193 lines (192 with data), 10.3 kB

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pytesseract\n",
    "pytesseract.pytesseract.tesseract_cmd = 'C:\\\\Program Files\\\\Tesseract-OCR\\\\tesseract.exe'\n",
    "tessdata_dir_config = '--tessdata-dir \"C:\\\\Program Files\\\\Tesseract-OCR\\\\tessdata\"'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import spacy\n",
    "from spacy import displacy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "colors = {'DOCTOR': '#55fdf9', 'PATIENT': '#54b52d', 'BIRTHDATE': '#8752a1', 'DATE': '#ff91af','TYPE': '#999900', 'ADRESS': '#cd0000', 'REFERING_DOCTOR': '#0071f1' }\n",
    "options = {'ents': ['DOCTOR', 'PATIENT', 'BIRTHDATE', 'DATE','TYPE', 'ADRESS', 'REFERING_DOCTOR'], 'colors':colors}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Anaconda\\lib\\site-packages\\spacy\\util.py:833: UserWarning: [W095] Model 'fr_pipeline' (0.0.0) was trained with spaCy v3.2 and may not be 100% compatible with the current version (3.2.0). If you see errors or degraded performance, download a newer compatible model or retrain your custom model with the current spaCy version. For more details and available updates, run: python -m spacy validate\n",
      "  warnings.warn(warn_msg)\n"
     ]
    }
   ],
   "source": [
    "nlp = spacy.load(\"./ner_model/output/model-best\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<span class=\"tex2jax_ignore\"><div class=\"entities\" style=\"line-height: 2.5; direction: ltr\">\n",
       "<mark class=\"entity\" style=\"background: #cd0000; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    Gien\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">ADRESS</span>\n",
       "</mark>\n",
       ", le \n",
       "<mark class=\"entity\" style=\"background: #ff91af; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    28/09/2010\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">DATE</span>\n",
       "</mark>\n",
       "</br>Monsieur \n",
       "<mark class=\"entity\" style=\"background: #54b52d; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    LEBROC Bernard\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">PATIENT</span>\n",
       "</mark>\n",
       "</br>BM/0077</br>Examen demandé par la Médecine Jaune.</br>\n",
       "<mark class=\"entity\" style=\"background: #999900; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    ECHOGRAPHIE\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">TYPE</span>\n",
       "</mark>\n",
       " VESICO-PROSTATIQUE ET RENALE (Hitachi Elite 8500</br>concept 26.02.2008)</br></br>Indication : néoplasie prostatique.</br>Rétention urinaire aigue.</br>RESULTAT :</br>Les coupes échographiques réalisées montrent des reins de volume et</br>d’échostructure normaux.</br></br>Il n’y à pas de dilatation des cavités pyélocalicielles ou de calcul rénal</br>visible.</br></br>Il existe une bonne différenciation cortico-médullaire et le parenchyme</br>rénal est d'épaisseur normale.</br></br>Il n’y à pas de syndrome tumoral tissulaire liquidien décelable.</br></br>Il n’y à pas d’épanchement péri ou para-rénal.</br>Au niveau pelvien, les bas uretères sont virtuels et la vessie est faiblement</br>remplie, transsonore à parois régulières sans image proliférative</br>endovésicale ou de calcul intraluminal.</br></br>La prostate par voie endorectale pèse environ 5 g ce jour et ses contours</br>sont réguliers et son échostructure est habituelle.</br></br>Le résidu post-mictionnel est non significatif.</br>CONCLUSION</br>L’examen retient un appareil urinaire échographiquement normal</br>ce jour et un petit résidu prostatique est d’environ 5g .</br>II n’y a pas d’autre anomalie.</br>\n",
       "<mark class=\"entity\" style=\"background: #55fdf9; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    Docteur J. NGUYEN HUU\n",
       "\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">DOCTOR</span>\n",
       "</mark>\n",
       "Nous avons conseillé au patient(e) de transmettre le compte rendu à son médecin traitant.</br></div></span>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with open('medical_report_text/repport_330.txt', 'r') as file:\n",
    "    data = file.read()\n",
    "    doc = nlp(data)\n",
    "displacy.render(doc, style=\"ent\", jupyter=True, options=options)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<span class=\"tex2jax_ignore\"><div class=\"entities\" style=\"line-height: 2.5; direction: ltr\">\n",
       "<mark class=\"entity\" style=\"background: #cd0000; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    Hirson\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">ADRESS</span>\n",
       "</mark>\n",
       " le \n",
       "<mark class=\"entity\" style=\"background: #ff91af; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    02/12/2019\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">DATE</span>\n",
       "</mark>\n",
       "</br>Mon Cher Confrère,</br>Voici les clichés du thorax de :</br>M. \n",
       "<mark class=\"entity\" style=\"background: #54b52d; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    HERVIEU VINCENT\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">PATIENT</span>\n",
       "</mark>\n",
       " 6160221066</br></br>IPP: 709805</br>ACCESSION NUMBER : 72134051</br>\n",
       "<mark class=\"entity\" style=\"background: #999900; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    Conventionnelle Radio\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">TYPE</span>\n",
       "</mark>\n",
       " DLP : 6.3 dGy.cm° Contraste : N/C cc N/C</br>INDICATION :</br>radio de contrôle</br>TECHNIQUE :</br>POUMONS FACE + PROFIL</br>RESULTAT :</br></br>Pas d’anomalie de la silhouette cardio-médiastinale. Pas d’anomalie en projection des hiles pulmonaires. Pas de</br>syndrome bronchique ni d’image de dilatation des bronches. Pas d’anomalie du parenchyme pulmonaire. Pas</br>d’anomalie pleurale. Pas d’anomalie du cadre osseux et des parties molles.</br></br>CONCLUSION DE L'EXAMEN DE Monsieur \n",
       "<mark class=\"entity\" style=\"background: #54b52d; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    HERVIEU Vincent\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">PATIENT</span>\n",
       "</mark>\n",
       " :</br>Pas d'anomalie radiologique visible.</br></br>RADIOLOGUE SIGNATAIRE : \n",
       "<mark class=\"entity\" style=\"background: #55fdf9; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    Radiologue MBAPTE WAMBA John\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">DOCTOR</span>\n",
       "</mark>\n",
       " , Validation électronique</br>MEDECIN REQUERANT : \n",
       "<mark class=\"entity\" style=\"background: #0071f1; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
       "    MIART LAURENT\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; vertical-align: middle; margin-left: 0.5rem\">REFERING_DOCTOR</span>\n",
       "</mark>\n",
       "</br></br>TRACABILITE HORAIRE : Demande : 15:59 - Protocole : 15:59 - Images reçues : 15:59 - Validation : 16:48</br></div></span>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with open('medical_report_text/repport_350.txt', 'r') as file:\n",
    "    data = file.read()\n",
    "    doc = nlp(data)\n",
    "displacy.render(doc, style=\"ent\", jupyter=True, options=options)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}