napari-tmidas 0.1.9__py3-none-any.whl → 0.2.1__py3-none-any.whl

napari_tmidas/processing_functions/trackastra_tracking.py

@@ -0,0 +1,303 @@
+#!/usr/bin/env python3
+"""
+TrackAstra Cell Tracking Module for napari-tmidas
+
+This module integrates TrackAstra deep learning-based cell tracking into the
+napari-tmidas batch processing framework. It uses a dedicated conda environment
+to manage TrackAstra dependencies separately from the main environment.
+"""
+
+import os
+import shutil
+import subprocess
+from pathlib import Path
+
+import numpy as np
+from skimage.io import imread
+
+# Add the registry import
+from napari_tmidas._registry import BatchProcessingRegistry
+
+
+class TrackAstraEnvManager:
+    """Manages the TrackAstra conda environment."""
+
+    @staticmethod
+    def get_conda_cmd():
+        """Get the conda/mamba command available on the system."""
+        # Try mamba first (faster)
+        if shutil.which("mamba"):
+            return "mamba"
+        elif shutil.which("conda"):
+            return "conda"
+        else:
+            raise RuntimeError(
+                "Neither conda nor mamba found. Please install Anaconda/Miniconda/Miniforge."
+            )
+
+    @staticmethod
+    def check_env_exists():
+        conda_cmd = TrackAstraEnvManager.get_conda_cmd()
+        try:
+            # Try running python --version in the env
+            result = subprocess.run(
+                [conda_cmd, "run", "-n", "trackastra", "python", "--version"],
+                capture_output=True,
+                text=True,
+                timeout=10,
+            )
+            return result.returncode == 0
+        except (subprocess.TimeoutExpired, subprocess.CalledProcessError):
+            return False
+
+    @staticmethod
+    def create_env():
+        """Create the TrackAstra conda environment if it doesn't exist."""
+        if TrackAstraEnvManager.check_env_exists():
+            print("TrackAstra environment already exists.")
+            return True
+
+        print("Creating TrackAstra conda environment...")
+        conda_cmd = TrackAstraEnvManager.get_conda_cmd()
+
+        # Create environment with Python 3.10 (required for TrackAstra)
+        env_create_cmd = [
+            conda_cmd,
+            "create",
+            "-n",
+            "trackastra",
+            "python=3.10",
+            "--no-default-packages",
+            "-y",
+        ]
+
+        try:
+            subprocess.run(env_create_cmd, check=True)
+
+            # Install ilpy first from conda-forge
+            ilpy_cmd = [
+                conda_cmd,
+                "install",
+                "-n",
+                "trackastra",
+                "-c",
+                "conda-forge",
+                "-c",
+                "gurobi",
+                "-c",
+                "funkelab",
+                "ilpy",
+                "-y",
+            ]
+            subprocess.run(ilpy_cmd, check=True)
+
+            # Install TrackAstra and other dependencies via pip
+            pip_packages = [
+                "trackastra[napari]",
+                "scikit-image",
+                "tifffile",
+                "torch",
+                "torchvision",
+            ]
+
+            pip_cmd = [
+                conda_cmd,
+                "run",
+                "-n",
+                "trackastra",
+                "pip",
+                "install",
+            ] + pip_packages
+
+            subprocess.run(pip_cmd, check=True)
+
+            print("TrackAstra environment created successfully!")
+            return True
+
+        except subprocess.CalledProcessError as e:
+            print(f"Error creating TrackAstra environment: {e}")
+            return False
+
+
+def create_trackastra_script(img_path, mask_path, model, mode, output_path):
+    """Create a Python script to run TrackAstra in the dedicated environment."""
+    script_content = f"""
+import sys
+import numpy as np
+from skimage.io import imread
+from tifffile import imwrite
+import torch
+from trackastra.model import Trackastra
+from trackastra.tracking import graph_to_ctc, graph_to_napari_tracks
+
+
+# Load images
+print('Loading images...')
+img = imread('{img_path}')
+mask = imread('{mask_path}')
+print(f'Img shape: {{img.shape}}, Mask shape: {{mask.shape}}')
+
+
+# Validate dimensions
+if mask.ndim not in [3, 4]:
+    raise ValueError(f'Expected 3D (TYX) or 4D (TZYX) mask, got {{mask.ndim}}D')
+
+if mask.shape[0] < 2:
+    raise ValueError(f'Need at least 2 timepoints, got {{mask.shape[0]}}')
+
+model = Trackastra.from_pretrained('{model}', device="automatic")
+track_graph = model.track(img, mask, mode='{mode}')
+_, masks_tracked = graph_to_ctc(track_graph, mask, outdir=None)
+
+# Save the tracked masks
