI to predict disease risks early. That’s why we launched the Smart Health Monitoring project. We started by developing a prototype for predicting cardiovascular diseases and are now expanding our focus to include diabetes prediction. The care costs for diabetes can exceed $4,500 annually, while complications can cost an additional $6,000. Our app aims to mitigate these risks by reducing complications for both diabetes and cardiovascular disease patients.

AI-Enhanced Disease Prediction and Prevention

Our app helps identify family members of diabetes patients who may be in the pre-diabetes stage using AI, guiding them toward lifestyle changes and weight management to prevent disease progression. Cardiovascular disease is the most expensive complication of type 2 diabetes, with 86% of type 2 diabetes patients eventually developing cardiovascular issues.

Comprehensive Patient Monitoring with Smart Health Devices

The app connects medical devices, patients, and clinical teams for seamless health management. It includes communication tools for patients and clinicians, as well as storage and retrieval of vital health measurements. The app pairs with the FDA-approved Checkme device, which measures vital signs like blood pressure, pulse rate, body temperature, oxygen levels, and even ECG data. All measurements are automatically sent to registered clinicians or guardians.

Checkme also offers continuous monitoring for heart rate, oxygen levels, and temperature, making it an essential tool for tracking patient health in real-time. The app’s diabetes diary feature allows users to log insulin intake, carbohydrate consumption, meal details, and HbA1c levels, which helps endocrinologists make informed decisions regarding treatment and prescriptions.

 


https://smarthealthmonitoring.com/assets/img/articles/dia_diary.png

 

 

AI Prediction Models for Health Risks

Our AI models can predict cardiovascular risks based on factors such as blood pressure, age, gender, fasting glucose level, smoking status, and alcohol intake. The model was trained on a dataset of 70,000 patients using the Gradient Boosting algorithm, achieving an AUC of 80%. The predictions are processed in the cloud using Flask and Python on Heroku servers. The system also supports pre-diabetes prediction based on BMI, family history, and glucose levels.

 

The following diagram show the whole patient monitoring cycle for diabetes patients
 

Diabetes use case cycle

 

The diagram below show the whole patient monitoring cycle for cardiovascular diseases patients:

 

Cardio use case cycle

 

 

 

 

 

 

 

 

 

 

We made AI models to predict cardiac arrest and we are in the process of linking it to our app. You can

see them on the links below:

1) [High Risk] Cardiovascular diseases Prediction using AI: 
2) [Low Risk] Cardiovascular diseases Prediction using AI: 

 

AI Prediction Work


We have a working prototype inside our smart health app that can predict if a patient is in risk of get a cardiovascular disease based on blood pressure, age, gender, fasting glucose level, smoking status and alcohol intake (see the images below from the app).


We have the knowledge to build AI healthcare prediction models and started working on diabetes prediction models based on BMI and fasting glucose measurements.

The current cardvascular diseases model was trained using a 70,000 patient dataset and our model uses Gradient Boosting algorithm that gave an AUC (Area under the curve) of 80%. The prediction is performed on the cloud, we use Flask framework & python language on Heroku servers so please enable internet on your phone before trying the prediction.

The image on the most right shows the result of a prediction.

 

AI prediction in the App

 

 

 

 

 

 

 

 

 

 

 

 

Key Features of the Smart Health App:

  1. AI-based prediction for cardiovascular diseases.
  2. Integration with Checkme device for monitoring vital signs and ECG.
  3. Usage of NEWS (National Early Warning Score) for early detection of patient deterioration.
  4. Live chat functionality for real-time communication between patients and clinical teams.
  5. Comprehensive diabetes diary for tracking glucose levels, insulin intake, and carbohydrate consumption.
  6. Secure connection between clinicians, guardians, and patients via approval-based requests.
  7. Real-time sharing of patient location and vital signs with authorized parties.
  8. Visualization of HbA1c estimates and blood glucose trends.
  9. Tagging of glucose measurements (e.g., fasting, after meals) for better diagnosis and treatment.

Glossary:

  • Cardiovascular Disease: A condition affecting the heart and blood vessels.
  • Gradient Boosting Algorithm: A machine learning technique that combines weak prediction models to create a strong predictive model.
  • Pre-diabetes: A stage with elevated glucose levels (100 to 125 mg/dL) that precedes type 2 diabetes.
  • Vital Signs: Essential health indicators, including heart rate, blood pressure, oxygen levels, and temperature.
  • Artificial Intelligence (AI): The use of mathematical and statistical models to predict outcomes in healthcare, including disease diagnosis and treatment.
  • ECG (Electrocardiogram): A test that measures the heart’s electrical activity to check for abnormalities.
  • Diabetes: A chronic condition where the body cannot produce enough insulin or properly use it.
  • Remote Patient Monitoring: A method for healthcare providers to monitor and analyze patient conditions remotely.
  • NEWS (National Early Warning Score): A scoring system to assess and monitor patient health using key physiological parameters.

Use Cases and AI Prediction Work:

Our AI models are actively predicting cardiac arrest risks and are being integrated into the app. You can explore our models below:

  1. High-Risk Cardiovascular Disease Prediction using AI
  2. Low-Risk Cardiovascular Disease Prediction using AI

Visualizing Health Data and Predictions

The app provides easy-to-read visualizations of health data and predictions, making it user-friendly for both patients and clinicians. From glucose trends to cardiovascular risk assessments, our Smart Health Monitoring app delivers insights that empower proactive healthcare.

Add Comment