endoreg-db 0.5.2__py3-none-any.whl → 0.6.0__py3-none-any.whl

endoreg_db/models/data_file/base_classes/abstract_video.py

@@ -1,201 +0,0 @@
-# import cv2
-from PIL import Image
-from django.core.files.base import ContentFile
-from django.db import models, transaction
-from tqdm import tqdm
-# import cv2
-import io
-from datetime import date
-
-BATCH_SIZE = 1000
-
-class AbstractVideo(models.Model):
-    file = models.FileField(upload_to="raw_videos", blank=True, null=True)
-    video_hash = models.CharField(max_length=255, unique=True)
-    patient = models.ForeignKey("Patient", on_delete=models.CASCADE, blank=True, null=True)
-    date = models.DateField(blank=True, null=True)
-    suffix = models.CharField(max_length=255)
-    fps = models.FloatField()
-    duration = models.FloatField()
-    width = models.IntegerField()
-    height = models.IntegerField()
-    endoscope_image_x = models.IntegerField(blank=True, null=True)
-    endoscope_image_y = models.IntegerField(blank=True, null=True)
-    endoscope_image_width = models.IntegerField(blank=True, null=True)
-    endoscope_image_height = models.IntegerField(blank=True, null=True)
-    center = models.ForeignKey("Center", on_delete=models.CASCADE, blank=True, null=True)
-    endoscopy_processor = models.ForeignKey("EndoscopyProcessor", on_delete=models.CASCADE, blank=True, null=True)
-    frames_extracted = models.BooleanField(default=False)
-
-    meta = models.JSONField(blank=True, null=True)
-
-    class Meta:
-        abstract = True
-
-    def get_roi_endoscope_image(self):
-        return {
-            'x': self.endoscope_image_content_x,
-            'y': self.endoscope_image_content_y,
-            'width': self.endoscope_image_content_width,
-            'height': self.endoscope_image_content_height,
-        }
-
-    def initialize_metadata_in_db(self, video_meta=None):
-        if not video_meta:
-            video_meta = self.meta
-        self.set_examination_date_from_video_meta(video_meta)
-        self.patient, created = self.get_or_create_patient(video_meta)
-        self.save()
-
-    def get_or_create_patient(self, video_meta=None):
-        from ...persons import Patient
-        if not video_meta:
-            video_meta = self.meta
-
-        patient_first_name = video_meta['patient_first_name']
-        patient_last_name = video_meta['patient_last_name']
-        patient_dob = video_meta['patient_dob']
-
-        # assert that we got all the necessary information
-        assert patient_first_name and patient_last_name and patient_dob, "Missing patient information"
-
-        patient, created = Patient.objects.get_or_create(
-            first_name=patient_first_name,
-            last_name=patient_last_name,
-            dob=patient_dob
-        )
-
-        return patient, created
-
-    def get_frame_model(self):
-        assert 1 == 2, "This method should be overridden in derived classes"
-
-    def get_video_model(self):
-        assert 1 == 2, "This method should be overridden in derived classes"
-
-    def get_frame_number(self):
-        """
-        Get the number of frames in the video.
-        """
-        frame_model = self.get_frame_model()
-        framecount = frame_model.objects.filter(video=self).count()
-        return framecount
-    
-    def set_frames_extracted(self, value:bool=True):
-        self.frames_extracted = value
-        self.save()
-        
-    def get_frames(self):
-        """
-        Retrieve all frames for this video in the correct order.
-        """
-        frame_model = self.get_frame_model()
-        return frame_model.objects.filter(video=self).order_by('frame_number')
-
-    def get_frame(self, frame_number):
-        """
-        Retrieve a specific frame for this video.
-        """
-        frame_model = self.get_frame_model()
-        return frame_model.objects.get(video=self, frame_number=frame_number)
-
-    def get_frame_range(self, start_frame_number:int, end_frame_number:int):
-        """
-        Expects numbers of start and stop frame.
-        Returns all frames of this video within the given range in ascending order.
-        """
-        frame_model = self.get_frame_model()
-        return frame_model.objects.filter(video=self, frame_number__gte=start_frame_number, frame_number__lte=end_frame_number).order_by('frame_number')
-
-    def _create_frame_object(self, frame_number, image_file):
-        frame_model = self.get_frame_model()
-        frame = frame_model(
-                video=self,
-                frame_number=frame_number,
-                suffix='jpg',
-            )
-        frame.image_file = image_file  # Temporary store the file-like object
-
-        return frame
-
-    def _bulk_create_frames(self, frames_to_create):
-        frame_model = self.get_frame_model()
-        with transaction.atomic():
-            frame_model.objects.bulk_create(frames_to_create)
-
-            # After the DB operation, save the ImageField for each object
-            for frame in frames_to_create:
-                frame_name = f"video_{self.id}_frame_{str(frame.frame_number).zfill(7)}.jpg"
-                frame.image.save(frame_name, frame.image_file)
-
-            # Clear the list for the next batch
-            frames_to_create = []
-
-    def set_examination_date_from_video_meta(self, video_meta=None):
-        if not video_meta:
-            video_meta = self.meta
-        date_str = video_meta['examination_date'] # e.g. 2020-01-01
-        if date_str:
-            self.date = date.fromisoformat(date_str)
-            self.save()
-
-    def extract_all_frames(self):
-        """
-        Extract all frames from the video and store them in the database.
-        Uses Django's bulk_create for more efficient database operations.
-        """
-        # Open the video file
-        video = cv2.VideoCapture(self.file.path)
-
-        # Initialize video properties
-        self.initialize_video_specs(video)
-
-        # Prepare for batch operation
-        frames_to_create = []
-
-        # Extract frames
-        for frame_number in tqdm(range(int(self.duration * self.fps))):
-            # Read the frame
-            success, image = video.read()
-            if not success:
-                break
-
-            # Convert the numpy array to a PIL Image object
-            pil_image = Image.fromarray(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
-
-            # Save the PIL Image to a buffer
-            buffer = io.BytesIO()
-            pil_image.save(buffer, format='JPEG')
-
-            # Create a file-like object from the byte data in the buffer
-            image_file = ContentFile(buffer.getvalue())
-
-            # Prepare Frame instance (don't save yet)
-            frame = self._create_frame_object(frame_number, image_file)
-            frames_to_create.append(frame)
-
-            # Perform bulk create when reaching BATCH_SIZE
-            if len(frames_to_create) >= BATCH_SIZE:
-                self._bulk_create_frames(frames_to_create)
-                frames_to_create = []
-
-
-        # Handle remaining frames
-        if frames_to_create:
-            self._bulk_create_frames(frames_to_create)
-            frames_to_create = []
-
-        # Close the video file
-        video.release()
-        self.set_frames_extracted(True)
-
-
-    def initialize_video_specs(self, video):
-        """
-        Initialize and save video metadata like framerate, dimensions, and duration.
-        """
-        self.fps = video.get(cv2.CAP_PROP_FPS)
-        self.width = int(video.get(cv2.CAP_PROP_FRAME_WIDTH))
-        self.height = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT))
-        self.duration = video.get(cv2.CAP_PROP_FRAME_COUNT) / self.fps
-        self.save()

