Use of an Artificial Neural Network to Construct a Model of Predicting Deep Fungal Infection in Lung Cancer Patients

Background: The statistical methods to analyze and predict the related dangerous factors of deep fungalinfection in lung cancer patients were several, such as logic regression analysis, meta-analysis, multivariate Coxproportional hazards model analysis, retrospective analysis, and so on, but the results are inconsistent. Materialsand Methods: A total of 696 patients with lung cancer were enrolled. The factors were compared employingStudent’s t-test or the Mann-Whitney test or the Chi-square test and variables that were significantly relatedto the presence of deep fungal infection selected as candidates for input into the final artificial neural networkanalysis (ANN) model. The receiver operating characteristic (ROC) and area under curve (AUC) were used toevaluate the performance of the artificial neural network (ANN) model and logistic regression (LR) model. Results:The prevalence of deep fungal infection from lung cancer in this entire study population was 32.04%(223/696),deep fungal infections occur in sputum specimens 44.05%(200/454). The ratio of candida albicans was 86.99%(194/223) in the total fungi. It was demonstrated that older (≥65 years), use of antibiotics, low serum albuminconcentrations (≤37.18g /L), radiotherapy, surgery, low hemoglobin hyperlipidemia (≤93.67g /L), long time ofhospitalization (≥14days) were apt to deep fungal infection and the ANN model consisted of the seven factors.The AUC of ANN model(0.829±0.019)was higher than that of LR model (0.756±0.021). Conclusions: The artificialneural network model with variables consisting of age, use of antibiotics, serum albumin concentrations, receivedradiotherapy, received surgery, hemoglobin, time of hospitalization should be useful for predicting the deepfungal infection in lung cancer.     

晚期肺癌患者肺部感染死亡危险因素分析

目的 分析肺癌伴肺部感染患者死亡的相关危险因素。方法　回顾2010年1月—2017年5月收治的186例晚期肺癌患者临床资料,其中肺部感染死亡者52例,并收集肺部感染死亡的相关危险因素。结果　晚期肺癌患者肺部感染死亡率为27.96%（52/186）。在诸多因素中,年龄（P=0.039）、BMI值（P=0.000）、组织分化程度低（P=0.003）、功能状态评分（P=0.022）、转移部位≥3（P=0.029）、合并症（P=0.034）、多重耐药菌感染（P=0.000）、血红蛋白浓度<90 g/L（P=0.014）等在死亡组和非死亡组之间差异具有统计学意义。结论　年龄（≥60岁）、BMI值（<20 kg/m²）、组织分化程度（低）、抗肿瘤治疗药物（≥3种）、靶向药物治疗、转移部位（≥3）、合并症、多重耐药菌感染、血红蛋白浓度（<90 g/L）等是晚期肺癌患者肺部感染死亡的主要危险因素。     

肺癌患者合并肺部真菌感染的临床分析

目的 分析肺癌患者并发肺部真菌感染的影响因素及临床特点,以有效预防和控制感染。方法收集新疆肿瘤医院2007年1月~12月出院的872例肺癌患者的临床资料,对其中合并肺部真菌感染的87例资料总结,分析真菌感染的影响因素及真菌种类特点。结果 872例肺癌患者中,肺部真菌感染87例,感染发生率9.9%。真菌类型主要为念珠菌菌属（96.6%）,其中白色念珠菌（81%）为主要菌种,主要影响因素有年龄≥50岁,Ⅲ～Ⅳ期的中晚期肺癌患者、住院时间≥14天、化疗、放疗,侵袭性操作、白细胞减少≥Ⅲ度,长时间使用抗生素及激素(P＜0.05)。而患者的性别,肺癌的病理分型,是否行手术治疗与肺部真菌感染无关(P＞0.05)。结论 减少易感因素,及时治疗是降低肺癌患者真菌感染的有效措施。     

Lung Cancer Risk Prediction Method Based on Feature Selection and Artificial Neural Network

A method to predict the risk of lung cancer is proposed, based on two feature selection algorithms: Fisherand ReliefF, and BP Neural Networks. An appropriate quantity of risk factors was chosen for lung cancer riskprediction. The process featured two steps, firstly choosing the risk factors by combining two feature selectionalgorithms, then providing the predictive value by neural network. Based on the method framework, an algorithmLCRP (lung cancer risk prediction) is presented, to reduce the amount of risk factors collected in practicalapplications. The proposed method is suitable for health monitoring and self-testing. Experiments showed itcan actually provide satisfactory accuracy under low dimensions of risk factors.     

