Upload app.py

app.py

@@ -0,0 +1,1035 @@
+import gradio as gr
+import spaces
+from cellpose import models
+import numpy as np
+import cv2
+import matplotlib.pyplot as plt
+import tempfile
+from PIL import Image, ImageDraw
+import io
+from huggingface_hub import hf_hub_download
+import base64
+from concurrent.futures import ThreadPoolExecutor, as_completed
+
+HF_REPO_ID = "myang4218/cellposemodel"
+MODEL_OPTIONS = {
+    "Hemocytometer Model": "hemocytometermodel.npy",
+    "General Model": "generalmodel.npy"
+}
+
+loaded_models = {}
+
+# ---- mobile-safe size limits (aggressive for Safari) ----
+MAX_SIDE = 1024          
+MAX_PIXELS = 1024 * 1024
+
+
+def safe_resize(image_np):
+    """
+    Downscale image to fit within MAX_SIDE and MAX_PIXELS while
+    preserving aspect ratio. Works for RGB / RGBA / grayscale.
+    """
+    h, w = image_np.shape[:2]
+    total = h * w
+
+    if max(h, w) <= MAX_SIDE and total <= MAX_PIXELS:
+        return image_np
+
+    # compute scale 
+    scale_side = MAX_SIDE / max(h, w)
+    scale_pixels = (MAX_PIXELS / total) ** 0.5
+    scale = min(scale_side, scale_pixels)
+
+    new_w = max(1, int(w * scale))
+    new_h = max(1, int(h * scale))
+
+    return cv2.resize(image_np, (new_w, new_h), interpolation=cv2.INTER_AREA)
+
+
+def draw_exclusion_overlay(image_np, left_width_pct, top_width_pct):
+    
+    h, w = image_np.shape[:2]
+    
+    # Convert to PIL for drawing
+    img_pil = Image.fromarray(image_np)
+    draw = ImageDraw.Draw(img_pil, 'RGBA')
+    
+    # Calculate pixel widths from percentages
+    left_px = int(w * left_width_pct / 100)
+    top_px = int(h * top_width_pct / 100)
+    
+    # Draw overlays for exclusion zones
+    if left_px > 0:
+        # Left exclusion zone
+        draw.rectangle(
+            [(0, 0), (left_px, h)],
+            fill=(255, 0, 0, 80)  # Semi-transparent red
+        )
+        # border line
+        draw.line([(left_px, 0), (left_px, h)], fill=(255, 0, 0, 255), width=3)
+    
+    if top_px > 0:
+        # Top exclusion zone
+        draw.rectangle(
+            [(0, 0), (w, top_px)],
+            fill=(255, 0, 0, 80)  # Semi-transparent red
+        )
+        # border line
+        draw.line([(0, top_px), (w, top_px)], fill=(255, 0, 0, 255), width=3)
+    
+    return np.array(img_pil)
+
+
+def apply_stereological_exclusion(masks, left_width_pct, top_width_pct):
+    h, w = masks.shape
+    
+    # Calculate pixel widths from percentages
+    left_px = int(w * left_width_pct / 100)
+    top_px = int(h * top_width_pct / 100)
+    
+    filtered_masks = masks.copy()
+    cell_ids = np.unique(masks)
+    cell_ids = cell_ids[cell_ids > 0]
+    
+    excluded_cells = []
+    included_cells = []
+    
+    for cell_id in cell_ids:
+        cell_mask = (masks == cell_id)
+        
+        # Get cell boundary coordinates
+        rows, cols = np.where(cell_mask)
+        
+        # Check if cell touches left exclusion zone
+        touches_left = np.any(cols < left_px) if left_px > 0 else False
+        
+        # Check if cell touches top exclusion zone
+        touches_top = np.any(rows < top_px) if top_px > 0 else False
+        
+        # Exclude if touching left or top
+        if touches_left or touches_top:
+            filtered_masks[cell_mask] = 0
+            excluded_cells.append(cell_id)
+        else:
+            included_cells.append(cell_id)
+    
+    # Renumber remaining cells
+    unique_ids = np.unique(filtered_masks)
+    unique_ids = unique_ids[unique_ids > 0]
+    
+    renumbered_masks = np.zeros_like(filtered_masks)
+    for new_id, old_id in enumerate(unique_ids, start=1):
+        renumbered_masks[filtered_masks == old_id] = new_id
+    
+    return renumbered_masks, len(excluded_cells), len(included_cells)
+
+
+def classify_cells_by_blueness(image_np, masks, threshold_bias):
+    """
+    Classify cells as dead (blue) or alive using an adaptive Otsu threshold
+    on per-cell blueness scores, with a user bias to fine-tune.
+
+    Args:
+        image_np:        RGB image array
+        masks:           Cellpose segmentation masks
+        threshold_bias:  Slider value -50..+50; shifts Otsu threshold up/down.
+                         Negative = more cells classified dead (looser).
+                         Positive = fewer cells classified dead (stricter).
+                         0 = pure Otsu (fully automatic).