endoreg_db/models/data_file/frame.py

@@ -1,45 +0,0 @@
-from endoreg_db.models.annotation.image_classification import ImageClassificationAnnotation
-from endoreg_db.models.label.label import Label
-from .base_classes import AbstractFrame
-from django.db import models
-
-class Frame(AbstractFrame):
-    video = models.ForeignKey("Video", on_delete=models.CASCADE, related_name="frames")
-
-class LegacyFrame(AbstractFrame):
-    video = models.ForeignKey("LegacyVideo", on_delete=models.CASCADE, related_name='frames')
-    image = models.ImageField(upload_to="legacy_frames", blank=True, null=True)
-    suffix = models.CharField(max_length=255)
-    # ImageClassificationAnnotation has a foreign key to this model (related name: image_classification_annotations)
-
-    class Meta:
-        unique_together = ('video', 'frame_number')
-        indexes = [
-            models.Index(fields=['video', 'frame_number']),
-        ]
-
-    def get_classification_annotations(self):
-        """
-        Get all image classification annotations for this frame.
-        """
-        return ImageClassificationAnnotation.objects.filter(legacy_frame=self)
-
-    def get_classification_annotations_by_label(self, label:Label):
-        """
-        Get all image classification annotations for this frame with the given label.
-        """
-        return ImageClassificationAnnotation.objects.filter(legacy_frame=self, label=label)
-
-    def get_classification_annotations_by_value(self, value:bool):
-        """
-        Get all image classification annotations for this frame with the given value.
-        """
-        return ImageClassificationAnnotation.objects.filter(legacy_frame=self, value=value)
-
-    def get_classification_annotations_by_label_and_value(self, label:Label, value:bool):
-        """
-        Get all image classification annotations for this frame with the given label and value.
-        """
-        return ImageClassificationAnnotation.objects.filter(legacy_frame=self, label=label, value=value)
-
-    