恶性肿瘤患者院内感染相关因素分析

目的分析肿瘤住院患者发生院内感染的相关因素。方法对2003年1月～2006年12月住院并发院内感染的108例恶性肿瘤患者进行回顾性调查分析。结果108例院内感染患者中,发病年龄在40～69岁之间较多;住院时间在10～29天之间患者发病为多;发病部位和发病病种以呼吸系统占首位;院内感染可发生于出现骨髓抑制或无骨髓抑制患者,但与是否进行相关侵入性治疗无关,而与激素、抗生素的使用相关。结论对住院的肿瘤患者应监测有无院内感染的发生,严格掌握激素、抗生素的使用适应证,降低院内感染发生率。     

肺癌患者医院获得性肺炎的病原菌和临床特点

目的 探讨肺癌患者医院获得性肺炎的病原菌和临床特点。方法 回顾性分析医院获得性肺炎126例患者的病例资料。结果 患者的临床表现不典型，痰培养分离出102株病原菌，其中革兰氏阴性菌占54.9%，革兰氏也是性菌占30.4%，真菌占14.7%，耐药情况较严重。结论 老年和机体免疫抑制是接受化疗或放疗的肺癌患者易发生医院获得性肺炎的主要因素。对医院获得性肺炎患者应及时诊断，并根据细菌药敏结果选用抗生素。     

影响肺癌化疗患者发生肺部感染的危险因素分析

目的探讨影响肺癌化疗患者发生肺部感染的危险因素。方法对296例肺癌化疗患者进行回归性分析,收集临床资料,统计肺部感染发生率,应用单因素检验及logistic回归模型筛选出影响肺部感染的危险因素。结果296例患者中54例发生肺部感染,感染率为18. 24%。单因素分析显示,肺癌化疗患者肺部感染发生与年龄、吸烟史、高血压、糖尿病、TNM分期、治疗方法、化疗用药方案、化疗周期、化疗前KPS评分、白蛋白含量有关(P <0. 05)。多因素Logisitic回归分析显示,年龄≥60岁、TNM分期为Ⅲ~Ⅳ期、化疗用药≥2种、化疗周期> 4个、化疗前KPS评分<80分、白蛋白水平均为肺癌化疗患者肺部感染发生的危险因素(P <0. 05)。结论肺癌化疗患者肺部感染发生率较高,危险因素众多,应积极采取针对措施加强控制。     

Assessing Breast Cancer Risk with an Artificial Neural Network

Objectives: Radiologists face uncertainty in making decisions based on their judgment of breast cancer risk.Artificial intelligence and machine learning techniques have been widely applied in detection/recognition of cancer.This study aimed to establish a model to aid radiologists in breast cancer risk estimation. This incorporated imagingmethods and fine needle aspiration biopsy (FNAB) for cyto-pathological diagnosis. Methods: An artificial neuralnetwork (ANN) technique was used on a retrospectively collected dataset including mammographic results, riskfactors, and clinical findings to accurately predict the probability of breast cancer in individual patients. Area underthe receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity, and positive and negative predictivevalues were used to evaluate discriminative performance. Result: The network incorporating the selected featuresperformed best (AUC = 0.955). Sensitivity and specificity of the ANN were respectively calculated as 0.82 and 0.90.In addition, negative and positive predictive values were respectively computed as 0.90 and 0.80. Conclusion: ANNhas potential applications as a decision-support tool to help underperforming practitioners to improve the positivepredictive value of biopsy recommendations.     

晚期非小细胞肺癌合并肺部感染的危险因素及相关性分析

目的 探讨晚期非小细胞肺癌（non-small cell lung cacer,NSCLC）合并肺部感染的危险因素。方法 通过回顾性临床研究的方法,收集2011年6月至2012年6月北京友谊医院肿瘤科收治107例住院非小细胞肺癌患者的临床资料,其中合并肺部感染患者33例,收集肺部感染的相关因素并进行统计。计数资料采用卡方检验,计量资料采用t检验进行分析,应用Logistic回归模型对非小细胞肺癌合并肺部感染的危险因素进行分析。结果 107例非小细胞肺癌患者中,男性70例,肺部感染发生率为27.1%（19/70）,女性37例,肺部感染发生率为37.8%（14/37）。在诸多因素中,体温状态（P<0.01）、血清白蛋白水平（P=0.01）、化疗（P=0.01）、粒细胞缺乏（P=0.04）,以及基础疾病的罹患等在肺部感染组和非感染组之间差异有统计学意义。结论 患者血清白蛋白水平降低、粒细胞减少、化疗、既往有糖尿病史等,均为晚期非小细胞肺癌合并肺部感染的主要危险因素。     

