Update src/streamlit_app.py

src/streamlit_app.py

@@ -9,10 +9,9 @@ import streamlit as st
 import tensorflow as tf
 from tensorflow import keras
 import pydicom
-import matplotlib.pyplot as plt
-
 from fpdf import FPDF
 
+
 # -----------------------------
 # Page config
 # -----------------------------
@@ -27,6 +26,7 @@ st.caption(
     "This tool is for decision support only and does not replace clinical judgment."
 )
 
+
 # -----------------------------
 # Paths / Model Loading
 # -----------------------------
@@ -34,6 +34,7 @@ REPO_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
 MODEL_PATH = os.path.join(REPO_ROOT, "model.keras")
 VERSION_PATH = os.path.join(REPO_ROOT, "model_version.json")  # optional
 
+
 @st.cache_resource
 def load_model():
     if not os.path.exists(MODEL_PATH):
@@ -47,6 +48,7 @@ def load_model():
         m = keras.models.load_model(MODEL_PATH, safe_mode=False)
     return m
 
+
 model = load_model()
 
 # model input details
@@ -54,6 +56,7 @@ input_shape = model.input_shape  # (None, H, W, C)
 img_size = int(input_shape[1]) if input_shape and input_shape[1] else 256
 exp_ch = int(input_shape[-1]) if input_shape and input_shape[-1] else 1
 
+
 def get_model_version():
     if os.path.exists(VERSION_PATH):
         try:
@@ -63,8 +66,33 @@ def get_model_version():
             return "ResNet50_v1"
     return "ResNet50_v1"
 
+
 MODEL_VERSION = get_model_version()
 
+
+# -----------------------------
+# Text safety (PDF + error messages)
+# -----------------------------
+def safe_text(s: str, max_len: int = 200) -> str:
+    if s is None:
+        return ""
+    s = str(s)
+
+    # replace common unicode characters that can break FPDF
+    s = s.replace("–", "-").replace("—", "-").replace("’", "'").replace("“", '"').replace("”", '"')
+
+    # add break opportunities for long tokens (UUIDs / filenames)
+    s = s.replace("-", "- ").replace("_", "_ ").replace("/", "/ ")
+
+    # keep latin-1 safe for default FPDF fonts
+    s = s.encode("latin-1", "replace").decode("latin-1")
+
+    # trim long strings
+    if len(s) > max_len:
+        s = s[:max_len] + "..."
+    return s
+
+
 # -----------------------------
 # Confidence interpretation
 # -----------------------------
@@ -76,6 +104,7 @@ def interpret_confidence(prob: float) -> str:
     else:
         return "High likelihood (>60%)"
 
+
 # -----------------------------
 # DICOM + preprocessing
 # -----------------------------
@@ -89,6 +118,7 @@ def dicom_bytes_to_img(data: bytes) -> np.ndarray:
 
     return img
 
+
 def preprocess(img_2d: np.ndarray) -> np.ndarray:
     # (H,W) -> (1,img_size,img_size,C) float32 0..1
     x = tf.convert_to_tensor(img_2d[..., np.newaxis], dtype=tf.float32)  # (H,W,1)
@@ -104,6 +134,7 @@ def preprocess(img_2d: np.ndarray) -> np.ndarray:
     x = np.expand_dims(x, axis=0)       # (1,img_size,img_size,C)
     return x.astype(np.float32)
 
+
 def predict_prob(x: np.ndarray) -> float:
     pred = model.predict(x, verbose=0)
     if isinstance(pred, (list, tuple)):
@@ -113,6 +144,40 @@ def predict_prob(x: np.ndarray) -> float:
     return max(0.0, min(1.0, prob))
 
 
+# -----------------------------
+# PDF report
+# -----------------------------
+def build_pdf_report(df_ok: pd.DataFrame, threshold: float, model_version: str) -> bytes:
+    pdf = FPDF()
+    pdf.set_auto_page_break(auto=True, margin=12)
+    pdf.add_page()
+
+    pdf.set_font("Helvetica", size=12)
+    w = pdf.w - pdf.l_margin - pdf.r_margin  # effective width
+
+    pdf.cell(0, 8, safe_text("Pneumonia Detection Report"), ln=True)
+    pdf.set_font("Helvetica", size=10)
+    pdf.cell(0, 6, safe_text(f"Generated: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}"), ln=True)
+    pdf.cell(0, 6, safe_text(f"Model Version: {model_version}"), ln=True)
+    pdf.cell(0, 6, safe_text(f"Decision Threshold: {threshold:.2f}"), ln=True)
+    pdf.ln(4)
+
+    for _, row in df_ok.iterrows():
+        lines = [
+            f"File: {row['file_name']}",
+            f"Probability: {float(row['probability']) * 100:.2f}%",
+            f"Confidence: {row['confidence_level']}",
+            f"Prediction: {row['prediction']}",
+        ]
+        for line in lines:
+            pdf.multi_cell(w, 6, safe_text(line))
+        pdf.ln(2)
+
+    out = pdf.output(dest="S")
+    if isinstance(out, str):
+        out = out.encode("latin-1", "ignore")
+    return out
+
 
 # -----------------------------
 # UI
@@ -128,8 +193,6 @@ threshold = st.slider(
     help="If predicted probability is greater than or equal to the threshold, output is Pneumonia. Otherwise Not Pneumonia."
 )
 
-show_gradcam = st.checkbox("Show Grad-CAM heatmap (explainability)", value=True)
-
 st.subheader("Upload Chest X-ray DICOM Files")
 uploaded_files = st.file_uploader(
     "Select one or multiple DICOM files (.dcm)",
@@ -152,7 +215,7 @@ if submit:
     if not uploaded_files:
         st.warning("Please upload at least one DICOM file before submitting.")
     else:
-        # cache bytes once (so we can read multiple times safely)
+        # cache bytes once (so we can read safely)
         file_bytes = {f.name: f.getvalue() for f in uploaded_files}
 
         rows = []
@@ -206,23 +269,6 @@ if submit:
                 f"'{r['prediction']}'."
             )
 
-        # Grad-CAM section
-        if show_gradcam:
-            st.markdown("### Grad-CAM Heatmaps")
-            for name, data in file_bytes.items():
-                try:
-                    img = dicom_bytes_to_img(data)
-                    x = preprocess(img)
-                    heatmap = make_gradcam_heatmap(x)
-                    fig = overlay_heatmap(img, heatmap)
-                    st.write(f"Heatmap for: {name}")
-                    st.pyplot(fig)
-                except Exception as e:
-                    st.warning(
-                        f"Could not generate Grad-CAM for {name}. "
-                        f"Reason: {safe_text(str(e), max_len=160)}"
-                    )
-
         # Downloads
         st.markdown("### Downloads")
 
@@ -239,15 +285,14 @@ if submit:
         if len(df_ok) > 0:
             pdf_bytes = build_pdf_report(df_ok, threshold, MODEL_VERSION)
             st.download_button(
-            "Download PDF Report",
-            data=pdf_bytes,
-            file_name="pneumonia_report.pdf",
-            mime="application/pdf",
-            use_container_width=True
-        )
+                "Download PDF Report",
+                data=pdf_bytes,
+                file_name="pneumonia_report.pdf",
+                mime="application/pdf",
+                use_container_width=True
+            )
         else:
             st.info("PDF report is available only when at least one file is successfully processed.")
-    
 
 st.divider()
 st.caption(