endoreg_db/models/data_file/import_classes/__init__.py

@@ -1,32 +0,0 @@
-import os
-from pathlib import Path
-
-from .raw_video import RawVideoFile
-from .raw_pdf import RawPdfFile
-
-# FileImporter class
-# This class is used to import data from a file into the database.
-# Expects a directory containing files to import.
-# creates correct import file object depending on file type
-
-# FileImporter class
-# This class is used to import data from a file into the database by creating objects for the files.
-# main method is import_files which expects a path to a directory containing files to import.
-# creates correct import file object depending on file type by checking the file extension
-
-class FileImporter:
-    def __init__(self, directory):
-        self.directory = directory
-
-    def import_files(self):
-        directory_path = Path(self.directory)
-        for file in directory_path.iterdir():
-            if file.is_file():
-                if file.suffix.lower() in ['.mov', '.mp4']:
-                    RawVideoFile.create_from_file(file)
-                else:
-                    raise ValueError(f"File type {file.suffix} not supported")
-            else:
-                raise ValueError(f"{file} is not a file")
-            
-        

endoreg_db/models/data_file/import_classes/processing_functions/__init__.py

@@ -1,35 +0,0 @@
-from .video import (
-    get_videos_scheduled_for_frame_extraction, 
-    extract_frames_from_video,
-    extract_frames_from_videos,
-
-    get_videos_scheduled_for_ocr,
-    videos_scheduled_for_ocr_preflight,
-    perform_ocr_on_video,
-    perform_ocr_on_videos,
-
-    videos_scheduled_for_initial_prediction_preflight,
-    get_videos_scheduled_for_initial_prediction,
-    get_multilabel_model,
-    get_multilabel_classifier,
-    get_crops,
-    perform_initial_prediction_on_video,
-    perform_initial_prediction_on_videos,
-
-    videos_scheduled_for_prediction_import_preflight,
-    get_videos_scheduled_for_prediction_import,
-    import_predictions_for_video,
-    import_predictions_for_videos,
-    
-
-    delete_frames_preflight,
-    get_videos_scheduled_for_frame_deletion,
-    delete_frames_for_video,
-    delete_frames,
-)
-
-from .pdf import (
-    get_pdf_files_scheduled_for_processing,
-    process_pdf_file,
-    process_pdf_files,
-)

endoreg_db/models/data_file/import_classes/processing_functions/pdf.py

@@ -1,28 +0,0 @@
-from ..raw_pdf import RawPdfFile
-from logging import getLogger
-#
-# # setup logging to pdf_import.log
-logger = getLogger('examination_pdf_import')
-
-def get_pdf_files_scheduled_for_processing():
-    reports = RawPdfFile.objects.filter(
-        state_report_processing_required=True
-    )
-    return reports
-
-def process_pdf_file(report:RawPdfFile):
-    if report.update(save=True, verbose=True):
-        logger.info(f"Report {report} processed successfully")
-        return True
-    else:
-        logger.error(f"Report {report} processing failed")
-        return False
-
-
-def process_pdf_files():
-    reports = get_pdf_files_scheduled_for_processing()
-    for report in reports:
-        process_pdf_file(report)
-    
-
-