人工神经网络在宫颈癌预后预测中的应用

目的探讨人工神经网络在宫颈癌术后5年生存期预测中的应用。方法收集125例宫颈癌患者的临床病理资料及治疗随访信息,按照4∶1的比例,随机分为训练组(100例)和测试组(25例),分别采用Logistics回归分析,筛选单因素分析有统计学意义的因素建立Logistics回归模型和概率神经网络模型（PNN）,用训练组训练网络模型,用测试组检测网络模型。结果PNN模型的准确性92%,敏感度为75%,特异性为95.23%,Logistics回归模型的准确性为84%,敏感度为50.0%,特异性为82.61%。结论神经网络在生存分析中有很大的灵活性;在模型中可以容纳非线性效应,不需要对数据的随机特征如分布等作出假设,不要求满足H0假定,具有较广泛的应用前景。     

老年肺癌住院死亡患者肺部感染影响因素分析

目的 探讨晚期老年肺癌住院患者死亡原因及肺部感染影响因素。方法 回顾性分析88例晚期老年肺癌住院患者死亡病例临床资料。结果 本组肺癌患者均≥65岁,功能状态评分>2分者占81.82%,90%以上至少有一处转移,基础病多,住院期间侵入性操作多,肺部感染是最主要的并发症,预后差,占死亡原因68.18%。病原菌以革兰氏阴性菌为主,常伴有真菌感染。3种以上抗肿瘤治疗、胸腔积液、肺部转移、住院时间长是肺部感染的危险因素。结论 肺部感染是晚期老年肺癌患者主要并发症和死亡原因。尽早发现转移灶,有效处理肺部及胸膜转移,减少易感因素,根据痰培养结果及时控制感染,是提高晚期老年肺癌患者治疗效果的关键。     

肿瘤患者医院感染相关因素调查

目的 :了解肿瘤患者医院感染发病率及漏报率 ,分析发生医院感染及漏报的原因 ,以便加强监控。方法 :查阅 1997～ 2 0 0 1年住院病例 2 2 6 17份 ,逐份进行回顾性调查分析。结果 :5年中总出院人数 2 2 6 17份 ,发生院内感染 974例次 ,感染率为 4 13%。各感染部位发生率以呼吸道为最高 ,占4 7 13% ,其次为切口感染 ,占 15 30 %。各月份感染率以 1月份为最高 ,占 5 0 6 %。漏报率 1997年为最高达 70 85 % ,2 0 0 1年为最低 7 19%。结论 :合理应用抗生素 ,减少各种侵袭性操作 ,提高医务人员对医院感染重要性的认识。     

数据挖掘技术在肺癌危险度预测模型中的应用

目的 使用数据挖掘技术建立肺癌危险度预测模型,比较C5.0决策树与人工神经网络用于肺癌风险预测的性能,并探讨其在肺癌风险预测中的价值.方法 选择180例肺癌患者及240例肺良性疾病患者,收集肺癌相关危险因素和临床症状共17个自变量,建立C5.0决策树与人工神经网络模型,比较模型的预测性能.结果 共收集420份病历资料,...     

Predicting the Survival of Gastric Cancer Patients Using Artificial and Bayesian Neural Networks

Introduction and purpose: In recent years the use of neural networks without any premises for investigation of prognosis in analyzing survival data has increased. Artificial neural networks (ANN) use small processors with a continuous network to solve problems inspired by the human brain. Bayesian neural networks (BNN) constitute a neural-based approach to modeling and non-linearization of complex issues using special algorithms and statistical methods. Gastric cancer incidence is the first and third ranking for men and women in Iran, respectively. The aim of the present study was to assess the value of an artificial neural network and a Bayesian neural network for modeling and predicting of probability of gastric cancer patient death. Materials and Methods: In this study, we used information on 339 patients aged from 20 to 90 years old with positive gastric cancer, referred to Afzalipoor and Shahid Bahonar Hospitals in Kerman City from 2001 to 2015. The three layers perceptron neural network (ANN) and the Bayesian neural network (BNN) were used for predicting the probability of mortality using the available data. To investigate differences between the models, sensitivity, specificity, accuracy and the area under receiver operating characteristic curves (AUROCs) were generated. Results: In this study, the sensitivity and specificity of the artificial neural network and Bayesian neural network models were 0.882, 0.903 and 0.954, 0.909, respectively. Prediction accuracy and the area under curve ROC for the two models were 0.891, 0.944 and 0.935, 0.961. The age at diagnosis of gastric cancer was most important for predicting survival, followed by tumor grade, morphology, gender, smoking history, opium consumption, receiving chemotherapy, presence of metastasis, tumor stage, receiving radiotherapy, and being resident in a village. Conclusion: The findings of the present study indicated that the Bayesian neural network is preferable to an artificial neural network for predicting survival of gastric cancer patients in Iran.     