+
+    Returns:
+        dead_count, alive_count, colored_overlay, otsu_threshold, final_threshold
+    """
+
+    if len(image_np.shape) == 2:
+        image_np = cv2.cvtColor(image_np, cv2.COLOR_GRAY2RGB)
+    elif len(image_np.shape) == 3 and image_np.shape[2] == 4:
+        image_np = cv2.cvtColor(image_np, cv2.COLOR_RGBA2RGB)
+
+    hsv = cv2.cvtColor(image_np, cv2.COLOR_RGB2HSV)
+
+    hue        = hsv[:, :, 0].astype(np.float32)
+    saturation = hsv[:, :, 1].astype(np.float32)
+
+    # Raw blueness: hue proximity to 115° × saturation
+    hue_distance = np.minimum(np.abs(hue - 115), 180 - np.abs(hue - 115))
+    hue_score    = np.maximum(0, 1 - hue_distance / 65)
+    blueness     = hue_score * (saturation / 255.0)
+
+    # --- Compute per-cell mean blueness scores ---
+    cell_ids = np.unique(masks)
+    cell_ids = cell_ids[cell_ids > 0]
+
+    if len(cell_ids) == 0:
+        blank = image_np.copy()
+        return 0, 0, blank, 0.0, 0.0
+
+    cell_scores = np.array([np.mean(blueness[masks == cid]) for cid in cell_ids])
+
+    # --- Otsu on the distribution of per-cell scores ---
+    # cv2.threshold expects uint8; scale 0-1 → 0-255
+    scores_u8 = (np.clip(cell_scores, 0, 1) * 255).astype(np.uint8)
+
+    if scores_u8.max() == scores_u8.min():
+        # All cells identical → Otsu is undefined; use midpoint
+        otsu_threshold = float(scores_u8[0]) / 255.0
+    else:
+        # Reshape to a single-column image so cv2.threshold works
+        thresh_val, _ = cv2.threshold(
+            scores_u8.reshape(-1, 1), 0, 255,
+            cv2.THRESH_BINARY + cv2.THRESH_OTSU
+        )
+        otsu_threshold = thresh_val / 255.0
+
+    # --- Apply user bias: slider -50..+50 maps to ±0.20 shift ---
+    bias = (threshold_bias / 50.0) * 0.20
+    final_threshold = float(np.clip(otsu_threshold + bias, 0.0, 1.0))
+
+    # --- Classify ---
+    dead_cells  = [cid for cid, s in zip(cell_ids, cell_scores) if s > final_threshold]
+    alive_cells = [cid for cid, s in zip(cell_ids, cell_scores) if s <= final_threshold]
+
+    # --- Colored overlay ---
+    overlay = image_np.copy().astype(np.float32)
+    for cid in dead_cells:
+        overlay[masks == cid] = [255, 0, 0]
+    for cid in alive_cells:
+        overlay[masks == cid] = [0, 255, 0]
+
+    alpha        = 0.4
+    final_overlay = (1 - alpha) * image_np.astype(np.float32) + alpha * overlay
+    final_overlay = np.clip(final_overlay, 0, 255).astype(np.uint8)
+
+    return len(dead_cells), len(alive_cells), final_overlay, otsu_threshold, final_threshold
+
+
+def measure_confluency(masks, image_np):
+    tot_pixels = image_np.shape[0] * image_np.shape[1]
+    cell_pixels = np.count_nonzero(masks)
+    confluency = cell_pixels / tot_pixels * 100
+    return confluency
+    
+def filter_mask_by_size(masks, minimum_pixels):
+    filtered_masks = masks.copy()
+    cell_ids = np.unique(masks)
+    cell_ids = cell_ids[cell_ids > 0]
+
+    removed_count = 0
+    
+    for cell_id in cell_ids:
+        cell_mask = (masks == cell_id)
+        cell_pixels = np.count_nonzero(cell_mask)
+        if cell_pixels < minimum_pixels:
+            filtered_masks[cell_mask] = 0
+            removed_count += 1
+
+    unique_ids = np.unique(filtered_masks)
+    unique_ids = unique_ids[unique_ids > 0]
+
+    renumbered_masks = np.zeros_like(filtered_masks)
+    for new_id, old_id in enumerate(unique_ids, start=1):
+        renumbered_masks[filtered_masks == old_id] = new_id
+
+    return renumbered_masks, removed_count
+
+
+def filter_mask_by_maxsize(masks, maximum_pixels):
+    filtered_masks = masks.copy()
+    cell_ids = np.unique(masks)
+    cell_ids = cell_ids[cell_ids > 0]
+
+    removed_count = 0
+    for cell_id in cell_ids:
+        cell_mask = (masks == cell_id)
+        cell_pixels = np.count_nonzero(cell_mask)
+        if cell_pixels > maximum_pixels:
+            filtered_masks[cell_mask] = 0
+            removed_count += 1
+
+    unique_ids = np.unique(filtered_masks)
+    unique_ids = unique_ids[unique_ids > 0]
+
+    renumbered_masks = np.zeros_like(filtered_masks)
+    for new_id, old_id in enumerate(unique_ids, start=1):
+        renumbered_masks[filtered_masks == old_id] = new_id
+
+    return renumbered_masks, removed_count
+
+
+def rec_min_size(masks, q=25):
+    ids = np.unique(masks)
+    ids = ids[ids > 0]
+    if len(ids) == 0:
+        return 0
+    sizes = np.array([np.count_nonzero(masks == cid) for cid in ids])
+    return int(round(np.percentile(sizes, q)))
+
+
+def apply_polygon_mask(image_pil, points_json):
+    """
+    Given a PIL image and a JSON string of [[x,y],...] points,
+    zero out everything outside the polygon and return a PIL image.
