diff --git a/__pycache__/api.cpython-37.pyc b/__pycache__/api.cpython-37.pyc index 1fffb8865ee29e17815a353075d3574315e0ebde..b673d274a46bd3b88af3b16dcff844f266f879af 100644 Binary files a/__pycache__/api.cpython-37.pyc and b/__pycache__/api.cpython-37.pyc differ diff --git a/__pycache__/give_color.cpython-37.pyc b/__pycache__/give_color.cpython-37.pyc new file mode 100644 index 0000000000000000000000000000000000000000..fdd666bba06371bea9abc851dc66216e167c05a4 Binary files /dev/null and b/__pycache__/give_color.cpython-37.pyc differ diff --git a/__pycache__/model.cpython-37.pyc b/__pycache__/model.cpython-37.pyc index f1ffe2c303b4415f93dd8b741b68649a3aa367d5..1174a9f10d46801817c080321b8377bbaac41312 100644 Binary files a/__pycache__/model.cpython-37.pyc and b/__pycache__/model.cpython-37.pyc differ diff --git a/api.py b/api.py index 2e974dc76d9ee5d0b157c34605b71b5df3f31ba8..eafaa290c5e6d8a4a7ea2a7a126e3137a28f0647 100644 --- a/api.py +++ b/api.py @@ -3,7 +3,7 @@ import copy import requests import uuid from flask import Flask, request, redirect, url_for, render_template, flash, jsonify, Blueprint - +from give_color import get_color from model import Oral oral_api = Blueprint('oral', __name__) @@ -38,7 +38,15 @@ def recognize(): print('' if verbose == 0 else 'verbose out:') oral = Oral(finding, conclusion, verbose = False if verbose == 0 else 1) data = oral.get_json() - return jsonify({'success': True, 'description': {'data': data}}), 200 + finding_list, conclusion_list, findind_replaced_word, conclusion_replaced_word, ImmunohistochemistryContent, MolecularResultsContent = oral.get_entity() + color_finding = get_color(finding, finding_list, findind_replaced_word, ImmunohistochemistryContent, + MolecularResultsContent) + color_conclusion = get_color(conclusion, conclusion_list, conclusion_replaced_word, + ImmunohistochemistryContent, MolecularResultsContent) + return jsonify({'success': True, + 'description': {'entity': data, + 'color': {'finding': color_finding, + 'conclusion': color_conclusion}}}), 200 except Exception as e: print("/n******ERROR SRART******/n") print(e) diff --git a/give_color.py b/give_color.py new file mode 100644 index 0000000000000000000000000000000000000000..2c2080852f52031865356860bf633ff0400b1440 --- /dev/null +++ b/give_color.py @@ -0,0 +1,226 @@ +## +from model import Oral +import re + +## +Finding = """ +左IV区淋巴结：1枚，直径2.5cm（1） +颏下：1枚，直径2cm（1） +左颊：部分上、下颌骨及颊部组织12*9*8cm，黏膜切面见一肿块6*3*2cm，灰白，界不清（1） +送检切缘：前内、前外、上内、上外、颊、翼内肌、翼外肌、腮腺、鼻底、鼻旁、鼻腔、软腭、舌根、颏突，均0.5cm +送检切缘：腮腺2,0.5cm +左颈大块：9*6*4cm，为肌肉、脂肪、血管，灰黄。 +右颈大块：10*8*4cm，一侧见一腺体4*3*2cm，灰黄分叶，余为脂肪血管。 +左I区: 1只直径1cm。 +左II区: 2只直径0.5-1cm。 +左III区: 4只直径0.8-1.3cm。 +左IV区: 3只直径1.2-1.8cm。 +左V区: 1只直径1cm。 +右I区: 3只直径1-3.5cm。 +右II区: 3只直径0.8-2cm。 +右III区: 4只直径0.5-1cm。 +右IV区: 4只直径0.5-1.2cm。 +右V区: 2只直径0.5-0.8cm。 +""" +Conclusion = """ +“左颊”黏膜鳞状细胞癌，高-中分化，DOI＞10mm +送检切缘：“腮腺”阳性（+）、“前内、前外、上内、上外、颊、翼内肌、翼外肌、鼻底、鼻旁、鼻腔、软腭、舌根、颏突、腮腺2”均阴性（-） +冰冻送检：“颏下淋巴结”1只内见鳞状细胞癌转移（+），侵犯至包膜外，“左IV区淋巴结”1只为软组织，伴出血 +“左颈大块”为软组织、“右颌下腺”轻度慢性炎 +送检淋巴结：“右I”1/3只有肿瘤转移（+），余及“左”“I”1只，“II”2只，“III”4只，“IV”3只，“V”1只，“右”“II”3只，“III”4只，“IV”4只，“V”2只均阴性（-） +免疫组化结果NI21-695 +AE1/AE3+ CKH+ CK5/6+ EGFR部分+ Ki67部分+ CD31- S-100- P16- +北院分子结果(NM2021-0316)：EGFR扩增探针 FISH（未见明显扩增（-）） +""" + +oral = Oral(Finding, Conclusion, verbose = False) + + +## +def get_color(text, entity_list, replaced_word, ImmunohistochemistryContent, MolecularResultsContent): + # print('replaced_word:', replaced_word) + # print(ImmunohistochemistryContent) + # print(MolecularResultsContent) + synonyms = [] + for item in entity_list: + if item['tag'] == 'ANATOMY': + words_anatomy = item['words'] + if item['words'] == 'I' or item['words'] == 'II' or item['words'] == 'III' or item['words'] == 'IV' or item[ + 'words'] == 'VI' or item['words'] == 'V' or item['words'] == '右' or item['words'] == '左': + words_anatomy = "“" + item['words'] + "”" + for i in re.finditer( + words_anatomy.replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + synonyms.append(i.span()) + if item['tag'] == 'OTHER': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + synonyms.append(i.span()) + + negative = [] + for i in re.finditer('无', text): + negative.append(i.span()) + for i in re.finditer('未见', text): + negative.append(i.span()) + for i in re.finditer('未发生', text): + negative.append(i.span()) + + numeracal = [] + for item in entity_list: + if item['tag'] == 'NUMBER': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + numeracal.append(i.span()) + if item['tag'] == 'SIZE': + # if item['words']=='12*9*8cm': + # print("^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^") + # print(text) + # for i in re.finditer(item['words'], text): + # print(i) + # print("^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^") + + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + numeracal.append(i.span()) + if item['tag'] == 'DOI': + if replaced_word == '': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), + text): + numeracal.append(i.span()) + else: + if item['words'] not in text: + if replaced_word == '大于': + doi_words = item['words'].replace(">", "大于") + elif replaced_word == '小于': + doi_words = item['words'].replace("<", "小于") + elif replaced_word == '大于等于': + doi_words = item['words'].replace("≥", "大于等于") + elif replaced_word == '小于等于': + doi_words = item['words'].replace("≤", "小于等于") + elif replaced_word == '＞': + doi_words = item['words'].replace(">", "＞") + elif replaced_word == '小于': + doi_words = item['words'].replace("<", "＜") + for i in re.finditer( + doi_words.replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), + text): + numeracal.append(i.span()) + + summary = [] + for item in entity_list: + if item['tag'] == 'ENE': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + summary.append(i.span()) + if item['tag'] == 'PN': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + summary.append(i.span()) + + classification = [] + for item in entity_list: + if item['tag'] == 'LEVEL': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + classification.append(i.span()) + + other = [] + for item in entity_list: + if item['tag'] == 'SQUAMOUS': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + other.append(i.span()) + if item['tag'] == 'INVASION': + for i in re.finditer( + item['words'].replace('*', '\*').replace('(', '\(').replace(')', '\)').replace('+', '\+'), text): + other.append(i.span()) + if ImmunohistochemistryContent != '': + for i in re.split('。