+imwrite('{output_path}', masks_tracked.astype(np.uint32), compression='zlib')
+print(f'Saved tracked masks to: {output_path}')
+
+"""
+
+    return script_content
+
+
+@BatchProcessingRegistry.register(
+    name="Track Cells with Trackastra",
+    suffix="_tracked",
+    description="Track cells across time using TrackAstra deep learning (expects TYX or TZYX label images)",
+    parameters={
+        "model": {
+            "type": str,
+            "default": "ctc",
+            "description": "general_2d (nuclei/cells/particles) or ctc (Cell Tracking Challenge; 2D/3D)",
+        },
+        "mode": {
+            "type": str,
+            "default": "greedy",
+            "description": "greedy (fast), ilp (accurate with divisions), greedy_nodiv",
+        },
+        "label_pattern": {
+            "type": str,
+            "default": "_labels.tif",
+            "description": " ",
+        },
+    },
+)
+def trackastra_tracking(
+    image: np.ndarray,
+    model: str = "ctc",
+    mode: str = "greedy",
+    label_pattern: str = "_labels.tif",
+) -> np.ndarray:
+    """
+    Track cells in time-lapse label images using TrackAstra.
+
+    This function takes a time series of segmentation masks and performs
+    automatic cell tracking using TrackAstra deep learning framework.
+
+    Expected input dimensions:
+    - TYX: Time series of 2D label images
+    - TZYX: Time series of 3D label images (will process each Z-slice separately)
+
+    Parameters:
+    -----------
+    image : np.ndarray
+        Input label image array with time as first dimension
+    model : str
+        TrackAstra model: 'general_2d' or 'ctc' (default: "ctc")
+    mode : str
+        Tracking mode: 'greedy', 'ilp', or 'greedy_nodiv' (default: "greedy")
+    label_pattern : str
+        To identify label images
+
+    Returns:
+    --------
+    np.ndarray
+        Tracked label image with consistent IDs across time
+    """
+    print(f"Input shape: {image.shape}, dtype: {image.dtype}")
+
+    # Validate input
+    if image.ndim < 3:
+        print(
+            "Input is not a time series (needs at least 3 dimensions). Returning unchanged."
+        )
+        return image
+
+    if image.shape[0] < 2:
+        print(
+            "Input has only one timepoint. Need at least 2 for tracking. Returning unchanged."
+        )
+        return image
+
+    # Ensure TrackAstra environment exists
+    if not TrackAstraEnvManager.check_env_exists():
+        print("TrackAstra environment not found. Creating it now...")
+        if not TrackAstraEnvManager.create_env():
+            print(
+                "Failed to create TrackAstra environment. Returning unchanged."
+            )
+            return image
+
+    # Get the current file path from the processing context
+    import inspect
+
+    img_path = None
+
+    for frame_info in inspect.stack():
+        frame_locals = frame_info.frame.f_locals
+        if "filepath" in frame_locals:
+            img_path = frame_locals["filepath"]
+            break
+
+    if img_path is None:
+        print("Could not determine input file path. Returning unchanged.")
+
+    temp_dir = Path(os.path.dirname(img_path))
+
+    # Save the mask data
+    mask_path = img_path.replace(".tif", "_labels.tif")
+    # Create the tracking script
+    script_path = temp_dir / "run_tracking.py"
+    output_path = temp_dir / os.path.basename(img_path).replace(
+        ".tif", "_tracked.tif"
+    )
+
+    script_content = create_trackastra_script(
+        str(img_path), str(mask_path), model, mode, str(output_path)
+    )
+
+    with open(script_path, "w") as f:
+        f.write(script_content)
+
+    if label_pattern in img_path:
+        pass
+    else:
+        # Run TrackAstra in the dedicated environment
+        conda_cmd = TrackAstraEnvManager.get_conda_cmd()
+        cmd = [
+            conda_cmd,
+            "run",
+            "-n",
+            "trackastra",
+            "python",
+            str(script_path),
+        ]
+        print(f"Running TrackAstra with model='{model}', mode='{mode}'...")
+        result = subprocess.run(cmd, capture_output=True, text=True)
+
+        if result.returncode != 0:
+            print("TrackAstra error:")
+            print(result.stdout)
+            print(result.stderr)
+            print("Returning original image unchanged.")
+            return image
+
+        print(result.stdout)
+
+    # Load and return the tracked result
+    if output_path.exists():
+        tracked = imread(str(output_path))
+        print(f"Tracking completed. Output shape: {tracked.shape}")
+        os.remove(script_path)
+        return tracked
+    else:
+        print("TrackAstra did not produce output. Returning unchanged.")
+        return image

