Presidio_Metrics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "IyuzxMJVIx14",
    "outputId": "aa3b76e7-b7cf-4404-8cc4-77422092670f"
   },
   "outputs": [],
   "source": [
    "!pip install -q presidio_image_redactor\n",
    "!python -m spacy download en_core_web_lg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "collapsed": true,
    "id": "noXvtTT5JA1i",
    "jupyter": {
     "outputs_hidden": true
    },
    "outputId": "ee358962-5c66-4234-8b24-dff951dbae40"
   },
   "outputs": [],
   "source": [
    "!sudo apt install tesseract-ocr\n",
    "!sudo apt install libtesseract-dev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "0vzo3hARJYjH",
    "outputId": "5982d0ab-da1b-419f-a487-270b7b36c3fc"
   },
   "outputs": [],
   "source": [
    "import inspect\n",
    "from presidio_image_redactor import DicomImagePiiVerifyEngine\n",
    "\n",
    "file_path = inspect.getfile(DicomImagePiiVerifyEngine)\n",
    "file_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "im0ARo1OJe2J"
   },
   "outputs": [],
   "source": [
    "!rm -rf {file_path}\n",
    "!cp ./dicom_image_pii_verify_engine.py /usr/local/lib/python3.11/dist-packages/presidio_image_redactor/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "9en0baG0JolU"
   },
   "outputs": [],
   "source": [
    "# RESTART SESSION!!!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "W7unRIjwJrCd"
   },
   "outputs": [],
   "source": [
    "def calculate_precision(gt, pred):\n",
    "    tp = [i for i in pred if i in gt]\n",
    "    try:\n",
    "        precision = len(tp) / len(pred)\n",
    "    except ZeroDivisionError:\n",
    "        precision = 0\n",
    "\n",
    "    return precision\n",
    "\n",
    "def calculate_recall(gt, pred):\n",
    "    tp = [i for i in pred if i in gt]\n",
    "    try:\n",
    "        recall = len(tp) / len(gt)\n",
    "    except ZeroDivisionError:\n",
    "        recall = 0\n",
    "\n",
    "    return recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "U34ts4U_JtO5",
    "outputId": "be28fda9-1376-4bcd-face-15aa42f84e79"
   },
   "outputs": [],
   "source": [
    "import json\n",
    "\n",
    "import glob\n",
    "import os\n",
    "import time\n",
    "import pydicom\n",
    "from pathlib import Path\n",
    "import matplotlib.pyplot as plt\n",
    "from PIL import Image\n",
    "from presidio_image_redactor import DicomImageRedactorEngine\n",
    "from presidio_image_redactor import DicomImagePiiVerifyEngine\n",
    "\n",
    "dicom_engine = DicomImageRedactorEngine()\n",
    "dicom_image_engine = DicomImagePiiVerifyEngine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "PUXXo46sJuVA"
   },
   "outputs": [],
   "source": [
    "!python prepare_data.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ds_Vb2HBJywO",
    "outputId": "f6f3a821-e221-4c0e-abdd-4f537310ad00"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Output written to presidio_result/dicom_files\n"
     ]
    }
   ],
   "source": [
    "input_path = \"./dicom_files/\"\n",
    "output_parent_dir = \"./presidio_result/\"\n",
    "\n",
    "start = time.time()\n",
    "\n",
    "dicom_engine.redact_from_directory(\n",
    "    input_dicom_path = input_path,\n",
    "    output_dir = output_parent_dir,\n",
    "    fill=\"contrast\",\n",
    "    save_bboxes=True\n",
    ")\n",
    "\n",
    "time_taken = round(time.time() - start, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "e1yyDjBuJ1nu",
    "outputId": "f84a2e61-264c-4ce7-9476-163cb265eb02"
   },
   "outputs": [],
   "source": [
    "# Original DICOM images\n",
    "p = Path(input_path).glob(\"**/*.dcm\")\n",
    "original_files = [x for x in p if x.is_file()]\n",
    "print(f\"Total Original Files {len(original_files)}\")\n",
    "\n",
    "# Redacted DICOM images\n",
    "p = Path(output_parent_dir).glob(\"**/*.dcm\")\n",