endoreg_db/models/data_file/import_classes/processing_functions/video.py

@@ -1,260 +0,0 @@
-from ..raw_video import RawVideoFile
-import json 
-
-# # Starting point
-# Automated tasks generate RawVideoFile objects in our db.
-# Each object has state_{NAME} attributes.
-# We will create functions which query the db for RawVideoFile 
-# objects with specific state_{NAME} attributes.
-# Then, we perform the necessary operations on the RawVideoFile and
-# update the state_{NAME} attributes accordingly.
-
-# # Step 1 - Frame Extraction
-# function to query for videos scheduled for frame extraction,
-# these have state_frames_required and state_frames_extracted
-def get_videos_scheduled_for_frame_extraction():
-    return RawVideoFile.objects.filter(
-        state_frames_required=True,
-        state_frames_extracted=False
-    )
-
-def extract_frames_from_video(video:RawVideoFile):
-    # extract frames from video
-    video.extract_frames()
-
-    # update state_frames_extracted
-    video.state_frames_extracted = True
-    video.save()
-
-    return video
-
-def extract_frames_from_videos():
-    videos = get_videos_scheduled_for_frame_extraction()
-    for video in videos:
-        extract_frames_from_video(video)
-
-# # Step 2 - OCR
-# function to query for videos scheduled for OCR,
-# these have 
-# state_ocr_required = True and state_ocr_completed = False and state_frames_extracted = True
-def get_videos_scheduled_for_ocr():
-    return RawVideoFile.objects.filter(
-        state_ocr_required=True,
-        state_ocr_completed=False,
-        state_frames_extracted=True
-    )
-
-# function to set state_frames_required to True for videos 
-# which are scheduled for OCR but have not had frames extracted
-def videos_scheduled_for_ocr_preflight():
-    videos = RawVideoFile.objects.filter(
-        state_ocr_required=True,
-        state_ocr_completed=False,
-        state_frames_extracted=False
-    )
-    for video in videos:
-        video.state_frames_required = True
-        video.save()
-
-def perform_ocr_on_video(video:RawVideoFile):
-    # perform OCR on video
-    video.update_text_metadata()
-
-    # update state_ocr_completed
-    video.state_ocr_completed = True
-    video.save()
-
-    return video
-
-def perform_ocr_on_videos():
-    videos = get_videos_scheduled_for_ocr()
-    for video in videos:
-        perform_ocr_on_video(video)
-
-
-# # Step 3 - initial Prediction
-# function to query for videos scheduled for initial prediction,
-# these have
-# state_initial_prediction_required = True and state_initial_prediction_completed = False and state_frames_extracted = True
-def videos_scheduled_for_initial_prediction_preflight():
-    videos = RawVideoFile.objects.filter(
-        state_initial_prediction_required=True,
-        state_initial_prediction_completed=False,
-        state_frames_extracted=False
-    )
-    for video in videos:
-        video.state_frames_required = True
-        video.save()
-
-def get_videos_scheduled_for_initial_prediction():
-    return RawVideoFile.objects.filter(
-        state_initial_prediction_required=True,
-        state_initial_prediction_completed=False,
-        state_frames_extracted=True
-    )
-
-from pathlib import Path
-def get_multilabel_model(model_path:Path):
-    from agl_predict_endo_frame.model_loader import MultiLabelClassificationNet
-    model_path_str = model_path.resolve().as_posix()
-    model = MultiLabelClassificationNet.load_from_checkpoint(model_path_str)
-    model.cuda()
-    model.eval()
-    return model
-
-def get_multilabel_classifier(model, verbose:bool=False):
-    from agl_predict_endo_frame.predict import Classifier
-    classifier = Classifier(model, verbose = verbose)
-    return classifier
-
-def get_crops(video, paths):
-    endo_roi_dict = video.get_endo_roi()
-    # dict with x, y, width height
-    # crops is list of touples with (y_min, y_max, x_min, x_max)
-    crop_tuple = (
-        endo_roi_dict["y"],
-        endo_roi_dict["y"] + endo_roi_dict["height"],
-        endo_roi_dict["x"],
-        endo_roi_dict["x"] + endo_roi_dict["width"],  
-    )
-    crops = [crop_tuple for _ in paths]
-    return crops
-
-# model = MultiLabelClassificationNet.load_from_checkpoint("model/colo_segmentation_RegNetX800MF_6.ckpt")
-def perform_initial_prediction_on_video(
-    video:RawVideoFile, model_path,
-    window_size_s, min_seq_len_s
-):
-
-    model = get_multilabel_model(model_path)