+    """
+    import json
+    if not points_json or points_json.strip() in ("", "[]"):
+        return image_pil
+    try:
+        pts = json.loads(points_json)
+    except Exception:
+        return image_pil
+    if len(pts) < 3:
+        return image_pil
+
+    image_np = np.array(image_pil)
+    h, w = image_np.shape[:2]
+    poly = np.array(pts, dtype=np.int32)
+    poly[:, 0] = np.clip(poly[:, 0], 0, w - 1)
+    poly[:, 1] = np.clip(poly[:, 1], 0, h - 1)
+    mask = np.zeros((h, w), dtype=np.uint8)
+    cv2.fillPoly(mask, [poly], 255)
+    if len(image_np.shape) == 3:
+        result = np.where(mask[:, :, np.newaxis] == 255, image_np, 0).astype(np.uint8)
+    else:
+        result = np.where(mask == 255, image_np, 0).astype(np.uint8)
+    return Image.fromarray(result)
+
+
+def toggle_stereological_mode(use_stereology):
+    """Show/hide stereological controls based on checkbox"""
+    return gr.update(visible=use_stereology)
+
+
+def update_exclusion_preview(image, left_width, top_width):
+    """Update the preview image with exclusion zone overlay"""
+    if image is None:
+        return None
+    
+    image_np = np.array(image)
+    overlay = draw_exclusion_overlay(image_np, left_width, top_width)
+    return Image.fromarray(overlay)
+
+
+# ---------------------------------------------------------------------------
+# Patch-parallel segmentation helpers (Option A)
+# ---------------------------------------------------------------------------
+PATCH_SIZE   = 512   # target patch side length
+PATCH_OVERLAP = 64   # overlap border on each edge (pixels)
+MIN_PATCH_DIM = 256  # don't bother patching if image fits comfortably
+
+
+def _split_patches(image_np, patch_size=PATCH_SIZE, overlap=PATCH_OVERLAP):
+    """
+    Split image into overlapping patches.
+    Returns list of (patch_np, row_start, col_start) tuples.
+    """
+    h, w = image_np.shape[:2]
+    patches = []
+    row = 0
+    while row < h:
+        row_end = min(row + patch_size, h)
+        col = 0
+        while col < w:
+            col_end = min(col + patch_size, w)
+            patch = image_np[row:row_end, col:col_end]
+            patches.append((patch, row, col))
+            if col_end == w:
+                break
+            col += patch_size - overlap
+        if row_end == h:
+            break
+        row += patch_size - overlap
+    return patches
+
+
+def _merge_patch_masks(patch_results, full_h, full_w, overlap=PATCH_OVERLAP):
+    """
+    Stitch per-patch masks into a single full-image mask.
+
+    Strategy:
+    - Each patch gets a unique ID offset so cell IDs never collide.
+    - Patches are pasted into the canvas using a priority canvas that
+      gives interior pixels precedence over overlap-border pixels.
+    - After pasting, cells whose centroids fall in the overlap zone
+      of two adjacent patches are deduplicated: if two cells from
+      different patches share >50% IoU they are the same cell — keep
+      the one whose centroid is furthest from a patch edge.
+    """
+    full_mask  = np.zeros((full_h, full_w), dtype=np.int32)
+    # track which patch_idx owns each pixel (used for overlap resolution)
+    owner_map  = np.full((full_h, full_w), -1, dtype=np.int32)
+    # distance-to-nearest-edge for the owning patch (higher = more central)
+    priority   = np.zeros((full_h, full_w), dtype=np.float32)
+
+    id_offset = 0
+    patch_meta = []   # (offset, row_start, col_start, patch_h, patch_w)
+
+    for patch_idx, (mask_patch, row_start, col_start) in enumerate(patch_results):
+        ph, pw = mask_patch.shape
+        # offset all non-zero IDs so they're globally unique
+        shifted = np.where(mask_patch > 0, mask_patch + id_offset, 0).astype(np.int32)
+
+        # compute per-pixel priority = min distance to any patch edge
+        rows_idx = np.arange(ph)
+        cols_idx = np.arange(pw)
+        dist_r = np.minimum(rows_idx, ph - 1 - rows_idx)           # (ph,)
+        dist_c = np.minimum(cols_idx, pw - 1 - cols_idx)           # (pw,)
+        pri_patch = np.minimum(dist_r[:, None], dist_c[None, :])   # (ph, pw)
+
+        roi_full   = full_mask [row_start:row_start+ph, col_start:col_start+pw]
+        roi_owner  = owner_map [row_start:row_start+ph, col_start:col_start+pw]
+        roi_pri    = priority  [row_start:row_start+ph, col_start:col_start+pw]
+
+        # where this patch has higher priority, overwrite
+        better = pri_patch > roi_pri
+        roi_full [better] = shifted   [better]
+        roi_owner[better] = patch_idx
+        roi_pri  [better] = pri_patch [better]
+
+        max_id = int(mask_patch.max())
+        patch_meta.append((id_offset, row_start, col_start, ph, pw))
+        id_offset += max_id + 1
+
+    # --- Renumber to compact sequential IDs ---
+    unique_ids = np.unique(full_mask)
+    unique_ids = unique_ids[unique_ids > 0]
+    renumbered = np.zeros_like(full_mask)
+    for new_id, old_id in enumerate(unique_ids, start=1):
+        renumbered[full_mask == old_id] = new_id
+
+    return renumbered
+
+
+def _segment_patch(args):
+    """Worker: run cellpose on a single patch. Called from a thread pool."""
