User-facing badges, upload guidelines, remove jargon from header

app.py

@@ -275,27 +275,25 @@ if gallery_data:
 
 with gr.Blocks(title="UNIStainNet -- Virtual IHC Staining") as demo:
 
-    # ── Header (all inline styles for Gradio 6.x compatibility) ──
+    # ── Header ────────────────────────────────────────────────────
     gr.HTML("""
     <div style="text-align:center; padding:1.5rem 1rem 0.5rem 1rem;">
         <h1 style="font-size:1.8rem; font-weight:700; margin-bottom:0.3rem;">UNIStainNet</h1>
         <p style="font-size:1.05rem; color:#555; margin-top:0.2rem;">
-            Foundation-Model-Guided Virtual Staining of H&amp;E to IHC
+            Virtual Immunohistochemistry Staining from H&amp;E
         </p>
     </div>
     <p style="text-align:center; color:#555; font-size:0.95rem; margin-bottom:0.8rem;">
-        Translate H&amp;E histopathology images into immunohistochemistry (IHC) stains
-        for breast cancer biomarkers using a single unified deep learning model.
+        Generate HER2, Ki67, ER, and PR stains from a single H&amp;E breast tissue image
+        using one unified deep learning model.
     </p>
     <div style="display:flex; justify-content:center; gap:0.6rem; flex-wrap:wrap; margin-bottom:1rem;">
         <span style="display:inline-block; padding:0.25rem 0.75rem; border-radius:999px;
-                     font-size:0.8rem; font-weight:600; background:#dce3f9; color:#1a3a8a;">42M Parameters</span>
+                     font-size:0.8rem; font-weight:600; background:#dce3f9; color:#1a3a8a;">Breast Cancer Biomarkers</span>
         <span style="display:inline-block; padding:0.25rem 0.75rem; border-radius:999px;
-                     font-size:0.8rem; font-weight:600; background:#d4edda; color:#155724;">4 IHC Stains</span>
+                     font-size:0.8rem; font-weight:600; background:#d4edda; color:#155724;">HER2 / Ki67 / ER / PR</span>
         <span style="display:inline-block; padding:0.25rem 0.75rem; border-radius:999px;
-                     font-size:0.8rem; font-weight:600; background:#e8d5f5; color:#5b1a8a;">UNI Foundation Model</span>
-        <span style="display:inline-block; padding:0.25rem 0.75rem; border-radius:999px;
-                     font-size:0.8rem; font-weight:600; background:#f5ddc4; color:#7a3b10;">Single Forward Pass</span>
+                     font-size:0.8rem; font-weight:600; background:#e8d5f5; color:#5b1a8a;">One Model, 4 Stains</span>
     </div>
     """)
 
@@ -353,8 +351,16 @@ with gr.Blocks(title="UNIStainNet -- Virtual IHC Staining") as demo:
             )
         else:
             gr.Markdown(
-                "Upload an H&E image and select a target IHC stain. "
-                "The model generates the virtual stain in a single forward pass (~1 second on GPU)."
+                "Upload an H&E image and select a target IHC stain to generate."
+            )
+        with gr.Accordion("Image upload guidelines", open=False):
+            gr.Markdown(
+                "- **Tissue type:** H&E-stained breast cancer tissue\n"
+                "- **Magnification:** 20x recommended (trained on BCI and MIST datasets)\n"
+                "- **Size:** Images are center-cropped and resized to 512x512 internally\n"
+                "- **Format:** PNG, JPEG, or TIFF\n"
+                "- **Best results:** Regions with invasive carcinoma; "
+                "adipose or stromal tissue may produce lower quality output"
             )
         with gr.Row():
             with gr.Column(scale=1):
@@ -390,8 +396,16 @@ with gr.Blocks(title="UNIStainNet -- Virtual IHC Staining") as demo:
             )
         else:
             gr.Markdown(
-                "Generate **all 4 IHC stains** from a single H&E input. "
-                "This demonstrates the unified multi-stain capability of UNIStainNet."
+                "Generate **all 4 IHC stains** from a single H&E input."
+            )
+        with gr.Accordion("Image upload guidelines", open=False):
+            gr.Markdown(
+                "- **Tissue type:** H&E-stained breast cancer tissue\n"
+                "- **Magnification:** 20x recommended (trained on BCI and MIST datasets)\n"
+                "- **Size:** Images are center-cropped and resized to 512x512 internally\n"
+                "- **Format:** PNG, JPEG, or TIFF\n"
+                "- **Best results:** Regions with invasive carcinoma; "
+                "adipose or stromal tissue may produce lower quality output"
             )
         with gr.Row():
             with gr.Column(scale=1):