Update app.py

app.py

@@ -17,12 +17,12 @@ import time
 
 st.set_page_config(page_title="Scratch Assay Segmentation", layout="wide")
 
-APP_VERSION = "2.4"
+APP_VERSION = "2.5"
 DEFAULT_IMGSZ = 640
 
 MODEL_OPTIONS = {
-    "2.4": "24.pt",
-    "3.7": "37.pt",
+    "24": "24.pt",
+    "37": "37.pt",
 }
 
 
@@ -97,7 +97,13 @@ def find_or_create_folder(folder_name, parent=None):
     query = f"name='{folder_name}' and mimeType='application/vnd.google-apps.folder' and trashed=false"
     if parent:
         query += f" and '{parent}' in parents"
-    results = drive_service.files().list(q=query, spaces="drive", fields="files(id, name)").execute()
+
+    results = drive_service.files().list(
+        q=query,
+        spaces="drive",
+        fields="files(id, name)",
+    ).execute()
+
     folders = results.get("files", [])
     if folders:
         return folders[0]["id"]
@@ -105,6 +111,7 @@ def find_or_create_folder(folder_name, parent=None):
     file_metadata = {"name": folder_name, "mimeType": "application/vnd.google-apps.folder"}
     if parent:
         file_metadata["parents"] = [parent]
+
     file = drive_service.files().create(body=file_metadata, fields="id").execute()
     return file.get("id")
 
@@ -122,7 +129,7 @@ def process_image(uploaded_file, model, model_confidence, fov_um=None, pixel_siz
         image = Image.open(uploaded_file).convert("RGB")
         image_np = np.array(image)
 
-        width_px, _ = image.size
+        width_px, height_px = image.size
 
         effective_pixel_size_um = None
         if pixel_size_um is not None and pixel_size_um > 0:
@@ -141,6 +148,8 @@ def process_image(uploaded_file, model, model_confidence, fov_um=None, pixel_siz
                 "SemSegmentacao": True,
                 "Exibir": image,
                 "Original": get_image_bytes(image),
+                "Segmentada": None,
+                "Poligono": None,
             }
 
         result = results[0]
@@ -153,6 +162,8 @@ def process_image(uploaded_file, model, model_confidence, fov_um=None, pixel_siz
                 "SemSegmentacao": True,
                 "Exibir": image,
                 "Original": get_image_bytes(image),
+                "Segmentada": None,
+                "Poligono": None,
             }
 
         best_idx = 0
@@ -168,10 +179,12 @@ def process_image(uploaded_file, model, model_confidence, fov_um=None, pixel_siz
                 "SemSegmentacao": True,
                 "Exibir": image,
                 "Original": get_image_bytes(image),
+                "Segmentada": None,
+                "Poligono": None,
             }
 
         points = [
-            {"x": float(x * width_px), "y": float(y * image.height)}
+            {"x": float(x * width_px), "y": float(y * height_px)}
             for x, y in contour_norm
         ]
 
@@ -218,23 +231,30 @@ def process_image(uploaded_file, model, model_confidence, fov_um=None, pixel_siz
         return None
 
 
-def save_feedback(result, avaliacao, observacao):
+def save_feedback(result, avaliacao, observacao, selected_model_label):
     image_name = result["Imagem"]
+    image_base_name = image_name.rsplit(".", 1)[0]
 
     # 1) Sheet
-    sheet.append_row([image_name, avaliacao, observacao])
+    sheet.append_row([image_name, avaliacao, observacao, selected_model_label, APP_VERSION])
 
     # 2) Drive curation
     if avaliacao in ["Acceptable", "Bad", "No segmentation"]:
-        sufixo = "aceitavel" if avaliacao == "Acceptable" else "ruim" if avaliacao == "Bad" else "sem_segmentacao"
+        sufixo = (
+            "aceitavel" if avaliacao == "Acceptable"
+            else "ruim" if avaliacao == "Bad"
+            else "sem_segmentacao"
+        )
+
         parent_folder = find_or_create_folder("Feedback Segmentacoes")
-        subfolder = find_or_create_folder(image_name.replace(".png", ""), parent_folder)
+        model_folder = find_or_create_folder(f"model_{selected_model_label}", parent_folder)
+        subfolder = find_or_create_folder(image_base_name, model_folder)
 
