updates with biobert

.gitignore

@@ -0,0 +1 @@
+.env

app.py

@@ -1,8 +1,100 @@
 import gradio as gr
+from transformers import AutoTokenizer, AutoModelForTokenClassification
+from transformers import pipeline
+import torch
 
-def greet(name):
-    return "Hello " + name + "!!"
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
-demo.launch()
+# Load BioBERT model for NER (using a medical NER model based on BioBERT)
+# You can replace this with a specific PHI detection model if available
+MODEL_NAME = "dmis-lab/biobert-v1.1"
+# Alternative: Use a medical NER model if available, e.g., "alvaroalon2/biobert_diseases_ner"
 
+# Initialize the NER pipeline
+try:
+    # Try to load a tokenizer and model for token classification
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    # For PHI detection, we'll use a simple approach with the base model
+    # In production, you'd use a fine-tuned model for PHI detection
+    ner_pipeline = pipeline(
+        "token-classification",
+        model=MODEL_NAME,
+        tokenizer=MODEL_NAME,
+        aggregation_strategy="simple",
+    )
+except Exception as e:
+    print(f"Error loading model: {e}")
+    print("Falling back to a simpler approach...")
+    ner_pipeline = None
+
+
+def detect_phi(text: str) -> str:
+    """
+    Detect PHI (Protected Health Information) in the input text using BioBERT.
+
+    Args:
+        text: Input text containing potential PHI
+
+    Returns:
+        Formatted string showing detected entities
+    """
+    if not text or not text.strip():
+        return "Please enter some text to analyze."
+
+    if ner_pipeline is None:
+        return "Model not loaded. Please check the model configuration."
+
+    try:
+        # Run NER on the input text
+        entities = ner_pipeline(text)
+
+        if not entities:
+            return "No entities detected in the text."
+
+        # Format the results
+        result = "**Detected PHI Entities:**\n\n"
+
+        for entity in entities:
+            entity_text = entity.get("word", "")
+            entity_label = entity.get("entity_group", entity.get("label", "UNKNOWN"))
+            confidence = entity.get("score", 0.0)
+
+            result += (
+                f"- **{entity_text}** ({entity_label}) - Confidence: {confidence:.2%}\n"
+            )
+
+        # Also show the original text with highlights
+        result += "\n---\n\n**Original Text:**\n"
+        result += text
+
+        return result
+
+    except Exception as e:
+        return f"Error processing text: {str(e)}"
+
+
+# Create Gradio interface
+demo = gr.Interface(
+    fn=detect_phi,
+    inputs=gr.Textbox(
+        label="PHI Text Input",
+        placeholder="Enter text containing potential PHI (e.g., 'Patient John Doe, age 45, was admitted on 2024-01-15. SSN: 123-45-6789')",
+        lines=5,
+    ),
+    outputs=gr.Markdown(label="PHI Detection Results"),
+    title="BioBERT PHI Detection",
+    description="Enter text containing Protected Health Information (PHI) to detect entities using BioBERT model.",
+    examples=[
+        [
+            "Patient John Smith, age 52, was admitted to Memorial Hospital on January 15, 2024. Contact: [email protected]"
+        ],
+        [
+            "Dr. Sarah Johnson treated patient ID 12345 at the clinic located at 123 Main St, Boston, MA 02101."
+        ],
+        [
+            "The patient's date of birth is 03/15/1975 and their medical record number is MRN-987654."
+        ],
+    ],
+)
+
+if __name__ == "__main__":
+    demo.launch()