mohitrajdeo
commited on
Commit
Β·
698a414
1
Parent(s):
b18caa7
Update README.md with detailed project description, features, quick start guide, and application sections
Browse files
README.md
CHANGED
@@ -11,3 +11,353 @@ short_description: This tool provides early prediction and analysis for various
|
|
11 |
---
|
12 |
|
13 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
14 |
+
|
15 |
+
|
16 |
+
---
|
17 |
+
|
18 |
+
<!-- # π©Ί AI-Powered Health & Lifestyle Disease Prediction
|
19 |
+
|
20 |
+
Welcome to the **AI-Powered Health Prediction System**! π
|
21 |
+
|
22 |
+
This tool provides **early prediction and analysis** for various health conditions using **Machine Learning & NLP**. It is designed to assist users in understanding potential health risks based on their lifestyle and symptoms.
|
23 |
+
|
24 |
+
---
|
25 |
+
|
26 |
+
## π₯ Available Features:
|
27 |
+
|
28 |
+
β
**Lifestyle Disease Predictor** (Checkbox-based system using BiomedNLP-PubMedBERT)
|
29 |
+
|
30 |
+
π€ **AI Chatbot for Health Assistance** (Ask health-related questions)
|
31 |
+
|
32 |
+
π§ **Mental Health Assessment** (Analyze sentiment & well-being)
|
33 |
+
|
34 |
+
π©Έ **Disease Predictors:**
|
35 |
+
- Diabetes
|
36 |
+
- Asthma
|
37 |
+
- Stroke
|
38 |
+
- Cardiovascular Disease
|
39 |
+
|
40 |
+
π **Data Visualizer** (Analyze trends in health conditions)
|
41 |
+
|
42 |
+
π **User-friendly Interface** (Easy navigation and interactive elements)
|
43 |
+
|
44 |
+
π **Personalized Health Insights** (Recommendations based on user input)
|
45 |
+
|
46 |
+
π **Select an option from the sidebar to proceed!**
|
47 |
+
|
48 |
+
---
|
49 |
+
|
50 |
+
## π Quick Start Guide
|
51 |
+
|
52 |
+
1. Clone this repository:
|
53 |
+
```bash
|
54 |
+
git clone https://github.com/MOHITRAJDEO12345/early-prediction-for-ml_proj.git
|
55 |
+
```
|
56 |
+
2. Navigate to the project directory:
|
57 |
+
```bash
|
58 |
+
cd early-prediction-for-ml_proj
|
59 |
+
```
|
60 |
+
3. Install dependencies:
|
61 |
+
```bash
|
62 |
+
pip install -r requirements.txt
|
63 |
+
```
|
64 |
+
4. Run the application:
|
65 |
+
```bash
|
66 |
+
streamlit run app.py
|
67 |
+
```
|
68 |
+
|
69 |
+
---
|
70 |
+
|
71 |
+
## π₯ Application Sections
|
72 |
+
|
73 |
+
The application includes the following navigation options:
|
74 |
+
|
75 |
+
```python
|
76 |
+
options = [
|
77 |
+
'Home',
|
78 |
+
'Checkbox-to-disease-predictor',
|
79 |
+
'AI Health Consultant',
|
80 |
+
'Mental-Analysis',
|
81 |
+
'Diabetes Prediction',
|
82 |
+
'Asthma Prediction',
|
83 |
+
'Cardiovascular Disease Prediction',
|
84 |
+
'Stroke Prediction',
|
85 |
+
'Sleep Health Analysis',
|
86 |
+
'Data Visualization',
|
87 |
+
'Text-based Disease Prediction'
|
88 |
+
]
|
89 |
+
```
|
90 |
+
|
91 |
+
### π§ Mental Health Analysis
|
92 |
+
- NOTE: the trained model was not upto mark so we switched to gated transformer model
|
93 |
+
- Uses **mental/mental-roberta-base** for sentiment-based mental health assessment.
|
94 |
+
- Predicts **Depression and Anxiety** based on user input.
|
95 |
+
- Provides graphical risk assessment using **Seaborn & Matplotlib**.
|
96 |
+
|
97 |
+
### π¬ Disease Prediction Models
|
98 |
+
- NOTE: only those diseases have been taken that can be predicted wihtout diagnostic results and some of the features have been discared for training
|
99 |
+
- **Diabetes Model**: Predicts diabetes risk using medical indicators.
|
100 |
+
- **Asthma Model**: Uses preprocessed datasets to detect asthma likelihood.
|
101 |
+
- **Cardiovascular Model**: XGBoost-based prediction for heart disease.
|
102 |
+
- **Stroke Model**: Uses ML models to assess stroke risk factors.
|
103 |
+
|
104 |
+
### π Text-based Disease Prediction
|
105 |
+
- Uses **distilbert-base-uncased** for text-based disease prediction.
|
106 |
+
- Allows users to input symptoms via text or audio.
|
107 |
+
- Predicts possible lifestyle diseases based on user input.
|
108 |
+
- Provides graphical risk assessment using **Seaborn & Matplotlib**.
|
109 |
+
|
110 |
+
---
|
111 |
+
|
112 |
+
## πΈ Screenshots & UI Preview
|
113 |
+
|
114 |
+
π **Streamlit Application Interface:**
|
115 |
+
|
116 |
+
- NOTE: for functionality purpose only
|
117 |
+
- YOUTUBE: https://youtu.be/abrRqceVuDU
|
118 |
+

