example_data/generator.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100 random records have been generated and saved to 'random_medical_data.csv'\n",
      "\n",
      "Example of the first 3 generated records:\n",
      "\n",
      "Accession Number: 985440\n",
      "Report Text:\n",
      "Patient Name: Alexander Moore\n",
      "Age: 57 years\n",
      "Sex: Female\n",
      "Left breast not examined.\n",
      "Follow-up mammogram recommended in 6 months.\n",
      "Suspicious lesion identified in right breast, further investigation recommended.\n",
      "Hormone Therapy: Current hormone therapy\n",
      "Right breast examined.\n",
      "Right breast findings: No masses. No calcifications. No asymmetries. Skin retraction.\n",
      "Breast Density: c) Heterogeneously dense\n",
      "No suspicious lesions identified in left breast.\n",
      "Previous Biopsies: No previous breast biopsies\n",
      "Dense breast tissue may lower the sensitivity of mammography.\n",
      "Family History: Positive family history of breast cancer\n",
      "Source_RB: 1, Source_LB: 0, Cancer_RB: 1, Cancer_LB: 0\n",
      "\n",
      "Accession Number: 503958\n",
      "Report Text:\n",
      "Patient Name: Harper Davis\n",
      "Age: 74 years\n",
      "Sex: Male\n",
      "Follow-up mammogram recommended in 6 months.\n",
      "Right breast findings: Suspicious mass. Benign-appearing calcifications. Global asymmetry. Skin thickening.\n",
      "Suspicious lesion identified in right breast, further investigation recommended.\n",
      "Patient reported breast pain, which is likely unrelated to imaging findings.\n",
      "Dense breast tissue may lower the sensitivity of mammography.\n",
      "Suspicious lesion identified in left breast, biopsy advised.\n",
      "Left breast findings: Suspicious mass. Benign-appearing calcifications. Global asymmetry. Skin thickening.\n",
      "Previous Biopsies: No previous breast biopsies\n",
      "Left breast examined:\n",
      "Hormone Therapy: Current hormone therapy\n",
      "Family History: Positive family history of breast cancer\n",
      "Right breast examined.\n",
      "Additional ultrasound imaging may provide further clarification.\n",
      "Breast Density: c) Heterogeneously dense\n",
      "Source_RB: 1, Source_LB: 1, Cancer_RB: 1, Cancer_LB: 1\n",
      "\n",
      "Accession Number: 894772\n",
      "Report Text:\n",
      "Patient Name: Liam Jackson\n",
      "Age: 60 years\n",
      "Sex: Male\n",
      "Additional ultrasound imaging may provide further clarification.\n",
      "Right breast not examined.\n",
      "Patient reported breast pain, which is likely unrelated to imaging findings.\n",
      "Routine screening mammogram recommended in 1 year.\n",
      "No suspicious lesions identified in right breast.\n",
      "Breast Density: b) Scattered areas of fibroglandular density\n",
      "Left breast not examined.\n",
      "Hormone Therapy: Current hormone therapy\n",
      "No suspicious lesions identified in left breast.\n",
      "Family History: Positive family history of breast cancer\n",
      "Previous Biopsies: No previous breast biopsies\n",
      "Source_RB: 0, Source_LB: 0, Cancer_RB: 0, Cancer_LB: 0\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "import csv\n",
    "\n",
    "def generate_random_data(num_records, simple):\n",
    "    random.seed(0)\n",
    "    data = []\n",
    "    \n",
    "    first_names = [\"Emma\", \"Liam\", \"Olivia\", \"Noah\", \"Ava\", \"Ethan\", \"Sophia\", \"Mason\", \"Isabella\", \"William\",\n",
    "                   \"Mia\", \"James\", \"Charlotte\", \"Benjamin\", \"Amelia\", \"Lucas\", \"Harper\", \"Henry\", \"Evelyn\", \"Alexander\"]\n",
    "    last_names = [\"Smith\", \"Johnson\", \"Williams\", \"Brown\", \"Jones\", \"Garcia\", \"Miller\", \"Davis\", \"Rodriguez\", \"Martinez\",\n",
    "                  \"Hernandez\", \"Lopez\", \"Gonzalez\", \"Wilson\", \"Anderson\", \"Thomas\", \"Taylor\", \"Moore\", \"Jackson\", \"Martin\"]\n",
    "    \n",
    "    accession_numbers = random.sample(range(100000, 999999), num_records)\n",
    "    \n",
    "    for i in range(num_records):\n",
    "        accession_number = accession_numbers[i]\n",
    "        patient_name = f\"{random.choice(first_names)} {random.choice(last_names)}\"\n",
    "        age = random.randint(20, 80)\n",
    "        sex = random.choice([\"Male\", \"Female\"])\n",
    "        \n",
    "        source_rb = random.choice([0, 1])\n",
    "        source_lb = random.choice([0, 1])\n",
    "        if source_rb == 0:\n",
    "            cancer_rb = 0\n",
    "        else:\n",
    "            cancer_rb = random.choice([0, 1])\n",
    "        \n",
    "        if source_lb == 0:\n",
    "            cancer_lb = 0\n",
    "        else:\n",
    "            cancer_lb = random.choice([0, 1])\n",
    "        \n",
    "        breast_density = random.choice([\"a) Almost entirely fatty\", \"b) Scattered areas of fibroglandular density\", \n",
    "                                        \"c) Heterogeneously dense\", \"d) Extremely dense\"])\n",