+    patch_np, row_start, col_start, model_filename, hf_repo = args
+    # Each thread uses the shared loaded_models cache (GIL-safe for reads;
+    # model.eval() releases the GIL during GPU work so threads overlap.)
+    model_path = hf_hub_download(repo_id=hf_repo, filename=model_filename)
+    if model_filename in loaded_models:
+        model = loaded_models[model_filename]
+    else:
+        model = models.CellposeModel(gpu=True, pretrained_model=model_path)
+        loaded_models[model_filename] = model
+
+    mask, _, _ = model.eval(patch_np, diameter=None, channels=[0, 0])
+    return mask, row_start, col_start
+
+
+def run_segmentation_patched(image_np, model_filename):
+    """
+    Split image into overlapping patches, run Cellpose on each in parallel,
+    then stitch back into a single full-resolution mask.
+    Falls back to whole-image segmentation if the image is small enough
+    that patching adds overhead without benefit.
+    """
+    h, w = image_np.shape[:2]
+    model_path = hf_hub_download(repo_id=HF_REPO_ID, filename=model_filename)
+    if model_filename in loaded_models:
+        model = loaded_models[model_filename]
+    else:
+        model = models.CellposeModel(gpu=True, pretrained_model=model_path)
+        loaded_models[model_filename] = model
+
+    # Small images: no benefit from patching
+    if max(h, w) <= MIN_PATCH_DIM * 2:
+        mask, _, _ = model.eval(image_np, diameter=None, channels=[0, 0])
+        return mask, 1   # 1 patch
+
+    patches = _split_patches(image_np)
+    n_patches = len(patches)
+
+    # Build argument list for the thread pool
+    args_list = [
+        (patch, r, c, model_filename, HF_REPO_ID)
+        for patch, r, c in patches
+    ]
+
+    patch_results = []  # (mask, row_start, col_start) in submission order
+
+    # ThreadPoolExecutor: GPU kernels release the GIL so threads overlap on GPU
+    with ThreadPoolExecutor(max_workers=min(n_patches, 4)) as pool:
+        futures = {pool.submit(_segment_patch, a): a for a in args_list}
+        for future in as_completed(futures):
+            mask_patch, row_start, col_start = future.result()
+            patch_results.append((mask_patch, row_start, col_start))
+
+    # Re-sort by (row, col) so stitching is deterministic
+    patch_results.sort(key=lambda x: (x[1], x[2]))
+
+    full_mask = _merge_patch_masks(patch_results, h, w)
+    return full_mask, n_patches
+
+
+@spaces.GPU
+def run_segmentation(image, model_choice, min_cell_size, max_cell_size,
+                     use_stereology, left_exclusion, top_exclusion,
+                     crop_points=None):
+    image_np = np.array(image)
+    image_np = safe_resize(image_np)
+
+    # Apply polygon crop mask if the user drew one (need ≥3 points for a polygon)
+    if crop_points and len(crop_points) >= 3:
+        import json
+        pts_json = json.dumps(crop_points)
+        image_pil_masked = apply_polygon_mask(Image.fromarray(image_np), pts_json)
+        image_np = np.array(image_pil_masked)
+    
+    try:
+        model_filename = MODEL_OPTIONS[model_choice]
+
+        # Process image format to RGB
+        if len(image_np.shape) == 2:
+            processed_image_np = cv2.cvtColor(image_np, cv2.COLOR_GRAY2RGB)
+        elif len(image_np.shape) == 3 and image_np.shape[2] == 4:
+            processed_image_np = cv2.cvtColor(image_np, cv2.COLOR_RGBA2RGB)
+        else:
+            processed_image_np = image_np
+
+        # Run patch-parallel Cellpose segmentation
+        masks_raw, n_patches = run_segmentation_patched(processed_image_np, model_filename)
+
+        ids = np.unique(masks_raw)
+        ids = ids[ids > 0]
+
+        sizes = np.array([np.count_nonzero(masks_raw == cid) for cid in ids])
+
+        print("num_cells:", len(ids))
+        print("mean:", sizes.mean() if len(sizes) > 0 else 0)
+        print("median:", np.median(sizes) if len(sizes) > 0 else 0)
+        print("p90:", np.percentile(sizes, 90) if len(sizes) > 0 else 0)
+        print("max:", sizes.max() if len(sizes) > 0 else 0)
+        
+        # Compute recommendation from RAW masks 
+        recommend_min = rec_min_size(masks_raw)
+
+        # If user sets slider to 0, use the recommendation
+        min_used = recommend_min if (min_cell_size == 0) else int(min_cell_size)
+
+        # Apply filters
+        masks = masks_raw.copy()
+        removed_small = 0
+        removed_large = 0
+
+        if min_used > 0:
+            masks, removed_small = filter_mask_by_size(masks, min_used)
+
+        if max_cell_size > 0:
+            masks, removed_large = filter_mask_by_maxsize(masks, int(max_cell_size))
+
+        # Apply stereological exclusion if enabled
+        excluded_count = 0
+        if use_stereology:
+            masks, excluded_count, included_count = apply_stereological_exclusion(