|
119 |
+
|
120 |
+
π **Data Visualization Example:**
|
121 |
+
- NOTE: currently showing datasets
|
122 |
+
it will be used for visualizing anomalies in user predictions it will become personalized
|
123 |
+

|
124 |
+

|
125 |
+
|
126 |
+
π₯ **Separate Frontend Interface:**
|
127 |
+
- NOTE: the frontend is currently not connected with ml models and it may behave wrongly
|
128 |
+
- WORKING: https://v0.dev/chat/community/lifestyle-disease-prediction-ADp1mOc0hKg
|
129 |
+
- YOUTUBE: https://youtu.be/DU4FW-8hSoU
|
130 |
+

|
131 |
+
|
132 |
+
---
|
133 |
+
|
134 |
+
## β οΈ Disclaimer
|
135 |
+
|
136 |
+
This application has been developed using real-world healthcare datasets sourced from Kaggle:
|
137 |
+
|
138 |
+
- **Stroke Prediction Dataset**
|
139 |
+
- **Asthma Analysis & Prediction Dataset**
|
140 |
+
- **Diabetes Dataset**
|
141 |
+
- **Cardiovascular Disease Dataset**
|
142 |
+
- **Sentiment Analysis for Mental Health**
|
143 |
+
|
144 |
+
The predictions are generated using machine learning models trained on these datasets, incorporating evaluation metrics and graphical insights to enhance interpretability.
|
145 |
+
|
146 |
+
However, this tool has **not undergone clinical validation** and should be used for **informational and educational purposes only**. It is not intended to serve as a substitute for **professional medical diagnosis or treatment**. Always consult a qualified healthcare provider for medical advice.
|
147 |
+
|
148 |
+
---
|
149 |
+
|
150 |
+
# colab
|
151 |
+
- https://colab.research.google.com/drive/1DpOH7KgTWubr5qQjj13EDqxIqsbPLDQe?usp=sharing#scrollTo=EgbDF0U5L1l2
|
152 |
+
- https://colab.research.google.com/drive/1GI7Z1GPPUi67X6UssCQVJXr_QoysfJrz#scrollTo=XkcDpRRzFCIX
|
153 |
+
- https://colab.research.google.com/drive/1eZIBboyY_x0ZsJp5G10XrFFu4aG4eCuf#scrollTo=3NDJOlrEpmoL
|
154 |
+
- http://colab.research.google.com/drive/11KO6cvyTeYY_v5PnYqTwheEupJtNjfCr?usp=sharing#scrollTo=7EyXbXJkPnqf
|
155 |
+
- https://colab.research.google.com/drive/1-B7Q8hXHD0iIBvVldnLkvCiWGhJ2iYNL?usp=sharing
|
156 |
+
- https://colab.research.google.com/drive/1inXO2_JvTw6fOXiJGaW_0pJvI_3sNo0T?usp=sharing
|
157 |
+
- https://colab.research.google.com/drive/1NpwO0NBOKQBtUuN9cC-CXE4vuP5TCavY?usp=sharing
|
158 |
+
- https://colab.research.google.com/drive/10W68SdZHS3IvJAjFTBoqEFI5g7USZVo9?usp=sharing
|
159 |
+
- https://colab.research.google.com/drive/1J8xvEs7rDn0NLYIzH5S2UgFt-lOk7TA6?usp=sharing
|
160 |
+
- https://colab.research.google.com/drive/1BeDmCVjVLb3uqUHdnafgLMLItAtgsAsN?usp=sharing
|
161 |
+
-
|
162 |
+
---
|
163 |
+
|
164 |
+
## π Modular Features (Pending Integration)
|
165 |
+
|
166 |
+
Several functionalities have been implemented but are pending Streamlit integration for optimization:
|
167 |
+
|
168 |
+
β
**User Login & Basic Inputs**: Secure authentication and user profile management.
|
169 |
+
β
**Personalized Email Reports**: Automated daily, weekly, and monthly health insights.
|
170 |
+
β
**Anomaly Visualization**: Analyzes past predictions to detect anomalies.
|
171 |
+
β
**Workout Plans**: AI-driven personalized workout routines based on health data.
|
172 |
+
β
**Sleep Analysis**: AI-powered sleep tracking and recommendations.
|
173 |
+
β
**Medication Adherence**: Reminders and tracking for prescribed medications.
|
174 |
+
β
**Nutrition Recommendations**: AI-based meal planning and dietary suggestions.
|
175 |
+
β
**Community & Resources**: A section for health articles, discussions, and expert Q&A.
|
176 |
+
|
177 |
+
---
|
178 |
+
|
179 |
+
## π¬ Ongoing Research & Future Enhancements
|
180 |
+
|
181 |
+
π§ **Fitbit API Integration** β Real-time health monitoring with wearable devices.
|
182 |
+
π§ **LSTM Models for Realtime Fitbit Data** β Developing deep learning models for dynamic health tracking.
|
183 |
+
π§ **Enhanced Mental Health Analysis** β Exploring transformer-based sentiment models for deeper insights.
|
184 |
+
π§ **Hybrid ML & NLP Systems** β Combining structured health data with unstructured text for more accurate predictions.
|
185 |
+
|
186 |
+
---
|
187 |
+
|
188 |
+
## π¨βπ» Author
|
189 |
+
|
190 |
+
Developed by **Mohit Rajdeo**
|
191 |
+
GitHub: [MOHITRAJDEO12345](https://github.com/MOHITRAJDEO12345)
|
192 |
+
|
193 |
