1.Study of relation between sleep architecture and cognitive behavior in children with epilepsy
Wenjing JIANG ; Zhaofu CHI ; Lin MA
Journal of Clinical Neurology 1993;0(03):-
Objective To observe the relations of sleep structure changes and cognitive behavior abnormalities in children with idiopathic epilepsy.Methods All night polysomnographies, day attention test and Achenbach child behavior checklist were done on 64 children with idiopathic epilepsy and 20 healthy controls the requirement. Spearman correlations were made to evaluate the correlations between the parameters of sleep structure and the results of attention and cognitive behavior abnormalities.Results All children with epilepsy had longer stage Ⅰ sleep percentage and latency of rapid eye movement (REM) sleep compared with controls (all P
2.Imaging findings of acute urinary tract obstruction and perinephritis
Zhaofu JIANG ; Yang ZHOU ; Bing LU ; Yufei CHEN ; Lili KONG ; Shu JIANG
Journal of Practical Radiology 2016;32(10):1566-1569
Objective To analyze the reasons and imaging features of acute urinary tract obstruction and perinephritis and to evaluate the diagnostic efficacy of CT and MR in this kind of patients.Methods Imaging features of 72 patients with acute urinary tract obstruction were analyzed.The advantages of CT and MR were evaluated in diagnosis of the diseases.Results In all 70 patients (72 sites)with ureteral calculus,the accurate rates of diagnosis were 100% on CT and 59.7% on MR.In the 72 patients with perinephritis,the detection rate of effusion in perinephric soft tissue on MR was better than that on CT.Conclusion CT or MR has individually diagnostic advantage in acute urinary tract obstruction and perinephritis,CT is better in ureteral calculus and MR in perinephritis.
3.Application of magnetic resonance images in patients with adrenoleukodystrophy
Wenjing JIANG ; Zhaofu CHI ; Binfeng DU ; Lin MA ; Qingshi ZENG ; Jinwen HOU ; Xiaohua WANG
Chinese Journal of Neurology 2008;41(2):106-109
Objective To explore the role of proton MR spectroscopic imaging(1H-MRSI)and diffusion tensor imaging(DTI)in the diagnosis of adrenoleukodystrophy and to demonstrate the involvement of fibers by using the technique of DTT.Methods 1H-MRSI,DTI and routine imaging examinations were performed in 6 patients with ALD.The values of NAA,Cho,Cr,ADC,and FA were evaluated in different regions of lesion.The involved fibers were demonstrated by using the technique of DTT.Results The ratios of NAA/Cr(0.55±0.19)and NAA/Cho(0.22±0.11)were lower(F=7.693,7.751),and Cho/Cr(2.54±0.37)was higher(F=6.348)in the initial lesions,where higher ADC values(1.49±0.36,F=5.226)and the lowest FA values(0.21±0.08,F=5.139)were also observed(P<0.05).The decreases of NAA/Cr(1.16±0.03)and NAA/Cho(0.45±0.17)in adjacent regions were more distinct than those of remote regions(t=1.769,1.842,P<0.05).In the developing regions,the ADC values(0.89±0.03)were lower and the FA values(0.45±0.07)were the highest.There was negative correlation between NAA/Cho and ADC values(r=-0.71,P<0.05),and there was positive correlation between NAA/Cho and FA values(r=0.31,P<0.05).Discontinuations and fragments of fibres were observed in corpus callosum and pyramidal tract.Conclusions Combination of 1H-MRSI and DTI can offer a sensitive method for the early diagnosis and monitor the progress of white matter.DTT can be used to directly observe the involvement of fibers.
4.Advances in the application of non-contrast CT radiomics and machine learning in the diagnosis and treat-ment process of hypertensive Intracerebral hemorrhage
Yanwen JIANG ; Hu QIN ; Zhaofu LENG ; Aikel-Amu PAZILIYA ; Yongxin WANG
Chinese Journal of Nervous and Mental Diseases 2023;49(10):609-614
Hypertensive intracerebral hemorrhage(HICH)is a disease with a rapid onset,rapid progression,high mortality rate,and long-term impact on the ability to function.Non-contrast agent-based CT(NCCT)is a common method for evaluating and identifying HICH.Recent radiomics in image processing and machine learning(ML)have enabled the extraction of high-dimensional feature information from medical images,which can be used to rapidly and accurately diagnose HICH and predict its course of disease.The paper describes the application of radiomics and ML techniques in HICH diagnosis and treatment,and identifies possible directions for future research.