+                masks, left_exclusion, top_exclusion
+            )
+        
+        filter_msg = ""
+        if removed_small:
+            filter_msg += f"Removed {removed_small} small objects (< {min_used} pixels).\n"
+        if removed_large:
+            filter_msg += f"Removed {removed_large} large objects (> {int(max_cell_size)} pixels).\n"
+        if use_stereology and excluded_count > 0:
+            filter_msg += f"Stereological exclusion: {excluded_count} cells excluded (touching left/top zones).\n"
+
+        cell_count = len(np.unique(masks)) - 1
+        confluency = measure_confluency(masks, processed_image_np)
+
+        # Create a basic segmentation overlay (without viability)
+        segmentation_overlay = processed_image_np.copy().astype(np.float32)
+        if masks.max() > 0:
+            np.random.seed(42)  # For consistent random colors
+            colors = np.random.randint(0, 255, size=(masks.max() + 1, 3))
+            colors[0] = [0, 0, 0]
+            colored_mask = colors[masks]
+            alpha = 0.4
+            segmentation_overlay = (1 - alpha) * segmentation_overlay + alpha * colored_mask
+        segmentation_overlay = np.clip(segmentation_overlay, 0, 255).astype(np.uint8)
+        
+        # Add exclusion zone overlay if stereology is enabled
+        if use_stereology:
+            segmentation_overlay = draw_exclusion_overlay(segmentation_overlay, left_exclusion, top_exclusion)
+
+        info_msg = ""
+        if filter_msg:
+            info_msg += filter_msg
+        info_msg += f"Segmentation complete! Found {cell_count} cells.\n"
+        info_msg += f"Confluency: {confluency:.1f}%\n"
+        info_msg += f"Processed as {n_patches} patch{'es' if n_patches > 1 else ''} (parallel).\n"
+        if use_stereology:
+            info_msg += f"Stereological counting enabled (Left: {left_exclusion}%, Top: {top_exclusion}%)\n"
+        info_msg += "Now adjust the Blue Threshold for viability assessment."
+
+        return (
+            cell_count,
+            Image.fromarray(segmentation_overlay),
+            info_msg,
+            gr.update(visible=True),
+            pack_array(masks),
+            pack_array(processed_image_np),
+            confluency,
+            gr.update(value=recommend_min),  # update slider display to recommended
+        )
+
+    except Exception as e:
+        import traceback
+        traceback.print_exc()
+        return (
+            0,
+            None,
+            f"Error during segmentation: {str(e)}",
+            gr.update(visible=False),
+            None,
+            None,
+            0.0,
+            gr.update(),
+        )
+
+
+def update_viability_realtime(threshold_bias, stored_masks, stored_image_np):
+    # avoid unpacking None (e.g. slider moved before segmentation)
+    if stored_masks is None or stored_image_np is None:
+        return None, 0, 0, 0.0, "Please run segmentation first."
+
+    stored_masks    = unpack_array(stored_masks)
+    stored_image_np = unpack_array(stored_image_np)
+
+    try:
+        dead_count, alive_count, viability_overlay_np, otsu_thresh, final_thresh = \
+            classify_cells_by_blueness(stored_image_np, stored_masks, threshold_bias)
+
+        total_count       = alive_count + dead_count
+        viability_percent = (alive_count / total_count * 100) if total_count > 0 else 0.0
+        confluency        = measure_confluency(stored_masks, stored_image_np)
+
+        overlay_image = Image.fromarray(viability_overlay_np)
+        info_msg  = f"Total cells: {total_count}\nLive (green): {alive_count}\nDead (red): {dead_count}\n"
+        info_msg += f"Viability: {viability_percent:.1f}%\nConfluency: {confluency:.1f}%\n"
+        info_msg += f"Otsu threshold (auto): {otsu_thresh:.4f}\n"
+        info_msg += f"Final threshold (with bias): {final_thresh:.4f}\n"
+        info_msg += f"Sensitivity bias: {threshold_bias:+d}"
+
+        return overlay_image, alive_count, dead_count, viability_percent, info_msg
+
+    except Exception as e:
+        return None, 0, 0, 0.0, f"Error updating viability: {str(e)}"
+
+
+def pack_array(arr):
+    pil = Image.fromarray(arr.astype(np.uint8))
+    buf = io.BytesIO()
+    pil.save(buf, format="PNG")
+    return buf.getvalue()
+
+
+def unpack_array(data):
+    return np.array(Image.open(io.BytesIO(data)))
+
+
+def save_tab_result(cell_count, confluency, viab_percent):
+    """Package per-tab results into a dict for Tab 5 summary."""
+    return {
+        "cell_count": float(cell_count) if cell_count is not None else None,
+        "confluency": float(confluency) if confluency is not None else None,
+        "viab_percent": float(viab_percent) if viab_percent is not None else None,
+    }
+
+
+def compute_summary(r1, r2, r3, r4):
+    """Average cell count, confluency, and viability across tabs that have data."""