-    classifier = get_multilabel_classifier(model, verbose = True)
-
-    paths = video.get_frame_paths()
-    string_paths = [p.resolve().as_posix() for p in paths]
-    crops = get_crops(video, string_paths)
-    fps = video.get_fps()
-
-    predictions = classifier.pipe(string_paths, crops)
-    readable_predictions = [classifier.readable(p) for p in predictions]
-    result_dict = classifier.post_process_predictions_serializable(
-        readable_predictions,
-        window_size_s = window_size_s,
-        min_seq_len_s = min_seq_len_s,
-        fps = fps
-    )
-
-    
-    # Predictions
-    _path = video.get_predictions_path()
-    with open(_path, "w") as f:
-        json.dump(result_dict["predictions"], f, indent = 4)
-
-    # smooth_predictions
-    _path = video.get_smooth_predictions_path()
-    with open(_path, "w") as f:
-        json.dump(result_dict["smooth_predictions"], f, indent = 4)
-
-    # binary_predictions
-    _path = video.get_binary_predictions_path()
-    with open(_path, "w") as f:
-        json.dump(result_dict["binary_predictions"], f, indent = 4)
-
-    # Raw Sequences
-    _path = video.get_raw_sequences_path()
-    with open(_path, "w") as f:
-        json.dump(result_dict["raw_sequences"], f, indent = 4)
-
-    # filtered_sequences
-    _path = video.get_filtered_sequences_path()
-    with open(_path, "w") as f:
-        json.dump(result_dict["filtered_sequences"], f, indent = 4)
-
-    
-    # update state_initial_prediction_completed
-    video.state_initial_prediction_required = False
-    video.state_initial_prediction_completed = True
-    video.state_initial_prediction_import_required = True
-    video.state_initial_prediction_import_completed = False
-    video.save()
-
-    return video
-
-def perform_initial_prediction_on_videos(
-    model_path,
-    window_size_s, min_seq_len_s
-):
-    videos = get_videos_scheduled_for_initial_prediction()
-    for video in videos:
-        perform_initial_prediction_on_video(
-        video,
-        model_path, window_size_s, min_seq_len_s
-    )
-
-def videos_scheduled_for_prediction_import_preflight():
-    videos = RawVideoFile.objects.filter(
-        state_initial_prediction_completed=True,
-        state_initial_prediction_import_completed=False
-    )
-    for video in videos:
-        video.state_initial_prediction_required = True
-        video.save()
-    
-def get_videos_scheduled_for_prediction_import():
-    return RawVideoFile.objects.filter(
-        state_initial_prediction_import_required=True,
-        state_initial_prediction_import_completed=False,
-        state_initial_prediction_completed=True
-    )
-
-def import_predictions_for_video(video:RawVideoFile):
-    # import predictions for video
-    pass
-
-    # update state_prediction_import_completed
-    # video.state_initial_prediction_import_required = False
-    # video.state_initial_prediction_import_completed = True
-    # video.save()
-
-    return video
-
-def import_predictions_for_videos():
-    videos = get_videos_scheduled_for_prediction_import()
-    for video in videos:
-        import_predictions_for_video(video)
-
-# # Step 4 - Delete Frames if not needed anymore
-# function to query for videos scheduled for frame deletion,
-# first we need to set state_frames_required = False for videos with:
-# state_ocr_required = False and state_ocr_completed = True and
-# state_initial_prediction_required = False and state_initial_prediction_completed = True
-def delete_frames_preflight():
-    videos = RawVideoFile.objects.filter(
-        state_ocr_required=False,
-        state_ocr_completed=True,
-        state_initial_prediction_required=False,
-        state_initial_prediction_completed=True
-    )
-    for video in videos:
-        video.state_frames_required = False
-        video.save()
-
-# function to query for videos scheduled for frame deletion,
-# frames should be deleted if state_frames_required = False
-def get_videos_scheduled_for_frame_deletion():
-    return RawVideoFile.objects.filter(
-        state_frames_required=False
-    )
-
-def delete_frames_for_video(video:RawVideoFile):
-    # delete frames for video
-
-    # update state_frames_deleted
-    video.state_frames_extracted = False
-    video.save()
-
-    return video
-
-def delete_frames():
-    videos = get_videos_scheduled_for_frame_deletion()
-    for video in videos:
-        delete_frames_for_video(video)