Determining of Prognostic Factors in Gastric Cancer PatientsUsing Artificial Neural Networks

Background & Objectives: The aim of this study is to determine diagnostic factors for Iranian gastric cancerpatients and their importance using artificial neural network and Weibull regression models. Methods: Thisstudy was a historical cohort study with data gathered from 436 registered gastric cancer patients who underwentsurgery between 2002 and 2007 at the Taleghani Hospital (a referral center for gastrointestinal cancers), Tehran,Iran. In order to determine risk factors and their importance, neural network and Weibull regression modelswere used. Results: The Weibull regression analysis showed that lymph node metastasis and histopathology oftumor were selected as important variables. Based on the neural network model, staging, lymph node metastasis,histopathology of tumor, metastasis, and age at diagnosis were selected as important variables. The true predictionof neural network was 82.6%, and for the Weibull regression model, 75.7%. Conclusion: The present studyshowed that the neural network model is a more powerful tool in determining the important variables for gastriccancer patients compared to Weibull regression model. Therefore, this model is recommended for determiningof risk factors of such patients.     

原发性肝癌患者医院感染的相关因素分析

目的探讨住院原发性肝癌患者医院感染的相关因素。方法记录122例原发性肝癌患者医院感染的相关因素并进行分析。结果 122例患者医院感染发生率为23.0%(28/122),感染部位构成比依次为:呼吸道、腹膜炎、泌尿系、消化道、皮肤及其它;28例医院感染患者共检出25株病原体,其中革兰氏阴性菌占48.0%,真菌占28.0%,革兰氏阳性菌占24.0%;贫血、肝功能分级高、3次以上侵入性操作、合并糖尿病、放化疗史及年龄≥60岁患者医院感染发生率高(P<0.05)。结论原发性肝癌患者医院感染的发生与患者贫血、肝功能分级、侵入性操作次数、血糖水平等因素相关。     

膀胱癌术后医院感染患者细胞因子与急性时相反应蛋白的变化观察

目的观察及研究膀胱癌术后医院感染患者细胞因子与急性时相反应蛋白的变化情况。方法选取行手术治疗且发生术后医院感染的48例膀胱癌患者为A组,选取同时期的未发生医院感染的48例膀胱癌术后患者为B组,同时期的48例健康人员为C组,然后将三组的血清细胞因子及急性时相反应蛋白水平进行比较,并比较A组中不同感染程度患者的上述血清指标检测结果。结果 A组的血清IL-2水平均低于B组及C组,B组低于C组,其他血清细胞因子与急性时相反应蛋白水平高于B组及C组,B组高于C组,且A组中重度感染患者高于轻度及中度感染患者,中度感染患者则高于轻度感染患者,P均<0.05,均有显著性差异。结论膀胱癌术后医院感染患者细胞因子与急性时相反应蛋白均呈现高表达状态,且不同感染程度患者的表达水平也存在明显差异。     

X刀治疗肺癌脑转移39例疗效分析

目的评价X刀治疗肺癌脑转移瘤的效果并分析其预后因素。方法回顾性分析1994年2月-2002年11月在巴黎Tenon医院放疗科行X刀治疗的肺癌脑转移患者39例，共有转移灶49个，其中13例患者在行X刀前或后接受了全脑放疗（WBRT）。计算患者的生存期及局控率，并对潜在预后因素分别进行单因素和多因素分析。结果39例患者总体局控率和1年局控率分别为95．5％和95．0％，中位生存期为14个月。KPS〉70分、年龄≤60岁是独立的预后因素。结论X刀是肺癌脑转移的有效治疗方式，多种因素影响其预后。