Abstract： Diseases prognosis is often influenced by multiple factors, and some intricate non-linear relationships exist among those factors. Artificial neural network （ANN）, an artificial intelligence model, simulates the work mode of biological neurons and has a b capability to analyze multi-factor non-linear relationships. In recent years, ANN is increasingly applied in clinical medical fields, especially for the prediction of disease prognosis. This article focuses on the basic principles of ANN and its application in disease prognosis research.
Objective To improve the awareness of acute exacerbation of idiopathic pulmonary fibrosis ( AEIPF) and discuss its clinical characteristics, diagnosis, treatment and outcome. Methods The clinical data of patients with AEIPF from June 2006 to June 2011 in 11 hospitals in Jiangsu were collected and analyzed. Resluts There were 18 males and 3 females in the AEIPF patients with mean age of ( 67.4 ± 8.1) years. The duration from IPF diagnosis was ( 7.4 ±8.2) months. The duration of acute symptom before admission was ( 7.0 ±5.3) days. The distribution pattern of new groud-glass opacity was peripheral in 3 patients,multifocal in 5 patients, and diffuse in13 patients. All patients were treated with corticosteroid pulse therapy. Nine patients survived and 12 patients died. The mortality rate was 57.1% . Conclusions AEIPF progresses quickly and the mortality rate is very high. Corticosteroid pulse therapy is the mainstay of therapy in AEIPF patients.
Objective To study the factors affecting the prognosis of vitrectomy in mechanically injured eyes. Methods One hundred and thirty-Seven eyes undergone vitrectomy were in cluded.Recorded to EPIINFO data base were visual acuity (VA) immediately after injury,type and location of injury,the interval from injury to surgical intervention,retinal detachment,basic surgical maneuvers,proliferative vitreo-retinopathy (PVR),stage of trauma,complications,final VA,and final retinal status.Data were analyzed by SAS. Results We got anatomic success in 107 eyes (75.9%).Final VA of 74 eyes (54.1%) were better than 0.02.Multiple-variable analysis showed that the factors significantly affecting prognosis of vitrectomied eyes of mechanical injury were retinal detachment (RD) before surgery,delay of surgical intervention,complexity of surgical maneuvers,complications after surgery. Conclusions RD before surgery,delayed surgical intervention,complexity of surgery and post vitrectomy complications are the most important factors contributing to the poor prognosis of VA and failure of surgery. (Chin J Ocul Fundus Dis，2000，16：139-212)
目的:探讨引起早产的原因和影响预后的因素。方法:对我院126例早产进行回顾性分析。结果:早产发生率为5.0%,胎膜早破(PROM)、妊娠高血压综合征(PIH)是引起早产的主要因素。剖宫产与阴道分娩对早产儿结局无明显差异。破膜时间12h内和12h以上两组间宫内感染发生率差异无统计学意义。结论:早产儿预后与早产病因相关。不同分娩方式对新生儿结局没有影响。
Objective To explore the influencing factors of visual prognosis of macular edema secondary to branch retinal vein occlusion (BRVO-ME) after treatment with ranibizumab, and construct and verify the nomogram model. MethodsA retrospective study. A total of 130 patients with BRVO-ME diagnosed by ophthalmology examination in the Department of Ophthalmology, Liuzhou Red Cross Hospital from January 2019 to December 2021 were selected in this study. All patients received intravitreal injection of ranibizumab. According to the random number table method, the patients were divided into the training set and the test set with a ratio of 3:1, which were 98 patients (98 eyes) and 32 patients (32 eyes), respectively. According to the difference of logarithm of the minimum angle of resolution (logMAR) best corrected visual acuity (BCVA) at 6 months after treatment and logMAR BCVA before treatment, 98 patients (98 eyes) in the training set were divided into good prognosis group (difference ≤－0.3) and poor prognosis group (difference ＞－0.3), which were 58 patients (58 eyes) and 40 patients (40 eyes), respectively. The clinical data of patients in the two groups were analyzed, univariate and multivariate logistic regression analysis were carried out for the different indicators, and the visualization regression analysis results were obtained by using R software. The consistency index (C-index), convolutional neural network (CNN), calibration curve and receiver operating characteristic (ROC) curve were used to verify the accuracy of the nomogram model. ResultsUnivariate analysis showed that age, disease course, outer membrane (ELM) integrity, elliptical zone (EZ) integrity, BCVA, center macular thickness (CMT), outer hyperreflective retinal foci (HRF), inner retina HRF, and the blood flow density of retinal deep capillary plexus (DCP) were risk factors affecting the visual prognosis after treatment with ranibizumab in BRVO-ME patients (P＜0.05). Multivariate logistic regression analysis showed that course of disease, ELM integrity, BCVA and outer HRF were independent risk factors for visual prognosis after ranibizumab treatment for BRVO-ME patients (P＜0.05). The ROC area under the curve of the training set and the test set were 0.846[95% confidence interval (CI) 0.789-0.887) and 0.852 (95%CI 0.794 -0.873)], respectively; C-index were 0.836 (95%CI 0.793-0.865) and 0.845 (95%CI 0.780-0.872), respectively. CNN showed that the error rate gradually stabilized after 300 cycles, with good model accuracy and strong prediction ability. ConclusionsCourse of disease, ELM integrity, BCVA and outer HRF were independent risk factors of visual prognosis after ranibizumab treatment in BRVO-ME patients. The nomogram model based on risk factors has good differentiation and accuracy.