{napari_tmidas-0.1.9.dist-info → napari_tmidas-0.2.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: napari-tmidas
-Version: 0.1.9
+Version: 0.2.1
 Summary: A plugin for batch processing of confocal and whole-slide microscopy images of biological tissues
 Author: Marco Meer
 Author-email: marco.meer@pm.me
@@ -100,7 +100,7 @@ Now you can install `napari-tmidas` via [pip]:
 
     pip install napari-tmidas
 
-It is recommended to install the latest development version:
+It is recommended to install the latest development version. Please also regularly execute this command in the activated environment:
 
     pip install git+https://github.com/macromeer/napari-tmidas.git
 
@@ -108,19 +108,26 @@ It is recommended to install the latest development version:
 
 To use the Batch Microscopy Image Conversion pipeline, we need some libraries to read microscopy formats:
 
+    # mamba activate napari-tmidas
     pip install nd2 readlif tiffslide pylibCZIrw acquifer-napari
 
-For the Batch Crop Anything pipeline, we need to install MobileSAM and its dependencies:
-
-    pip install git+https://github.com/ChaoningZhang/MobileSAM.git
-
-
 If you want to batch compress images using [Zstandard](https://github.com/facebook/zstd), use the package manager of your operating system to install it:
 
     sudo apt-get install zstd    # for Linux
     brew install zstd            # for macOS
     choco install zstandard      # for Windows
 
+To use the Batch Crop Anything pipeline, we need to install SAM2 in the napari-tmidas environment:
+
+    # mamba activate napari-tmidas
+    cd /opt
+    git clone https://github.com/facebookresearch/sam2.git && cd sam2
+    pip install -e .
+    wget https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_large.pt -P checkpoints/
+    pip install decord
+
+
+
 
 ## Usage
 
@@ -153,7 +160,7 @@ You can start this pipeline via `Plugins > T-MIDAS > Batch Microscopy Image Conv
 ![image](https://github.com/user-attachments/assets/05929660-6672-4f76-89da-4f17749ccfad)
 
 4. You can click on the images in the table to show them in the viewer. For example first click on one of the `Original Files`, and then the corresponding `Processed File` to see an overlay.
-   
+
 <img src="https://github.com/user-attachments/assets/cfe84828-c1cc-4196-9a53-5dfb82d5bfce" alt="Image Processing Widget" style="width:75%; height:auto;">
 
 