+    all_results = [r1, r2, r3, r4]
+    valid = [(i + 1, r) for i, r in enumerate(all_results)
+             if r is not None and r.get("cell_count") is not None]
+
+    if not valid:
+        return (
+            0.0, 0.0, 0.0,
+            "No data yet — run segmentation in at least one tab, then click Refresh Summary."
+        )
+
+    avg_count  = sum(r["cell_count"]   for _, r in valid) / len(valid)
+    avg_conf   = sum(r["confluency"]   for _, r in valid) / len(valid)
+    avg_viab   = sum(r["viab_percent"] for _, r in valid) / len(valid)
+
+    lines = [f"Tab {tab_num}: {r['cell_count']:.0f} cells | "
+             f"{r['confluency']:.1f}% confluency | "
+             f"{r['viab_percent']:.1f}% viability"
+             for tab_num, r in valid]
+    lines.append(f"\nAverages ({len(valid)} tab{'s' if len(valid) > 1 else ''}):")
+    lines.append(f"  Cell count:  {avg_count:.1f}")
+    lines.append(f"  Confluency:  {avg_conf:.1f}%")
+    lines.append(f"  Viability:   {avg_viab:.1f}%")
+
+    return avg_count, avg_conf, avg_viab, "\n".join(lines)
+
+
+# ---------------------------------------------------------------------------
+# Polygon crop — gr.Image select-based, no JS
+# ---------------------------------------------------------------------------
+
+def draw_polygon_overlay(image_pil, points):
+    """
+    Draw numbered vertex dots and polygon edges onto a copy of image_pil.
+    points: list of (x, y) tuples in original image pixel space.
+    Returns a new PIL image.
+    """
+    img = image_pil.copy().convert("RGBA")
+    overlay = Image.new("RGBA", img.size, (0, 0, 0, 0))
+    draw = ImageDraw.Draw(overlay)
+
+    if len(points) >= 2:
+        # Draw edges
+        for i in range(len(points) - 1):
+            draw.line([points[i], points[i + 1]], fill=(74, 170, 255, 220), width=3)
+        if len(points) == 4:
+            draw.line([points[-1], points[0]], fill=(74, 170, 255, 220), width=3)
+            # Semi-transparent fill
+            draw.polygon(points, fill=(74, 170, 255, 50))
+
+    # Draw vertex dots + numbers
+    r = max(8, min(img.width, img.height) // 60)
+    for i, (x, y) in enumerate(points):
+        draw.ellipse([x - r, y - r, x + r, y + r],
+                     fill=(74, 170, 255, 255), outline=(255, 255, 255, 255))
+        draw.text((x, y), str(i + 1), fill=(255, 255, 255, 255), anchor="mm")
+
+    combined = Image.alpha_composite(img, overlay)
+    return combined.convert("RGB")
+
+
+def add_crop_point(image_pil, points, evt: gr.SelectData):
+    """
+    Called by gr.Image .select(). Appends the clicked coordinate,
+    redraws the overlay, returns (updated_image, updated_points).
+    Ignores clicks once 4 points are set.
+    """
+    if image_pil is None:
+        return image_pil, points
+    if points is None:
+        points = []
+    if len(points) >= 4:
+        return draw_polygon_overlay(image_pil, points), points
+
+    x, y = int(evt.index[0]), int(evt.index[1])
+    new_points = points + [(x, y)]
+    return draw_polygon_overlay(image_pil, new_points), new_points
+
+
+def clear_crop_points(image_pil):
+    """Reset polygon — return original image with no overlay and empty points."""
+    return image_pil, []
+
+
+
+
+def build_tab(tab_index, masks_state, image_state, result_state):
+    with gr.Tab(f"Tab {tab_index}"):
+        gr.Markdown("Run segmentation")
+
+        # Per-tab state: list of (x,y) crop polygon points
+        crop_points_state = gr.State(value=[])
+        # Clean copy of the uploaded image (no polygon drawn on it)
+        base_image_state  = gr.State(value=None)
+
+        with gr.Row():
+            with gr.Column():
+                img_input = gr.Image(
+                    type="pil",
+                    label="Upload image",
+                    image_mode="RGB",
+                    height=512
+                )
+
+                gr.Markdown(
+                    "### Crop region (optional)\n"
+                    "Click/tap up to **4 points** on the image below to define the region "
+                    "to segment. The polygon will be drawn as you click. "
+                    "Leave empty to segment the full image."