+
---
|
194 |
+
|
195 |
+
## π€ Contributions
|
196 |
+
|
197 |
+
Contributions are always welcome! Feel free to open an issue or submit a pull request if you have suggestions or improvements.
|
198 |
+
|
199 |
+
---
|
200 |
+
|
201 |
+
## π¬ Contact
|
202 |
+
|
203 |
+
For any questions or feedback, feel free to reach out:
|
204 |
+
|
205 |
+
π§ Email: [email protected]
|
206 |
+
π¦ Twitter: [@mohitrajdeo](https://twitter.com/mohitrajdeo) -->
|
207 |
+
|
208 |
+
|
209 |
+
|
210 |
+
# π©Ί Early Prediction of Health & Lifestyle Diseases
|
211 |
+
|
212 |
+
Welcome to the **AI-Powered Health Prediction System**! π
|
213 |
+
|
214 |
+
This tool provides **early prediction and analysis** for various health conditions using **Machine Learning & NLP**. It assists users in understanding potential health risks based on their lifestyle, medical indicators, and symptoms.
|
215 |
+
|
216 |
+
---
|
217 |
+
|
218 |
+
## π₯ Available Features:
|
219 |
+
|
220 |
+
β
**Diabetes Prediction** β Predict diabetes risk using medical indicators.
|
221 |
+
|
222 |
+
β
**Hypertension Prediction** β Assess the risk of high blood pressure.
|
223 |
+
|
224 |
+
β
**Cardiovascular Disease Prediction** β XGBoost-based prediction for heart disease.
|
225 |
+
|
226 |
+
β
**Stroke Prediction** β Machine Learning-based stroke risk analysis.
|
227 |
+
|
228 |
+
β
**Asthma Prediction** β Detect asthma likelihood using preprocessed datasets.
|
229 |
+
|
230 |
+
β
**Sleep Health Analysis** β AI-driven analysis of sleep patterns and health.
|
231 |
+
|
232 |
+
β
**Mental Health Assessment** β Sentiment-based analysis using **mental-roberta-base**.
|
233 |
+
|
234 |
+
β
**Medical Consultant AI Chatbot** β Ask health-related questions for AI-driven insights.
|
235 |
+
|
236 |
+
β
**Data Visualization** β Graphical representation of health trends and anomalies.
|
237 |
+
|
238 |
+
π **Select an option from the sidebar to proceed!**
|
239 |
+
|
240 |
+
---
|
241 |
+
|
242 |
+
## π Quick Start Guide
|
243 |
+
|
244 |
+
1. Clone this repository:
|
245 |
+
```bash
|
246 |
+
git clone https://github.com/MOHITRAJDEO12345/early-prediction-for-ml_proj.git
|
247 |
+
```
|
248 |
+
2. Navigate to the project directory:
|
249 |
+
```bash
|
250 |
+
cd early-prediction-for-ml_proj
|
251 |
+
```
|
252 |
+
3. Install dependencies:
|
253 |
+
```bash
|
254 |
+
pip install -r requirements.txt
|
255 |
+
```
|
256 |
+
4. Run the application:
|
257 |
+
```bash
|
258 |
+
streamlit run app.py
|
259 |
+
```
|
260 |
+
|
261 |
+
---
|
262 |
+
|
263 |
+
## π₯ Application Sections
|
264 |
+
|
265 |
+
The application includes the following navigation options:
|
266 |
+
|
267 |
+
```python
|
268 |
+
options = [
|
269 |
+
'Home',
|
270 |
+
'Diabetes Prediction',
|
271 |
+
'Hypertension Prediction',
|
272 |
+
'Cardiovascular Disease Prediction',
|
273 |
+
'Stroke Prediction',
|
274 |
+
'Asthma Prediction',
|
275 |
+
'Sleep Health Analysis',
|
276 |
+
'Mental-Analysis',
|
277 |
+
'Medical Consultant',
|
278 |
+
'Data Visualization'
|
279 |
+
]
|
280 |
+
```
|
281 |
+
|
282 |
+
### π§ Mental Health Analysis
|
283 |
+
- Uses **mental/mental-roberta-base** for sentiment-based mental health assessment.
|
284 |
+
- Predicts **Depression and Anxiety** based on user input.
|
285 |
+
- Provides graphical risk assessment using **Seaborn & Matplotlib**.
|
286 |
+
|
287 |
+
### π¬ Disease Prediction Models
|
288 |
+
- **Diabetes Model**: Predicts diabetes risk based on medical data.
|
289 |
+
- **Hypertension Model**: Evaluates high blood pressure risk.
|
290 |
+
- **Cardiovascular Model**: Uses XGBoost for heart disease prediction.
|
291 |
+
- **Stroke Model**: ML-based assessment of stroke risk factors.
|
292 |
+
- **Asthma Model**: Machine learning model for asthma detection.
|
293 |
+
|
294 |
+
### π Data Visualization
|
295 |
+
- Interactive graphs to analyze health trends.
|
296 |
+
- Anomaly detection for user predictions.
|
297 |
+
|
298 |
+
### π€ AI Medical Consultant
|
299 |
+
- AI-powered chatbot for answering health-related queries.
|
300 |
+
- Uses NLP models for better understanding and recommendations.
|
301 |
+
|
302 |
+
---
|
303 |
+
|
304 |
+
## πΈ Screenshots & UI Preview
|
305 |
+
|
306 |
+
π **Streamlit Application Interface:**
|
307 |
+