    "redacted_files = [x for x in p if x.is_file()]\n",
    "print(f\"Total Redacted Files {len(redacted_files)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6bVi-3_rJ4kW",
    "outputId": "7c439c6f-ffab-456d-a22b-4119f9024d34"
   },
   "outputs": [],
   "source": [
    "print(f\"Time Taken : {round(time_taken, 2)}\")\n",
    "print(f\"Average Time Taken : {round( time_taken / len(redacted_files), 2)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "PxJSVtZ4J5nT"
   },
   "outputs": [],
   "source": [
    "# Set paths\n",
    "data_dir = \"./dicom_files/\"\n",
    "gt_path = \"./gt_main.json\"\n",
    "\n",
    "with open(gt_path) as json_file:\n",
    "    gt = json.load(json_file)\n",
    "\n",
    "# Get list of files\n",
    "gt_dicom_files = list(gt.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "collapsed": true,
    "id": "nxTQ0PORJ7sn",
    "jupyter": {
     "outputs_hidden": true
    },
    "outputId": "6a5b3134-a55b-4593-9dc2-29043581d136"
   },
   "outputs": [],
   "source": [
    "collect_eval_results = {}\n",
    "\n",
    "for item in gt_dicom_files:\n",
    "    gt_file_of_interest = gt[item]\n",
    "\n",
    "    instance = pydicom.dcmread(os.path.join(data_dir, item))\n",
    "\n",
    "    verify_image, ocr_results, analyzer_results = dicom_image_engine.verify_dicom_instance(instance)\n",
    "    _, eval_results = dicom_image_engine.eval_dicom_instance(instance, gt_file_of_interest, use_metadata=True)\n",
    "\n",
    "    collect_eval_results[item] = eval_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RwYyrfMHJ9MJ",
    "outputId": "390e7c93-c182-4d0b-f22c-778d26c21f3e"
   },
   "outputs": [],
   "source": [
    "total_precision = 0.0\n",
    "total_recall = 0.0\n",
    "\n",
    "collect_results = {}\n",
    "\n",
    "for key, val in collect_eval_results.items():\n",
    "  presidio_pred = [i[\"label\"].strip().lower() for i in val[\"all_positives\"]]\n",
    "  gt_pred = [i[\"label\"].strip().lower() for i in val[\"ground_truth\"]]\n",
    "\n",
    "  precision = calculate_precision(gt_pred, presidio_pred)\n",
    "  recall = calculate_recall(gt_pred, presidio_pred)\n",
    "\n",
    "  collect_results[key] = {\"gt\" : gt_pred, \"pred\" : presidio_pred, \"precision\" : precision, \"recall\" : recall}\n",
    "\n",
    "  total_precision += precision\n",
    "  total_recall += recall\n",
    "\n",
    "final_precision = round(total_precision / len(collect_eval_results), 3)\n",
    "final_recall = round(total_recall / len(collect_eval_results), 3)\n",
    "f1_score = round(2 * ((final_precision * final_recall) / (final_precision + final_recall)), 3 )\n",
    "\n",
    "print(f\"Precision : {final_precision}\")\n",
    "print(f\"Recall : {final_recall}\")\n",
    "print(f\"F1-Score : {f1_score}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "BIbJmOKRKMZo"
   },
   "outputs": [],
   "source": [
    "with open(\"./results/ner_result/presidio_result.json\", \"w\") as file_out:\n",
    "  json.dump(collect_results, file_out, indent=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "DaYaiqEALFCK"
   },
   "outputs": [],
   "source": [
    "os.makedirs(\"./image_result\", exist_ok=True)\n",
    "os.makedirs(\"./results/deid_image_result/presidio/\", exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "tbzGZXnEK6i5"
   },
   "outputs": [],
   "source": [
    "presidio_dicom_result = glob.glob(\"./presidio_result/dicom_files/*.dcm\")\n",
    "\n",
    "for dicom_file in presidio_dicom_result:\n",
    "    base_name = os.path.basename(dicom_file).replace(\".dcm\", \".jpg\")\n",
    "    ds = pydicom.dcmread(dicom_file)\n",
    "\n",
    "    image = Image.fromarray(ds.pixel_array).convert(\"L\")\n",
    "\n",
    "    file_out = os.path.join(\"./results/deid_image_result/presidio/\", base_name)\n",
    "    image.save(file_out)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}