The aim of the current study was to investigate the spectra in the different organs of the rats which died of massive hemorrhage;to explore their spectral changes 15 days postmortem and the best mathematical model wit...The aim of the current study was to investigate the spectra in the different organs of the rats which died of massive hemorrhage;to explore their spectral changes 15 days postmortem and the best mathematical model with different band absorption ratio changes to postmortem interval(PMI);and to compare the spectral changes of different temperature.Thirty male Sprague-Dawley rats were sacrificed by cutting abdominal aorta,and the cadavers were divided equally and kept at 4 ℃,20 ℃ and 30 ℃ in the control chamber.From the same rat,seven different organs were sampled at intervals of 1-15 days postmortem,and then measured by Fourier transfom infrared(FTIR) spectrometer.Six mathematical model functions were explored.The absorbance of bands and band absorbance ratios of absorption peak in each organ showed a time-dependent increase or decrease,most band absorbance ratios remaining stable for 7-15 days postmortem.Cubic model functions of the various bands absorbance ratios against PMI showed a stronger related coefficient.The absorbance bands with obvious changes at 20 ℃ showed stabilized tendencies at 4 ℃ and significant changes at 30 ℃ within 15 days postmortem.In addition,FTIR spectroscopy revealed a time-dependent metabolic process,with potential of being used to estimate PMI during 7 days postmortem,which merits further investigation.展开更多
文摘目的探究基于MRI的影像组学模型及临床因素模型对缺血性脑卒中溶栓后出血转化风险的价值。材料与方法回顾性纳入并分析2个中心就诊的730例急性缺血性脑卒中(acute ischemic stroke,AIS)患者的临床及影像学资料。我们将中心1中的数据集的624例患者以7∶3的比例随机分为训练集(436例)与内部验证集(188例),对临床特征进行单因素多因素逻辑回归分析,筛选出与出血转化(hemorrhagic transformation,HT)有关的独立影响因素建立临床因素模型,基于MRI图像建立影像组学模型以及结合临床因素和影像组学特征的联合模型,预测AIS患者溶栓后发生HT的风险。中心2的106例患者数据用于外部验证。通过受试者工作特征(receiver operating characteristic,ROC)曲线及曲线下面积(area under the curve,AUC)评估模型的预测效能。为了比较不同模型之间AUC值差异是否有统计学意义,进一步采用DeLong检验进行统计分析,评估各模型之间的AUC差异的显著性。结果ROC曲线分析显示,训练集中临床因素模型、影像组学模型及联合模型预测出血转化的AUC分别为0.810(95%CI:0.756~0.864)、0.896(95%CI:0.865~0.928)和0.928(95%CI:0.899~0.958),在内部验证集中,临床因素模型、影像组学模型及联合模型预测出血转化的AUC分别为0.757(95%CI:0.671~0.843)、0.852(95%CI:0.791~0.913)和0.872(95%CI:0.809~0.935),此外,各模型在外部验证集中同样显示出良好的性能,AUC分别为0.720(95%CI:0.602~0.839)、0.804(95%CI:0.711~0.897)、0.828(95%CI:0.751~0.905)。决策曲线分析显示联合模型的净获益率最高。结论基于MRI的影像组学模型及临床因素模型对缺血性脑卒中后出血转化均有一定的预测价值;二者联合建立的联合模型预测价值最高,可应用于卒中后出血转化的风险预测。
基金funded by the Council of National Natural Science Foundation of China(No.81001350)
the Science and Technology Commission of Shanghai Municipality(No.09ZR1432900)
文摘The aim of the current study was to investigate the spectra in the different organs of the rats which died of massive hemorrhage;to explore their spectral changes 15 days postmortem and the best mathematical model with different band absorption ratio changes to postmortem interval(PMI);and to compare the spectral changes of different temperature.Thirty male Sprague-Dawley rats were sacrificed by cutting abdominal aorta,and the cadavers were divided equally and kept at 4 ℃,20 ℃ and 30 ℃ in the control chamber.From the same rat,seven different organs were sampled at intervals of 1-15 days postmortem,and then measured by Fourier transfom infrared(FTIR) spectrometer.Six mathematical model functions were explored.The absorbance of bands and band absorbance ratios of absorption peak in each organ showed a time-dependent increase or decrease,most band absorbance ratios remaining stable for 7-15 days postmortem.Cubic model functions of the various bands absorbance ratios against PMI showed a stronger related coefficient.The absorbance bands with obvious changes at 20 ℃ showed stabilized tendencies at 4 ℃ and significant changes at 30 ℃ within 15 days postmortem.In addition,FTIR spectroscopy revealed a time-dependent metabolic process,with potential of being used to estimate PMI during 7 days postmortem,which merits further investigation.