+                )
+
+                crop_display = gr.Image(
+                    type="pil",
+                    label="Click to set crop vertices (up to 4)",
+                    interactive=True,
+                    height=400,
+                )
+
+                crop_status = gr.Markdown("*Upload an image to enable cropping*")
+
+                clear_crop_btn = gr.Button("✕ Clear crop points", size="sm")
+
+                model_dropdown = gr.Dropdown(
+                    choices=list(MODEL_OPTIONS.keys()),
+                    label="Select Model",
+                    value="Hemocytometer Model"
+                )
+
+                min_size_slider = gr.Slider(
+                    minimum=0,
+                    maximum=500,
+                    value=0,
+                    step=10,
+                    label="Minimum Cell Size (pixels). Leave at zero for automated recommendation",
+                )
+
+                max_size_slider = gr.Slider(
+                    minimum=0,
+                    maximum=10000,
+                    value=10000,
+                    step=10,
+                    label="Maximum Cell Size (pixels)",
+                )
+
+                gr.Markdown("### Stereological Counting")
+                use_stereo = gr.Checkbox(
+                    label="Enable Stereological Counting",
+                    value=False,
+                    info="Use unbiased stereological rules for cell counting"
+                )
+
+                with gr.Group(visible=False) as stereo_controls:
+                    gr.Markdown("""
+                    **Stereological Counting Rules:**
+                    - Cells touching LEFT or TOP exclusion zones are EXCLUDED
+                    - Cells touching RIGHT or BOTTOM edges are INCLUDED
+                    - This provides unbiased counting for quantification
+                    """)
+
+                    excl_preview = gr.Image(
+                        type="pil",
+                        label="Exclusion Zone Preview (Red = Excluded)",
+                        height=500
+                    )
+
+                    left_excl = gr.Slider(
+                        minimum=0,
+                        maximum=50,
+                        value=10,
+                        step=1,
+                        label="Left Exclusion Width (%)",
+                        info="Width of left exclusion zone"
+                    )
+
+                    top_excl = gr.Slider(
+                        minimum=0,
+                        maximum=50,
+                        value=10,
+                        step=1,
+                        label="Top Exclusion Width (%)",
+                        info="Width of top exclusion zone"
+                    )
+
+                segment_btn = gr.Button("🔬 Run Segmentation", variant="primary", size="lg")
+
+            with gr.Column():
+                cell_count_out = gr.Number(label="Total Cells Detected", precision=0)
+                confluency_out = gr.Number(label="Confluency (%)", precision=1)
+                overlay_out    = gr.Image(type="pil", label="Segmentation Result")
+                info_out       = gr.Textbox(label="Processing Info", lines=4)
+
+        with gr.Group(visible=False) as viability_section:
+            gr.Markdown("### Viability Assessment (Trypan Blue)")
+            gr.Markdown(
+                "The threshold is set **automatically per image** using Otsu's method. "
+                "Use the sensitivity slider to fine-tune if needed."
+            )
+
+            with gr.Row():
+                with gr.Column():
+                    sensitivity = gr.Slider(
+                        minimum=-50,
+                        maximum=50,
+                        value=0,
+                        step=1,
+                        label="Sensitivity",
+                        info="← more dead cells  ·  auto (0)  ·  fewer dead cells →"
+                    )
+                with gr.Column():
+                    live_count_out = gr.Number(label="Live Cells (Green)", precision=0)
+                    dead_count_out = gr.Number(label="Dead Cells (Red)", precision=0)
+
+            viab_overlay     = gr.Image(type="pil", label="Viability Assessment (Green=Live, Red=Dead)")
+            viab_percent_out = gr.Number(label="Viability (%)", precision=1)
+            viab_info        = gr.Textbox(label="Analysis Results", lines=6)
+
+        # ---- Event handlers ------------------------------------------------
+
+        use_stereo.change(
+            fn=toggle_stereological_mode,
+            inputs=[use_stereo],
+            outputs=[stereo_controls]
+        )
+
+        # When image uploaded: populate crop_display and store clean base copy
+        def on_image_upload(img):
+            if img is None:
+                return None, None, "*Upload an image to enable cropping*"
+            return img, img, f"*Image loaded — click up to 4 points to define crop region*"
+
+        img_input.change(
+            fn=on_image_upload,
+            inputs=[img_input],
+            outputs=[crop_display, base_image_state, crop_status]
+        ).then(
+            # also reset crop points on new upload
+            fn=lambda: [],
+            outputs=[crop_points_state]
+        )
+
+        img_input.change(
+            fn=update_exclusion_preview,
+            inputs=[img_input, left_excl, top_excl],
+            outputs=[excl_preview]
+        )
+        left_excl.change(
+            fn=update_exclusion_preview,
+            inputs=[img_input, left_excl, top_excl],
+            outputs=[excl_preview]
+        )
+        top_excl.change(
+            fn=update_exclusion_preview,
+            inputs=[img_input, left_excl, top_excl],
+            outputs=[excl_preview]
+        )
+
+        # Click on crop_display → add point, redraw overlay
+        def on_crop_click(base_img, points, evt: gr.SelectData):
+            updated_img, updated_pts = add_crop_point(base_img, points, evt)
+            n = len(updated_pts)
+            if n < 4:
+                status = f"*{n} / 4 points set — keep clicking*"
