RESUMO
Objective To investigate the time relationship of the change,and diagnostic accuracy and sensitivity between retinal light threshold fluctuations (LTF) and retinal nerve fiber layer (RNFL) and ganglion cell complex(GCC) thickness on high-risk primary open-angle glaucoma (POAG).Methods Totally 319 patients (319 eyes) with high-risk in POAG from the First Affiliated Hospital of Kunming Medical Universityand during December 2009 and December 2017,50 healthy individuals (50 eyes) as control were collected in this longitudinal cohort study.Visual field and OCT were reviewed every 6 months on the high-risk group and every 12 months on the control group.High-risk groups inclusion criteria:vertical C/D≥0.6;early visual field defect (according to glaucoma visual field damage GSS2 quantitative grading standards,mean deviation and pattern standard deviation of central field exceeds the border as an early visual field defect);continuous repeatable results.The first field and OCT results in the absence of visual field defects and C/D≥0.6,which were conformed reliability indicators and removed learning effects as a baseline.When patients achieve POAG diagnosis criteria first time which was recorded as a turning point.And they were divided into early group meanwhile were ended of follow-up.After the last follow-up,the inspection data was segmented counted in yearly interval.The changes of LTF,thickness of RNFL and GCC during the follow-up period in the early POAG group and the control group were observed.The loss rate and change rate in each period were compared for the assessment of their trends with time.Followed by calculation of the area under receiver operating curves (AUC) to compare the predicted value of POAG and the sensitivity at 95％ specificity in each period.Results After last follow-up,totally 67 patients 67 eyes (early POAG group,37 males and 30 females) were entered the turning point.The mean follow-up of the early POAG group and the control group were 6.6 and 6.4 years.The average RNFL thickness was 79.05± 8.09 μm,GCC thickness was 71.58 ± 8.41 μm,LTF was -6.05 ± 7.02 dB in early POAG group.The average RNFL thickness was 93.49± 6.24 μrm,GCC thickness was 79.72± 6.32 μm,LTF was-0.31 ± 0.58 dB in the control group.The differences of LTF and the thickness of RNFL and GCC were statistically significant (t=-5.97,-10.42,-5.60;P＜0.001).The AUC of RNFL,GCC thickness and LTF increased with time in the early POAG group.The sensitivity was gradually increased at 95％ specificity:5th year before to at turning point,RNFL thickness AUC was 0.15,0.65,0.71,0.77,0.85,0.92,and sensitivity was 20％,56％,61％,65％,70％,76％,respectively;GCC thickness AUC was 0.12,0.53,0.69,0.74,0.82,0.90,and sensitivity was 14％,53％,69％,74％,82％,90％,respectively;the AUC of LTF was 0.10,0.21,0.33,0.75,0.86,0.91,and sensitivity was 7％,17％,44％,65％,78％,87％,respectively.Conclusions The earliest time of structural functional damage of POAG is at the 4th year before confirmed,simultaneous RNFL diagnosis accuracy and sensitivity are better than GCC and LTF.The earliest time of visual functional damage of POAG is at the 2th year before confirmed,simultaneous LTF diagnosis accuracy and sensitivity are better than RNFL and GCC.
RESUMO
This paper explores the relationship between the cardiac volume and time, which is applied to control dynamic heart phantom. We selected 50 patients to collect their cardiac computed tomography angiography (CTA) images, which have 20 points in time series CTA images using retrospective electrocardiograph gating, and measure the volume of four chamber in 20-time points with cardiac function analysis software. Then we grouped patients by gender, age, weight, height, heartbeat, and utilize repeated measurement design to conduct statistical analyses. We proposed structured sparse learning to estimate the mathematic expression of cardiac volume variation. The research indicates that all patients' groups are statistically significant in time factor ( = 0.000); there are interactive effects between time and gender groups in left ventricle ( = 8.597, = 0.006) while no interactive effects in other chambers with the remaining groups; and the different weight groups' volume is statistically significant in right ventricle ( = 9.004, = 0.005) while no statistical significance in other chambers with remaining groups. The accuracy of cardiac volume and time relationship utilizing structured sparse learning is close to the least square method, but the former's expression is more concise and more robust. The number of nonzero basic function of the structured sparse model is just 2.2 percent of that of least square model. Hence, the work provides more the accurate and concise expression of the cardiac for cardiac motion simulation.