| |\n', ImmunohistochemistryContent): + # print(i) + if i == "" or i == "\n": + continue + for j in re.finditer(i, text): + # print(j) + other.append(j.span()) + if MolecularResultsContent != '': + for i in re.split('。| |\n', MolecularResultsContent): + # print(i) + if i == "" or i == "\n": + continue + for j in re.finditer(i, text): + # print(j) + other.append(j.span()) + + synonyms = list(set(synonyms)) + negative = list(set(negative)) + numeracal = list(set(numeracal)) + summary = list(set(summary)) + classification = list(set(classification)) + other = list(set(other)) + + return { + '医学术语解析': other, + '同义词判断': synonyms, + '识别否定修饰词及其修饰对象': negative, + '数值解析': numeracal, + '对可归纳总结的信息进行解析后归纳判断': summary, + '对定义明确的分类变量进行赋值': classification + } + + +## +if __name__ == '__main__': + + finding_list, conclusion_list, findind_replaced_word, conclusion_replaced_word, ImmunohistochemistryContent, MolecularResultsContent = oral.get_entity() + + for i in finding_list: + print(i) + r1 = get_color(Finding, finding_list, findind_replaced_word, ImmunohistochemistryContent, MolecularResultsContent) + print("\n医学术语解析:") + for i in r1['医学术语解析']: + print(Finding[i[0]: i[1]], i[0], i[1]) + print("\n同义词判断:") + for i in r1['同义词判断']: + print(Finding[i[0]: i[1]], i[0], i[1]) + print("\n识别否定修饰词及其修饰对象:") + for i in r1['识别否定修饰词及其修饰对象']: + print(Finding[i[0]: i[1]], i[0], i[1]) + print("\n数值解析:") + for i in r1['数值解析']: + print(Finding[i[0]: i[1]], i[0], i[1]) + print("\n对可归纳总结的信息进行解析后归纳判断:") + for i in r1['对可归纳总结的信息进行解析后归纳判断']: + print(Finding[i[0]: i[1]], i[0], i[1]) + print("\n对定义明确的分类变量进行赋值:") + for i in r1['对定义明确的分类变量进行赋值']: + print(Finding[i[0]: i[1]], i[0], i[1]) + + print("\n*****************************************\n") + + for i in conclusion_list: + print(i) + r2 = get_color(Conclusion, conclusion_list, conclusion_replaced_word, ImmunohistochemistryContent, + MolecularResultsContent) + print("\n医学术语解析:") + for i in r2['医学术语解析']: + print(Conclusion[i[0]: i[1]], i[0], i[1]) + print("\n同义词判断:") + for i in r2['同义词判断']: + print(Conclusion[i[0]: i[1]], i[0], i[1], ) + print("\n识别否定修饰词及其修饰对象:") + for i in r2['识别否定修饰词及其修饰对象']: + print(Conclusion[i[0]: i[1]], i[0], i[1]) + print("\n数值解析:") + for i in r2['数值解析']: + print(Conclusion[i[0]: i[1]], i[0], i[1]) + print("\n对可归纳总结的信息进行解析后归纳判断:") + for i in r2['对可归纳总结的信息进行解析后归纳判断']: + print(Conclusion[i[0]: i[1]], i[0], i[1]) + print("\n对定义明确的分类变量进行赋值:") + for i in r2['对定义明确的分类变量进行赋值']: + print(Conclusion[i[0]: i[1]], i[0], i[1]) + +# ## +# print(Conclusion[235:236]) +# print(Conclusion[179:180]) +# print(Conclusion[161:162]) diff --git a/model.py