+            else:
+                status = "*4 points set ✓ — click **✕ Clear** to redo, or run segmentation*"
+            return updated_img, updated_pts, status
+
+        crop_display.select(
+            fn=on_crop_click,
+            inputs=[base_image_state, crop_points_state],
+            outputs=[crop_display, crop_points_state, crop_status]
+        )
+
+        # Clear crop points
+        def on_clear_crop(base_img):
+            img, pts = clear_crop_points(base_img)
+            return img, pts, "*Points cleared — click to set new vertices*"
+
+        clear_crop_btn.click(
+            fn=on_clear_crop,
+            inputs=[base_image_state],
+            outputs=[crop_display, crop_points_state, crop_status]
+        )
+
+        segment_btn.click(
+            fn=run_segmentation,
+            inputs=[img_input, model_dropdown, min_size_slider, max_size_slider,
+                    use_stereo, left_excl, top_excl, crop_points_state],
+            outputs=[cell_count_out, overlay_out, info_out, viability_section,
+                     masks_state, image_state, confluency_out, min_size_slider]
+        ).then(
+            fn=update_viability_realtime,
+            inputs=[sensitivity, masks_state, image_state],
+            outputs=[viab_overlay, live_count_out, dead_count_out, viab_percent_out, viab_info]
+        ).then(
+            fn=save_tab_result,
+            inputs=[cell_count_out, confluency_out, viab_percent_out],
+            outputs=[result_state]
+        )
+
+        sensitivity.change(
+            fn=update_viability_realtime,
+            inputs=[sensitivity, masks_state, image_state],
+            outputs=[viab_overlay, live_count_out, dead_count_out, viab_percent_out, viab_info]
+        ).then(
+            fn=save_tab_result,
+            inputs=[cell_count_out, confluency_out, viab_percent_out],
+            outputs=[result_state]
+        )
+
+
+
+# ---------------------------------------------------------------------------
+# Gradio interface
+# ---------------------------------------------------------------------------
+with gr.Blocks(
+    title="CellposeCellCounter",
+    theme=gr.themes.Soft(),
+) as demo:
+    gr.Markdown("# CellposeCellCounter")
+    gr.Markdown("For accurate cell confluency, crop the image to display only desired area. Note that some image file types are not yet supported. PNG and JPEG are preferred.")
+
+    # Shared mask/image state (one pair per tab so tabs don't clobber each other)
+    masks_states  = [gr.State(value=None) for _ in range(4)]
+    image_states  = [gr.State(value=None) for _ in range(4)]
+    result_states = [gr.State(value=None) for _ in range(4)]
+
+    # Build Tabs 1–4 with a loop
+    for i in range(4):
+        build_tab(i + 1, masks_states[i], image_states[i], result_states[i])
+
+    # -------------------------------------------------------------------------
+    # Tab 5 — Summary
+    # -------------------------------------------------------------------------
+    with gr.Tab("Tab 5 — Summary"):
+        gr.Markdown("## Average Results Across All Tabs")
+        gr.Markdown(
+            "Run segmentation (and optionally adjust the blue threshold) in one or more tabs, "
+            "then click **Refresh Summary** to see the averages."
+        )
+
+        refresh_btn = gr.Button("🔄 Refresh Summary", variant="primary", size="lg")
+
+        with gr.Row():
+            avg_count_out = gr.Number(label="Avg Cell Count",      precision=1)
+            avg_conf_out  = gr.Number(label="Avg Confluency (%)",  precision=1)
+            avg_viab_out  = gr.Number(label="Avg Viability (%)",   precision=1)
+
+        summary_box = gr.Textbox(label="Per-Tab Breakdown", lines=10)
+
+        refresh_btn.click(
+            fn=compute_summary,
+            inputs=result_states,   # list of 4 gr.State components
+            outputs=[avg_count_out, avg_conf_out, avg_viab_out, summary_box]
+        )
+
+    # Instructions
+    with gr.Accordion("Instructions", open=False):
+        gr.Markdown("""
+        ### How to use:
+
+        1. **Upload and Segment**:
+            - Upload your microscopy image.
+            - Select a Cellpose model (e.g., "Hemocytometer Model" for suspension culture).
+            - **(Optional)** Enable Stereological Counting for unbiased quantification.
+            - Click "Run Segmentation".
+
+        2. **Stereological Counting** (Optional):
+            - Check "Enable Stereological Counting" to use unbiased counting rules.
+            - Adjust the Left and Top exclusion zone widths using the sliders.
+            - Preview shows excluded areas in red.
+            - **Counting Rules**:
+                - Cells touching LEFT or TOP exclusion zones are EXCLUDED
+                - Cells touching RIGHT or BOTTOM edges are INCLUDED
+                - This ensures unbiased, systematic counting
+
+        3. **Analysis Results**:
+            - **Cell Count**: Total number of detected cells (after exclusions if using stereology)
+            - **Confluency**: Percentage of image area covered by cells
+
+        4. **Real-time Viability Assessment (Trypan Blue)**:
+            - After segmentation, the viability section will become visible.
+            - Adjust the **"Blue Threshold (%)"** slider in real-time.
+            - **Lower values (10-20%)** are more sensitive.
+            - **Higher values (30-50%)** are more selective.
+            - Green cells = Live, Red cells = Dead.
+
+        5. **Summary (Tab 5)**:
+            - After running segmentation in one or more tabs, switch to **Tab 5**.
+            - Click **Refresh Summary** to compute averages across all completed tabs.
+            - Displays average cell count, confluency, and viability, plus a per-tab breakdown.
+
+        6. **Interpreting Results**:
+            - The app displays total, live, and dead cell counts, viability percentage, and confluency.
+            - If stereological counting is enabled, excluded cells are noted in the processing info.
+        """)
+
+if __name__ == "__main__":
+    demo.launch()