|
308 |
+
|
309 |
+
π **Data Visualization Example:**
|
310 |
+

|
311 |
+

|
312 |
+
|
313 |
+
π₯ **Separate Frontend Interface:**
|
314 |
+

|
315 |
+
|
316 |
+
---
|
317 |
+
|
318 |
+
## β οΈ Disclaimer
|
319 |
+
|
320 |
+
This application has been developed using real-world healthcare datasets sourced from Kaggle:
|
321 |
+
|
322 |
+
- **Diabetes Dataset**
|
323 |
+
- **Hypertension Dataset**
|
324 |
+
- **Cardiovascular Disease Dataset**
|
325 |
+
- **Stroke Prediction Dataset**
|
326 |
+
- **Asthma Analysis & Prediction Dataset**
|
327 |
+
- **Sentiment Analysis for Mental Health**
|
328 |
+
|
329 |
+
The predictions are generated using machine learning models trained on these datasets, incorporating evaluation metrics and graphical insights to enhance interpretability.
|
330 |
+
|
331 |
+
However, this tool has **not undergone clinical validation** and should be used for **informational and educational purposes only**. It is not intended to serve as a substitute for **professional medical diagnosis or treatment**. Always consult a qualified healthcare provider for medical advice.
|
332 |
+
|
333 |
+
---
|
334 |
+
|
335 |
+
# π¬ Ongoing Research & Future Enhancements
|
336 |
+
|
337 |
+
π§ **Fitbit API Integration** β Real-time health monitoring with wearable devices.
|
338 |
+
π§ **LSTM Models for Realtime Fitbit Data** β Developing deep learning models for dynamic health tracking.
|
339 |
+
π§ **Enhanced Mental Health Analysis** β Exploring transformer-based sentiment models for deeper insights.
|
340 |
+
π§ **Hybrid ML & NLP Systems** β Combining structured health data with unstructured text for more accurate predictions.
|
341 |
+
|
342 |
+
---
|
343 |
+
|
344 |
+
## π¨βπ» Author
|
345 |
+
|
346 |
+
Developed by **Mohit Rajdeo**
|
347 |
+
GitHub: [MOHITRAJDEO12345](https://github.com/MOHITRAJDEO12345)
|
348 |
+
|
349 |
+
---
|
350 |
+
|
351 |
+
## π€ Contributions
|
352 |
+
|
353 |
+
Contributions are always welcome! Feel free to open an issue or submit a pull request if you have suggestions or improvements.
|
354 |
+
|
355 |
+
---
|
356 |
+
|
357 |
+
## π¬ Contact
|
358 |
+
|
359 |
+
For any questions or feedback, feel free to reach out:
|
360 |
+
|
361 |
+
π§ Email: [email protected]
|
362 |
+
π¦ Twitter: [@mohitrajdeo](https://twitter.com/mohitrajdeo)
|
363 |
+
|