Update app.py

app.py

@@ -173,7 +173,7 @@ def generate_image(prompt,
 
 # Gradio interface
 with gr.Blocks() as demo:
-    gr.Markdown(value="# Multimodal Understanding")
+    gr.Markdown(value="# AutoMultimodal")
     with gr.Row():
         image_input = gr.Image()
         with gr.Column():
@@ -186,66 +186,22 @@ with gr.Blocks() as demo:
     understanding_output = gr.Textbox(label="Response")
 
     examples_inpainting = gr.Examples(
-        label="Multimodal Understanding examples",
+        label="Multimodal Medical Prompts",
         examples=[
             [
-                """Analyze the provided fundus image in exhaustive detail, following the standard ophthalmological protocol for fundus examination.  Output an HTML report structured as a formal medical document.  The report MUST:
-
-1.  **Image Quality Assessment:** Begin with a concise assessment of image quality, noting focus, illumination, field of view, and any artifacts (and their impact on assessability).
-
-2.  **Detailed Clinical Findings:**  Describe each of the following areas with the utmost precision and specificity, using proper ophthalmological terminology:
-    *   **Optic Disc:**
-        *   Size and shape (including any abnormalities).
-        *   Color (specifically noting any pallor and its location).
-        *   Cup-to-Disc Ratio (CDR), providing both vertical and horizontal estimates.
-        *   Neuroretinal Rim:  Assess rim thickness in all quadrants (superior, inferior, nasal, temporal).  Explicitly state whether the ISNT rule is followed or violated.  Describe any notching or focal thinning.
-        *   Peripapillary Region:  Describe the presence/absence of peripapillary atrophy (PPA), differentiating between alpha and beta zones. Note any hemorrhages.
-    *   **Retinal Vasculature:**
-        *   Arterioles:  Describe caliber (narrowing, dilation), tortuosity, and any focal abnormalities.
-        *   Venules:  Describe caliber, tortuosity, and any abnormalities.
-        *   Arteriovenous (A/V) Ratio: Estimate the A/V ratio.
-        *   Crossing Changes:  Note any arteriovenous nicking or other crossing abnormalities.
-        *  Vessel Course: Describe the course of the major vessels, and check for abnormalities.
-    *   **Macula:**
-        *   Foveal Reflex:  Describe the presence/absence and quality of the foveal reflex.
-        *   Pigment Changes: Note any pigmentary abnormalities, drusen, or other lesions.
-        *   Edema/Exudates:  Describe any signs of macular edema or exudates.
-    *   **Peripheral Retina:**
-        *   Mid-Periphery: Describe any abnormalities (hemorrhages, exudates, tears, etc.).
-        *   Far Periphery: Note the extent of visualization and any findings.
-
-3.  **Differential Diagnosis:**  Based solely on the image findings, provide a prioritized differential diagnosis.  Include the most likely diagnosis and any other plausible possibilities.  For each diagnosis, explain the reasoning based on the observed features.
-
-4.  **Diagnostic Confidence:** Indicate the confidence level for the primary diagnosis.  List the key image findings that support the diagnosis.
-
-5.  **Simulated AI Attention Metrics:**  Create a table representing a *simulated* AI attention distribution.  This should reflect the expected focus areas for the most likely diagnosis, based on the known importance of different features.  Provide percentages for:
-    *   Optic Disc (Total)
-        *   Cup
-        *   Neuroretinal Rim (subdivided by region if significant differences exist)
-    *   Peripapillary Atrophy
-    *   Vessels
-    *   Macula
-    *   Periphery
-
-6.  **Summary and Impression:**  Provide a concise summary of the key findings and the overall impression.
-
-7. **Recommendations:**
-     *   Provide specific, actionable recommendations based on the image findings.
-     *    If referral is warranted, clearly state the urgency and the type of specialist.
-     *   List any recommended investigations (e.g., OCT, visual fields).
-
-8.  **Disclaimer:** Include a disclaimer stating that the report is based on image analysis alone and does not replace a full clinical examination.
-
-9. **HTML Structure:**  Use semantic HTML elements (h1-h3, p, ol, ul, table, div) to create a well-structured, readable report.  Include:
-    *  A report header with a title ("EyeUnit.ai | AI for Ophthalmology") and a logo 
-    *  Clearly labeled sections for each part of the analysis.
-    *  Tables for the "Overall Analysis Coverage" and "AI-Driven Attention Metrics."
-
-10. **CSS Styling:**  Apply CSS styles to make the report visually appealing and professional.  The report should be suitable for both screen viewing and printing (use a `@media print` block to optimize for print).
-     * **Crucial Details:**
-     * **PATIENT ID, NAME, AGE and DATE OF EXAM**
-
-11. **Crucial Details:** Output ONLY the complete HTML code. Do not provide any surrounding text or explanations. Focus solely on generating the HTML report.
+                """
+You are an AI assistant trained to analyze medical images. Analyze the attached fundus photograph in extreme detail, following a structured approach analogous to an ophthalmologist's examination. Provide a differential diagnosis solely based on this image analysis, without assuming any clinical history. Structure your response as follows:
+
+Analysis Methodology (Concise): List, very briefly, the key anatomical areas/features you will assess, in the order of assessment (e.g., Optic Disc, Vessels, Macula, Periphery, Overall Quality). Do not describe the analysis process in detail here – just name the areas.
+Detailed Image Analysis & Percentage Breakdown: Analyze the image, addressing each area listed in Step 1. For each area:
+Provide a highly detailed, objective description, using precise ophthalmological terminology. Quantify observations whenever possible (e.g., cup-to-disc ratio, A/V ratio).
+State the percentage of that area you were able to confidently analyze, based on image quality and clarity. For example: "Optic Disc: 90% analyzable (10% obscured by slight blurring at the superior margin)." "Macula: 100% analyzable." "Peripheral Retina (Nasal): 60% analyzable (40% not visible in the image)." Be precise.
+For any areas where analysis is incomplete (<100%), briefly explain the limiting factor (e.g., poor focus, limited field of view, artifact).
+Differential Diagnosis (Image-Based Only): Based solely on your Step 2 analysis, provide:
+Most Likely Diagnosis (from image findings).
+Other Possible Diagnoses (from image findings).
+Rationale: For each diagnosis, briefly link specific image findings to the diagnostic criteria.
+                
 """      ,          "fundus.png",
             ],
         ],
@@ -253,7 +209,7 @@ with gr.Blocks() as demo:
     )
     
         
-    gr.Markdown(value="# Text-to-Image Generation")
+    gr.Markdown(value="# Text-to-Image Multimodal Generation")