{napari_tmidas-0.1.9.dist-info → napari_tmidas-0.2.1.dist-info}/RECORD

@@ -1,13 +1,13 @@
 napari_tmidas/__init__.py,sha256=YNBLESwk8jr_TlDdkSC1CwH0tf0CKHF1i2_efzLjdpk,589
-napari_tmidas/_crop_anything.py,sha256=NItpE6uzfeKujh8a53TDDkFN2thpKC5NGiXMpAmSnnM,45446
+napari_tmidas/_crop_anything.py,sha256=KgGZhNEHaGbk6npDHXGi7FrgahsfivlVwvdNAAYdME0,101452
 napari_tmidas/_file_conversion.py,sha256=V6evJmggUwOFzJO203Y5ltboHXEWNJQckZPedGRkrLI,72203
-napari_tmidas/_file_selector.py,sha256=sZOY0QNwyAgugsEzG5pqZLfrVeCHwjPEW2C_BHndzyI,39595
-napari_tmidas/_label_inspection.py,sha256=hCxKE0zYk-qBh4ohqiZcEGLXa-3lL8p88y45p2WnE1g,7329
+napari_tmidas/_file_selector.py,sha256=JQ8t_nVzJXqlVUwIBjGE2jJDeyhuDKCXHJP_cPbzzBw,43091
+napari_tmidas/_label_inspection.py,sha256=74V36y5EnGs0vWK1FC7Kui4CPLBW_SIg885PSKeZsJ4,9184
 napari_tmidas/_reader.py,sha256=A9_hdDxtVkVGmbOsbqgnARCSvpEh7GGPo7ylzmbnu8o,2485
 napari_tmidas/_registry.py,sha256=Oz9HFJh41MKRLeKxRuc7x7yzc-OrmoTdRFnfngFU_XE,2007
 napari_tmidas/_roi_colocalization.py,sha256=OVjdHvtFN07DgrtTX8uqbrxZL6jVwl2L3klorgW2C9k,43196
 napari_tmidas/_sample_data.py,sha256=khuv1jemz_fCjqNwEKMFf83Ju0EN4S89IKydsUMmUxw,645
-napari_tmidas/_version.py,sha256=bhntibG3PKk5Ai3XlSNEV8gj-ffItuKloY6vzWn6swo,511
+napari_tmidas/_version.py,sha256=UoNvMtd4wCG76RwoSpNCUtaFyTwakGcZolfjXzNVSMY,511
 napari_tmidas/_widget.py,sha256=u9uf9WILAwZg_InhFyjWInY4ej1TV1a59dR8Fe3vNF8,4794
 napari_tmidas/_writer.py,sha256=wbVfHFjjHdybSg37VR4lVmL-kdCkDZsUPDJ66AVLaFQ,1941
 napari_tmidas/napari.yaml,sha256=1Am1dA0-ZtCXk6veIT6jrMz3zwQ7dF8_p9tZTFx_vTg,2641
@@ -17,16 +17,22 @@ napari_tmidas/_tests/test_sample_data.py,sha256=D1HU_C3hWpO3mlSW_7Z94xaYHDtxz0XU
 napari_tmidas/_tests/test_widget.py,sha256=I_d-Cra_CTcS0QdMItg_HMphvhj0XCx81JnFyCHk9lg,2204
 napari_tmidas/_tests/test_writer.py,sha256=4_MlZM9a5So74J16_4tIOJc6pwTOw9R0-oAE_YioIx4,122
 napari_tmidas/processing_functions/__init__.py,sha256=osXY9jSgDsrwFaS6ShPHP0wGRxMuX1mHRN9EDa9l41g,1891
-napari_tmidas/processing_functions/basic.py,sha256=TJFvJ9AfUp7MBseUAgryLJXdqj0gSLSKqlEPxE3s1n0,6694
-napari_tmidas/processing_functions/cellpose_env_manager.py,sha256=zngS5eborsJUimFn_g1Lm_YOZk2ZNIKxceWNxOjpxEg,4885
-napari_tmidas/processing_functions/cellpose_segmentation.py,sha256=7BCHj_QA1QJEl1NrsuoIFAJ040_SCWw0U-U_xjCyk18,16187
+napari_tmidas/processing_functions/basic.py,sha256=NJj7pjVPGZwH2H8lnDtxxK-p3JLpcayAqfmTduuPFDw,26777
+napari_tmidas/processing_functions/careamics_denoising.py,sha256=DFE_6lefeqckAvx-1EqwzJSU3iR3g3ujBGRnF_fnpoM,11638
