Application of MR Imaging Characteristics in the Differentiation of Renal Changes Between Patients with Stage III Type 2 Diabetic Kidney Disease and Healthy People

Figures & data

Figure 1 Basic flow chart of the experiment.

Figure 2 Respective MR images of the right kidney with stage III type 2 DKD and normal persons (NP). The ROI were curved over the renal parenchyma of the right kidney (red curve). In this study, the images analyzed were the FS-T2WI and ADC maps. (Female patient with DKD, 54 years old, stage III diabetes nephropathy, creatinine value was about 135.7, BMI index was 28.41).

Table 1 Patient Characteristics in the Training and Test Set

Variable	Training Set (n=50)				Test Set (n=38)
	DKD(n=30)	NP(n=20)	t/x^2	P value	DKD(n=25)	NP(n=13)	t/x^2	p value
Age (years) Mean ± SD	52.31±2.02	49.36±1.13	3.235	0.075	52.26±2.27	54.79±3.01	6.714	0.097
Sex			4.433	0.035			4.027	0.045
Male	21	8			18	5
Female	9	12			7	8
Urea (mmol/L)	133.36±22.41	71.85±19.05	13.173	<0.001	140±0.235	73.36±1.135	14.502	<0.001
Creatinine (μmmol/L)	9.81±1.22	4.36±0.83	19.216	<0.001	8.47±1.105	4.64±1.305	18.817	<0.001
Urinary albumin/creatinine ratio (mg/g)	113.63±4.99	0.23±0.12	38.217	<0.001	116.33±7.31	0.21±0.14	43.621	<0.001
Glomerular filtration rate (mL/min)	47.66±2.73	96.34±3.18	26.748	<0.001	46.89±1.71	95.97±3.21	21.874	<0.001
BMI(kg/m^2)			2.257	0.133			2.923	0.087
Normal	13	13			10	9
Fat	17	7			15	4

Abbreviations: DKD, Diabetic kidney disease; NP, Normal population; BMI, Body Mass Index.

Table 2 The Imaging Features Based on FS-T2WI, ADC and United Model

Model	Imaging Features	Coef.
ADC model	lbp-3D-m2_glszm_SmallAreaHighGrayLevelEmphasis	0.97
	Wavelet-LHL-firstorder_90Percentile	−0.69
	lbp-2D_glrlm_LongRunLowGrayLevelEmphasis	0.72
	Original_firstorder_Skewness	1.32
FS-T2WI model	Exponential_glszm_GrayLevelNonUniformity	−1.31
	bp-3D-m1_glszm_LowGrayLevelZoneEmphasis	1.38
	Wavelet-HLH_glrlm_ShortRunLowGrayLevelEmphasis	−0.71
	Wavelet-LLL_firstorder_Median	0.64
	Wavelet-LHL_glcm_ClusterShade	0.31
	Wavelet-HHL_glrlm_ LongRunLowGrayLevelEmphasis	−1.51
	Wavelet-LLH_glrlm_LowGrayLevelZoneEmphasis	0.58
	Wavelet-HLL_ngtdm_Buyness	0.63
United model	ADC-lbp-3D-m2_glszm_SmallAreaHighGrayLevelEmphasis	0.82
	ADC-wavelet-LHL-firstorder_90Percentile	−0.42
	ADC-lbp-2D_glrlm_LongRunLowGrayLevelEmphasis	0.51
	ADC-original_firstorder_Skewness	0.97
	FS-T2WI-wavelet-HLL_ngtdm_Buyness	0.41
	FS-T2WI exponential_glszm_GrayLevelNonUniformity	−0.85
	FS-T2WI bp-3D-m1_glszm_LowGrayLevelZoneEmphasis	1.22
	FS-T2WI wavelet-HLH_glrlm_ShortRunLowGrayLevelEmphasis	0.08
	FS-T2WI wavelet-LLL_firstorder_Median	0.42
	FS-T2WI wavelet-LHL_glcm_ClusterShade	0.23
	FS-T2WI wavelet-HHL_glrlm_ LongRunLowGrayLevelEmphasis	−1.43

Abbreviations: Coef, coefficient; FS-T2WI, fat saturation-T2-weighted imaging; ADC, apparent diffusion coefficient.

Figure 3 The ICC of ADC.

Figure 4 The ICC of FS-T2WI.

Table 3 Performance of Each Model in Training Set and Test Set

	Model	AUC	ACC (%)	SEN (%)	SPE (%)	PPV (%)	NPV (%)
Training set	FS-T2WI	0.969	92.0	96.4	86.4	90.0	95.0
	ADC	0.911	90.0	89.5	93.3	85.0	93.3
	FS-T2WI+ADC	0.982	94.0	93.5	94.7	96.7	90.0
Test set	FS-T2WI	0.910	84.2	88.9	81.8	92.0	69.2
	ADC	0.891	84.2	88.0	76.9	88.0	76.9
	FS-T2WI+ADC	0.935	89.5	88.9	90.9	96.0	76.9

Abbreviations: FS-T2WI, fat saturation-T2-weighted imaging; ADC, apparent diffusion coefficient; ACC, accuracy; AUC, areas under curve; SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive value.

Figure 5 ROC curve analysis of the training set in between the stage III type 2 DKD and normal subjects. The solid lines in different colors indicate that the ROC curve for each model correspond to a different AUC.

Figure 6 ROC curve analysis of the test set between the stage III type 2 DKD and normal subjects. The solid lines in different colors indicate that the ROC curve for each model correspond to a different AUC.