b/model.py index f3ff0c7c4eee6218eae27bafe9853dcbd52ca288..cb8f6ce7e6c2749a1764edcc711016475dbe7b91 100644 --- a/model.py +++ b/model.py @@ -17,12 +17,39 @@ class Oral: self._Conclusion = '' self._Finding = '' + self.inversed = False + if '送检淋巴结' in ImagingConclusion: self._Conclusion = ImagingConclusion self._Finding = ImagingFindings else: self._Conclusion = ImagingFindings self._Finding = ImagingConclusion + self.inversed = True + + self.conclusion_replaced_word = '' + self.findind_replaced_word = '' + if "大于" in self._Conclusion: + self.conclusion_replaced_word = "大于" + if "小于" in self._Conclusion: + self.conclusion_replaced_word = "小于" + if "大于等于" in self._Conclusion: + self.conclusion_replaced_word = "大于等于" + if "小于等于" in self._Conclusion: + self.conclusion_replaced_word = "小于等于" + if "＞" in self._Conclusion: + self.conclusion_replaced_word = "＞" + + if "大于" in self._Finding: + self.findind_replaced_word = "大于" + if "小于" in self._Finding: + self.findind_replaced_word = "小于" + if "大于等于" in self._Finding: + self.findind_replaced_word = "大于等于" + if "小于等于" in self._Finding: + self.findind_replaced_word = "小于等于" + if "＞" in self._Finding: + self.findind_replaced_word = "＞" self._Conclusion = self._Conclusion.strip('"').strip() \ .replace("大于", ">").replace("小于", "<").replace("大于等于", "≥").replace("小于等于", "≤").replace("＞", ">").replace( @@ -694,6 +721,21 @@ class Oral: for i in self._y_pre_o: print(i) + def get_entity(self): + finding_list = [] + conclusion_list = [] + if self.inversed == True: + for i in self._y_pre: + finding_list.append({'tag': i[0], 'words': i[1]}) + for i in self._y_pre_o: + conclusion_list.append({'tag': i[0], 'words': i[1]}) + else: + for i in self._y_pre_o: + finding_list.append({'tag': i[0], 'words': i[1]}) + for i in self._y_pre: + conclusion_list.append({'tag': i[0], 'words': i[1]}) + return finding_list, conclusion_list, self.findind_replaced_word, self.conclusion_replaced_word, self.ImmunohistochemistryContent, self.MolecularResultsContent + def print_list_item(self, l): for i in l: print(i) @@ -735,21 +777,15 @@ class Oral: ## if __name__ == '__main__': Finding = """ -" 左颈大块：6*5*4cm，一侧见一腺体3*3*2cm，灰黄分叶，余为脂肪血管。 -左I区: 3只直径0.2-1.2cm。 -左II区: 3只直径0.5-1.2cm。 -左III区: 3只直径0.5-1cm。 -左IV区: 3只直径0.5-0.8cm。 -右I区：3只直径1-2cm。 -右II区: 3只直径1cm。 -右III区: 3只直径0.5-0.8cm。 -右IV区: 3只直径0.5-0.6cm。 -右颈淋巴：7*7*6cm，內见一腺体3*3*2cm，灰黄分叶，余为脂肪血管。" +"1、一带黏膜组织，1.8*0.8*0.5cm。灰黄。（2全）。 + +2、送检切缘：“前、后、左、右、底”均0.5cm。" """ Conclusion = """ -"“左颌下腺”慢性涎腺炎 -“右颌下腺”慢性涎腺炎 -送检淋巴结：“左”“I区”1/3只、“II区”1/3只、“III区”1/3只（肿瘤位于软组织内）及“右”“I区”2/3只（其中1只肿瘤侵犯至包膜外）、“II区”2/3只（肿瘤侵犯至包膜外）有肿瘤转移（+），余及“左IV区”3只、“右”“III区”3只、“IV区”3只均阴性（-）" +"“右口底”黏膜上皮中-重度异常增生及瘤样增生，癌变为鳞状细胞癌，高-中分化，DOI:小于5mm。 +送检切缘：“前、后、左、右、底”均阴性（-）。 + +南院免疫组化结果(I2021-3047)：CKH（+），CK5/6（+），P16（-），Ki67（约30－40%+），CD31（-），S100（-），EGFR（+），P53（少量+）。" """ oral = Oral(Finding, Conclusion) @@ -758,4 +794,4 @@ if __name__ == '__main__': oral.print_y_pred_o() print("-----------------") oral.get_Info() - print(oral.get_json()) + # print(oral.get_json())