Enhanced progression analysis with automatic timepoint organization

app.py

@@ -8,6 +8,7 @@ import io
 from datetime import datetime
 import pandas as pd
 from collections import defaultdict
+import re
 
 # Page configuration
 st.set_page_config(
@@ -195,6 +196,44 @@ def analyze_progression(timepoints_data):
     response = model.generate_content(content)
     return response.text
 
+def extract_date_from_filename(filename):
+    """Extract date from filename using common patterns"""
+    # Common date patterns (add more patterns if needed)
+    patterns = [
+        r'(\d{4}[-_/]\d{2}[-_/]\d{2})',  # YYYY-MM-DD, YYYY_MM_DD
+        r'(\d{2}[-_/]\d{2}[-_/]\d{4})',  # DD-MM-YYYY, DD_MM_YYYY
+        r'(\d{8})',  # YYYYMMDD
+    ]
+    
+    for pattern in patterns:
+        match = re.search(pattern, filename)
+        if match:
+            date_str = match.group(1)
+            try:
+                # Try different date formats
+                for fmt in ['%Y-%m-%d', '%Y_%m_%d', '%d-%m-%Y', '%d_%m_%Y', '%Y%m%d']:
+                    try:
+                        return datetime.strptime(date_str.replace('/', '-'), fmt).strftime('%Y-%m-%d')
+                    except ValueError:
+                        continue
+            except ValueError:
+                continue
+    return None
+
+def organize_scans_by_date(files):
+    """Organize uploaded files by their dates"""
+    organized_files = defaultdict(list)
+    unorganized_files = []
+    
+    for file in files:
+        date = extract_date_from_filename(file.name)
+        if date:
+            organized_files[date].append(file)
+        else:
+            unorganized_files.append(file)
+    
+    return organized_files, unorganized_files
+
 def main():
     # Header
     st.markdown("""
@@ -227,7 +266,7 @@ def main():
 
     # Scan Analysis Section
     st.markdown("### Pentacam Scan Analysis")
-    analysis_type = st.radio("Select Analysis Type", ["Single Timepoint", "Multiple Timepoint Progression"])
+    analysis_type = st.radio("Select Analysis Type", ["Single Timepoint", "Progression Analysis"])
 
     if analysis_type == "Single Timepoint":
         st.markdown('<div class="upload-section">', unsafe_allow_html=True)
@@ -261,54 +300,70 @@ def main():
         
         st.markdown('</div>', unsafe_allow_html=True)
     
-    else:  # Multiple Timepoint Progression
+    else:  # Progression Analysis
         st.markdown('<div class="upload-section">', unsafe_allow_html=True)
-        st.info("Upload scans from multiple timepoints to analyze progression")
+        st.info("""Upload all your Pentacam scans at once. The system will automatically organize them by date and analyze progression.
+                   For best results, ensure your scan filenames include dates (e.g., 'scan_2023-01-15.jpg' or 'pentacam_20230115.png')""")
         
-        # Create a dynamic form for multiple timepoints
-        timepoints_data = defaultdict(list)
-        num_timepoints = st.number_input("Number of Timepoints", min_value=2, max_value=10, value=3)
+        uploaded_files = st.file_uploader(
+            "Upload All Pentacam Scans",
+            type=['png', 'jpg', 'jpeg'],
+            accept_multiple_files=True
+        )
         
-        st.markdown('<div class="timeline-container">', unsafe_allow_html=True)
-        for i in range(num_timepoints):
-            st.markdown(f'<div class="timepoint-card">', unsafe_allow_html=True)
-            col1, col2 = st.columns([1, 3])
-            
-            with col1:
-                date = st.date_input(f"Date - Timepoint {i+1}", key=f"date_{i}")
+        if uploaded_files:
+            organized_files, unorganized_files = organize_scans_by_date(uploaded_files)
             
-            with col2:
-                files = st.file_uploader(
-                    f"Upload Scans - Timepoint {i+1}",
-                    type=['png', 'jpg', 'jpeg'],
-                    accept_multiple_files=True,
-                    key=f"files_{i}"
-                )
+            if organized_files:
+                st.markdown("### Organized Scans by Date")
+                st.markdown('<div class="timeline-container">', unsafe_allow_html=True)
+                
+                timepoints_data = defaultdict(list)
+                dates = sorted(organized_files.keys())
                 
-                if files:
+                for date in dates:
+                    st.markdown(f'<div class="timepoint-card">', unsafe_allow_html=True)
+                    st.markdown(f"#### Timepoint: {date}")
+                    
+                    files = organized_files[date]
                     images = []
-                    for file in files:
-                        image = Image.open(file)
-                        images.append(image)
-                        timepoints_data[date.strftime("%Y-%m-%d")].extend(images)
+                    cols = st.columns(len(files))
                     
-                    if images:
-                        cols = st.columns(len(images))
-                        for idx, (col, img) in enumerate(zip(cols, images)):
-                            with col:
-                                st.image(img, caption=f"Scan {idx + 1}", use_column_width=True)
-            
-            st.markdown('</div>', unsafe_allow_html=True)
-        st.markdown('</div>', unsafe_allow_html=True)
-        
-        if timepoints_data and len(timepoints_data) >= 2:
-            if st.button("Analyze Progression"):
-                with st.spinner("Analyzing progression across timepoints..."):
-                    progression_analysis = analyze_progression(timepoints_data)
-                    st.markdown("### Progression Analysis Results")
-                    st.markdown('<div class="analysis-container">', unsafe_allow_html=True)
-                    st.markdown(progression_analysis)
+                    for idx, (file, col) in enumerate(zip(files, cols)):
+                        with col:
+                            image = Image.open(file)
+                            images.append(image)
+                            st.image(image, caption=f"Scan {idx + 1}", use_column_width=True)
+                    
+                    timepoints_data[date].extend(images)
                     st.markdown('</div>', unsafe_allow_html=True)
+                
+                if unorganized_files:
+                    st.warning(f"{len(unorganized_files)} files couldn't be automatically dated. Please ensure filenames include dates.")
+                    with st.expander("Manually Assign Dates"):
+                        for file in unorganized_files:
+                            col1, col2 = st.columns([2, 1])
+                            with col1:
+                                st.text(file.name)
+                            with col2:
+                                date = st.date_input(f"Date for {file.name}", key=f"manual_{file.name}")
+                                image = Image.open(file)
+                                timepoints_data[date.strftime("%Y-%m-%d")].append(image)
+                
+                if len(timepoints_data) >= 2:
+                    if st.button("Analyze Progression"):
+                        with st.spinner("Analyzing progression across timepoints..."):
+                            progression_analysis = analyze_progression(timepoints_data)
+                            st.markdown("### Progression Analysis Results")
+                            st.markdown('<div class="analysis-container">', unsafe_allow_html=True)
+                            st.markdown(progression_analysis)
+                            st.markdown('</div>', unsafe_allow_html=True)
+                else:
+                    st.warning("Please upload scans from at least 2 different timepoints for progression analysis.")
+                
+                st.markdown('</div>', unsafe_allow_html=True)
+            else:
+                st.error("No dated scans found. Please ensure your filenames include dates (e.g., 'scan_2023-01-15.jpg').")
         
         st.markdown('</div>', unsafe_allow_html=True)