+napari_tmidas/processing_functions/careamics_env_manager.py,sha256=QfmhY5CaeFboUGTxeDlQDPi9WSfeBWp56Zz_qc2luew,11219
+napari_tmidas/processing_functions/cellpose_env_manager.py,sha256=EWNuHuY0PPw8_mL61ElZ58M0-DTduPKuWUdvsrmKV8I,6191
+napari_tmidas/processing_functions/cellpose_segmentation.py,sha256=miRPIsrkv0jL1jNdUFwlTkmr6-m9g7U7k9ijyeatUY0,13410
 napari_tmidas/processing_functions/colocalization.py,sha256=AiTTVAcVhKuuHZhrj5IHwbzns7-GE6ewvFqhYy1L-do,7657
 napari_tmidas/processing_functions/file_compression.py,sha256=mxR-yqBdc-T1XI3StIXpW8h5xGdCOtLQjt8uoRFpDSY,6859
+napari_tmidas/processing_functions/sam2_env_manager.py,sha256=WzKOLFeu1KZRRBryKdWkDm6QJolhs3rCj-KD6Q-z9dE,2897
+napari_tmidas/processing_functions/sam2_mp4.py,sha256=NF0dWar2uyP_yQWxC8e08J6198C2qxEIzQccSI_5g40,10352
 napari_tmidas/processing_functions/scipy_filters.py,sha256=kKpDAlQQ0ZNbkt77QUWi-Bwolk6MMDvtG_bZJV3MjOo,1612
-napari_tmidas/processing_functions/skimage_filters.py,sha256=6wSROKH71zwSFBOZ22zgp-4Nrq79GNd7znOitiH3Z3c,4030
-napari_tmidas-0.1.9.dist-info/licenses/LICENSE,sha256=tSjiOqj57exmEIfP2YVPCEeQf0cH49S6HheQR8IiY3g,1485
-napari_tmidas-0.1.9.dist-info/METADATA,sha256=NwevLeFpp4eNGWdUMohmeiY8OICKpFU7GNjDnOqvTBU,11560
-napari_tmidas-0.1.9.dist-info/WHEEL,sha256=SmOxYU7pzNKBqASvQJ7DjX3XGUF92lrGhMb3R6_iiqI,91
-napari_tmidas-0.1.9.dist-info/entry_points.txt,sha256=fbVjzbJTm4aDMIBtel1Lyqvq-CwXY7wmCOo_zJ-jtRY,60
-napari_tmidas-0.1.9.dist-info/top_level.txt,sha256=63ybdxCZ4SeT13f_Ou4TsivGV_2Gtm_pJOXToAt30_E,14
-napari_tmidas-0.1.9.dist-info/RECORD,,
+napari_tmidas/processing_functions/skimage_filters.py,sha256=tSBx0nal88ixxVbu5o7ojTn90HgsUTt-aA_T6XLvmyY,16320
+napari_tmidas/processing_functions/timepoint_merger.py,sha256=DwL5vZBSplXt9dBBrKtMm9aH_NvT3mY7cdbeGg2OU_Y,16567
+napari_tmidas/processing_functions/trackastra_tracking.py,sha256=4kswVZCRHJ68oY95ezZmSrXMNrhjjEeN5x8a7GIjh4E,9084
+napari_tmidas-0.2.1.dist-info/licenses/LICENSE,sha256=tSjiOqj57exmEIfP2YVPCEeQf0cH49S6HheQR8IiY3g,1485
+napari_tmidas-0.2.1.dist-info/METADATA,sha256=AYL6vxMkpxPlXBrYZdBOvH1_AuNAQ83X68hue3KfA_k,11874
+napari_tmidas-0.2.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+napari_tmidas-0.2.1.dist-info/entry_points.txt,sha256=fbVjzbJTm4aDMIBtel1Lyqvq-CwXY7wmCOo_zJ-jtRY,60
+napari_tmidas-0.2.1.dist-info/top_level.txt,sha256=63ybdxCZ4SeT13f_Ou4TsivGV_2Gtm_pJOXToAt30_E,14
+napari_tmidas-0.2.1.dist-info/RECORD,,

{napari_tmidas-0.1.9.dist-info → napari_tmidas-0.2.1.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (79.0.1)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any