         resized_original = resize_image(result["Exibir"])
         buf = BytesIO()
         resized_original.save(buf, format="PNG")
         buf.seek(0)
-        upload_to_drive(buf, f"original_{sufixo}.png", subfolder)
+        upload_to_drive(buf, f"original_model_{selected_model_label}_v{APP_VERSION}_{sufixo}.png", subfolder)
 
         if avaliacao != "No segmentation" and result.get("Segmentada") and result.get("Poligono"):
             resized_segmented = resize_image(Image.open(BytesIO(result["Segmentada"].getvalue())))
@@ -244,7 +264,11 @@ def save_feedback(result, avaliacao, observacao):
                 buf = BytesIO()
                 img_obj.save(buf, format="PNG")
                 buf.seek(0)
-                upload_to_drive(buf, f"{nome}_{sufixo}.png", subfolder)
+                upload_to_drive(
+                    buf,
+                    f"{nome}_model_{selected_model_label}_v{APP_VERSION}_{sufixo}.png",
+                    subfolder,
+                )
 
 
 def render_metrics(result):
@@ -258,7 +282,7 @@ def render_metrics(result):
         st.markdown(f"- {area_um2:.2f} µm²")
 
 
-def render_feedback_block(result, prefix_key=""):
+def render_feedback_block(result, selected_model_label, prefix_key=""):
     st.markdown("#### Segmentation quality feedback")
     st.caption("User evaluation used for future model refinement.")
 
@@ -273,7 +297,7 @@ def render_feedback_block(result, prefix_key=""):
         key=f"{prefix_key}obs_{result['Imagem']}",
     )
     if st.button("Save feedback", key=f"{prefix_key}btn_{result['Imagem']}"):
-        save_feedback(result, avaliacao, observacao)
+        save_feedback(result, avaliacao, observacao, selected_model_label)
         st.success("Feedback saved successfully.")
 
 
@@ -281,11 +305,10 @@ def render_feedback_block(result, prefix_key=""):
 # Layout / UI
 # =========================
 st.title("Scratch Assay Segmentation Tool")
-st.caption(f"Version {APP_VERSION} · Deep learning–based wound closure segmentation")
+st.caption(f"Platform version {APP_VERSION}")
 
 st.markdown("---")
 
-# Upload block
 st.markdown("### Input")
 col_input_1, col_input_2 = st.columns([2, 1])
 
@@ -297,7 +320,8 @@ with col_input_2:
 
 model = load_model(MODEL_OPTIONS[selected_model_label])
 
-# Advanced settings (collapsed by default)
+st.caption(f"Selected model checkpoint: {selected_model_label}")
+
 with st.expander("⚙️ Advanced Settings", expanded=False):
     model_confidence = st.slider("Model confidence (%)", 20, 100, 80)
     st.markdown(
@@ -352,15 +376,14 @@ if upload_option == "Single image":
             fov_um=fov_um,
             pixel_size_um=pixel_size_um,
         )
+
         if result:
             results.append(result)
 
             st.markdown(f"#### {result['Imagem']}")
 
             if result["SemSegmentacao"]:
-                col = st.columns(1)[0]
-                with col:
-                    st.image(result["Exibir"], caption="Original", use_container_width=True)
+                st.image(result["Exibir"], caption="Original", use_container_width=True)
                 st.warning("No segmentation was detected for this image.")
             else:
                 col1, col2, col3 = st.columns(3)
@@ -382,7 +405,7 @@ if upload_option == "Single image":
                 )
 
             st.markdown("---")
-            render_feedback_block(result, prefix_key="single_")
+            render_feedback_block(result, selected_model_label, prefix_key="single_")
 
 
 # =========================
@@ -408,7 +431,6 @@ elif upload_option == "Image folder":
                 pixel_size_um=pixel_size_um,
             )
 
-        # Local inference is more stable with a single worker.
         with ThreadPoolExecutor(max_workers=1) as executor:
             processed = list(executor.map(process_wrapper, uploaded_files))
 
@@ -442,15 +464,13 @@ elif upload_option == "Image folder":
 
                     render_metrics(result)
 
-                    # Build ZIP
                     zip_file.writestr(f"segmentada_{result['Imagem']}.png", result["Segmentada"].getvalue())
                     zip_file.writestr(f"poligono_{result['Imagem']}.png", result["Poligono"].getvalue())
 
-                render_feedback_block(result, prefix_key="folder_")
+                render_feedback_block(result, selected_model_label, prefix_key="folder_")
 
         zip_images_buffer.seek(0)
 
-        # Summary table + exports
         if results:
             st.markdown("---")
             st.markdown("### Quantitative results")