    "        family_history = random.choice([\"Positive family history of breast cancer\", \"No family history of breast cancer\"])\n",
    "        previous_biopsies = random.choice([\"Previous breast biopsies\", \"No previous breast biopsies\"])\n",
    "        hormone_therapy = random.choice([\"Current hormone therapy\", \"No current hormone therapy\"])\n",
    "        \n",
    "        calcifications = random.choice([\"No calcifications\", \"Benign-appearing calcifications\", \"Suspicious calcifications\"])\n",
    "        masses = random.choice([\"No masses\", \"Benign-appearing mass\", \"Suspicious mass\"])\n",
    "        asymmetries = random.choice([\"No asymmetries\", \"Focal asymmetry\", \"Global asymmetry\"])\n",
    "        skin_changes = random.choice([\"No skin changes\", \"Skin thickening\", \"Skin retraction\"])\n",
    "        \n",
    "        basic_info = [\n",
    "            f\"Patient Name: {patient_name}\",\n",
    "            f\"Age: {age} years\",\n",
    "            f\"Sex: {sex}\"\n",
    "        ]\n",
    "        \n",
    "        if simple:\n",
    "            additional_sentences = [\n",
    "                \"Right breast examined. \" if source_rb else \"Right breast not examined. \",\n",
    "                \"Left breast examined. \" if source_lb else \"Left breast not examined. \",\n",
    "                \"Cancer identified in right breast.\" if cancer_rb else \"No cancer identified in right breast.\",\n",
    "                \"Cancer identified in left breast.\" if cancer_lb else \"No cancer identified in left breast.\",\n",
    "            ]\n",
    "        else:\n",
    "            additional_sentences = [\n",
    "                f\"Breast Density: {breast_density}\",\n",
    "                f\"Family History: {family_history}\",\n",
    "                f\"Previous Biopsies: {previous_biopsies}\",\n",
    "                f\"Hormone Therapy: {hormone_therapy}\",\n",
    "                \"Right breast examined.\" if source_rb else \"Right breast not examined.\",\n",
    "                \"Left breast examined:\" if source_lb else \"Left breast not examined.\",\n",
    "                f\"Right breast findings: {masses}. {calcifications}. {asymmetries}. {skin_changes}.\" if source_rb else \"\",\n",
    "                f\"Left breast findings: {masses}. {calcifications}. {asymmetries}. {skin_changes}.\" if source_lb else \"\",\n",
    "                \"Suspicious lesion identified in right breast, further investigation recommended.\" if cancer_rb else \"No suspicious lesions identified in right breast.\",\n",
    "                \"Suspicious lesion identified in left breast, biopsy advised.\" if cancer_lb else \"No suspicious lesions identified in left breast.\",\n",
    "                \"Follow-up mammogram recommended in 6 months.\" if (cancer_rb or cancer_lb) else \"Routine screening mammogram recommended in 1 year.\",\n",
    "                \"Additional ultrasound imaging may provide further clarification.\" if random.choice([True, False]) else \"\",\n",
    "                \"Patient reported breast pain, which is likely unrelated to imaging findings.\" if random.choice([True, False]) else \"\",\n",
    "                \"Dense breast tissue may lower the sensitivity of mammography.\" if breast_density in [\"c) Heterogeneously dense\", \"d) Extremely dense\"] else \"\",\n",
    "            ]\n",
    "        \n",
    "        additional_sentences = [s for s in additional_sentences if s]\n",
    "        \n",
    "        random.shuffle(additional_sentences)\n",
    "        \n",
    "        all_sentences = basic_info + additional_sentences\n",
    "        \n",
    "        report_text = \"\\n\".join(all_sentences)\n",
    "        \n",
    "        data.append([accession_number, report_text, source_rb, source_lb, cancer_rb, cancer_lb])\n",
    "    \n",
    "    return data\n",
    "\n",
    "def save_to_csv(data, filename):\n",
    "    with open(filename, 'w', newline='', encoding='utf-8') as csvfile:\n",
    "        writer = csv.writer(csvfile)\n",
    "        writer.writerow([\"Accession Number\", \"Report Text\", \"Source_RB\", \"Source_LB\", \"Cancer_RB\", \"Cancer_LB\"])\n",
    "        writer.writerows(data)\n",
    "\n",
    "simple = False\n",
    "num_records = 100\n",
    "random_data = generate_random_data(num_records, simple)\n",
    "\n",
    "save_to_csv(random_data, \"random_medical_data.csv\")\n",
    "\n",
    "print(f\"{num_records} random records have been generated and saved to 'random_medical_data.csv'\")\n",
    "\n",
    "print(\"\\nExample of the first 3 generated records:\")\n",
    "for record in random_data[:3]:\n",
    "    print(f\"\\nAccession Number: {record[0]}\")\n",
    "    print(f\"Report Text:\\n{record[1]}\")\n",
    "    print(f\"Source_RB: {record[2]}, Source_LB: {record[3]}, Cancer_RB: {record[4]}, Cancer_LB: {record[5]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pytorch2.4",
   "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": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}