目的:探讨心肺复苏循环恢复患者早期评估预后的相关因素。方法:对56例心肺复苏循环恢复患者进行病例回顾分析,分别记录患者年龄、性别以及心肺复苏循环恢复1小时内的瞳孔直径、格拉斯高昏迷评分、血WBC计数、血清肌酐Cr、血清丙氨酸氨基转移酶ALT、肌酸磷酸激酶、D-二聚体定性、血钙、血钾、血清淀粉酶、复苏后1小时内是否使用亚低温治疗、pH值、动脉血氧分压PaO2、动脉血二氧化碳分压PaCO2、血葡萄糖、复苏时间等, 采用多因素logistic回归模型分析心肺复苏循环恢复患者的预后因素。结果:回顾56例心肺复苏循环恢复患者病例, logistic回归分析发现血清淀粉酶、血糖和复苏时间对早期评估心肺复苏循环恢复患者的预后有意义。结论:复苏后1小时内血清淀粉酶、血糖和复苏时间对早期评估心肺复苏循环恢复患者的预后有价值。
目的：探讨成人化脓性脑膜炎合理有效的诊治方法和影响预后的因素。方法：对我院2001年1月～2007年4月成人化脓性脑膜炎患者的病历资料进行回顾性分析。结果：20例痊愈，2例自动出院，2例死亡。结论：早期诊断，早期合理使用抗生素，适当使用肾上腺皮质激素和脱水剂，改善脑代谢药物，患者大多预后好。诊治开始时间、患者年龄及伴发疾病是决定预后的关键。
ObjectiveTo explore effect of preoperative prognostic nutritional index (PNI) on clinically related postoperative pancreatic fistula (CR-POPF) after distal pancreatectomy (DP) and analyze its influencing factors in order to provide a basis for clinical prediction of CR-POPF. MethodsThe clinicopathologic data of patients who successfully completed DP in the Affiliated Hospital of Xuzhou Medical University and met the inclusion and exclusion criteria of this study from January 1, 2017 to January 31, 2021 were collected retrospectively. The preoperative PNI value was calculated and the optimal cut-off value was obtained according to the receiver operative characteristic (ROC) curve. The patients were divided into low and high PNI based on the optimal cut-off value. The clinicopathologic characteristics were compared between the patients with low and high PNI and CR-POPF or not. At the same time, multivariate logistic regression was used to analyze the influencing factors of CR-POPF. ResultsA total of 143 patients who met the inclusion and exclusion criteria were included in this study. The CR-POPF occurred in 33 cases (23.08%) after DP, and the average preoperative PNI was 52.26 (39.20–65.10), the optimal cut-off value of PNI was 50.55, with 49 cases in the low PNI group and 94 cases in the high PNI group. In patient with low PNI, the proportions of patients aged ≥65 years and with CR-POPF were higher than those with high PNI (P<0.05). In the patients with CR-POPF, the proportions of patients with soft pancreatic texture and with low preoperative PIN were higher than those without CR-POPF (P<0.05). Further, the multivariate logistic regression showed that the the preoperative low PNI (OR=5.417, P<0.001) and soft pancreatic texture (OR=4.126, P=0.002) increased the risk of CR-POPF. ConclusionLow preoperative PNI and soft pancreatic texture increase risk of CR-POPF after DP, and it is necessary to preoperatively evaluate PNI status of patients.
Toexploretheinfluenceoflocalmassiveexcisionbeforeradicalsurgeryonprognosisofpatientswithbreastcancer,wecomparedtheprognosisbetweenthegroupunderwentlocalresectionpriortoradicalsurgery(106cases)andthegorupwithdirectradicalresection(143cases).Theresultsshowedthatthelocalrecurrencerate,distancemetastasisrateandthesurvivalrateat3,5yearsofthegroupunderwentlocalexcisionpriortoradicalsurgerywere16.0%,26.4%,79.2%,71.7%respectivelyandofthegroupunderwentdirectradicalresectionwere4.9%,16.1%,89.5%,82.5%respectively,thedeferencewassignificant(Plt;0.01,0.05,0.05,0.05respectively).Theresultsindicatethatthelocalexcisionbeforeradicalsurgerycanaffecttheprognosisofpatientswithbreastcancer.
【Abstract】Objective Stromal cell-derived factor-1(SDF-1, CXCL12) is a member of the CXC subfamily of chemokines which, through its cognate receptor (CXCR4), plays an important role in tumor invasion and metastasis. This study analyzed quantitatively the expression of SDF-1 and its relation with clinicopathologic feature and clinical outcome in human breast cancer.Methods Expression of SDF-1 mRNA in 8 breast cancer cell lines, an endothelial cell line HECV and a fibroblast cell MRC5 was studied by using RT-PCR. In addition, the expression of SDF-1 was investigated at both protein (immunohistochemistry) and mRNA(real-time PCR) levels in a group of human normal mammary(n=32) and tumour tissues(n=120). Results SDF-1 expression was identified in MRC5, MDA-MB435s, MDA-MB436, MCF7 cell lines, breast tumour and normal tissues. Significantly higher level of SDF-1 was seen in lymph node positive than in lymph node negative tumours (399.00±210.00 vs 0.89±0.47), P=0.048. The level of SDF-1 expression in patients who developed local recurrence or metastasis, or patients who died of breast cancer was higher than in patients who were disease free as well, (670.00±346.00 vs 0.83±0.35), P=0.01. It was most notable that level of SDF-1 was significantly correlated with over survival (P=0.01) and incidence free survival (P=0.035, by Cox proportion analysis).Conclusion SDF-1 is a factor that is expressed in both stromal cells and some breast cancer cells. Its level are correlated with lymph node involvement, prognosis and survival in patients with breast cancer. SDF-1 may therefore have a potential prognostic value in breast cancer.