Integrated assessment of diabetic nephropathy by multi-parametric magnetic resonance imaging with non-enhanced T1 mapping, diffusion tensor imaging, and diffusion kurtosis imaging.

Diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI) can provide quantitative parameters that show promise for evaluation of diabetic kidney disease (DKD). The combination of radiomics with DTI and DKI may hold potential clinical value in detecting DKD. To investigate radiomics models of DKI and DTI for predicting DKD in type 2 diabetes mellitus (T2DM) and evaluate their performance in automated renal parenchyma segmentation. Prospective. One hundred and sixty-three T2DM patients (87 DKD; 63 females; 27-80 years), randomly divided into training cohort (N = 114) and validation cohort (N = 49). 1.5-T, diffusion spectrum imaging (DSI) with 9 different b-values. The images of DSI were processed to generate DKI and DTI parameter maps, including fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD). The Swin UNETR model was trained with 5-fold cross-validation using 100 samples for renal parenchyma segmentation. Subsequently, radiomics features were automatically extracted from each parameter map. The performance of the radiomics models on the validation cohort was evaluated by utilizing the receiver operating characteristic (ROC) curve. Mann-Whitney U test, Chi-squared test, Pearson correlation coefficient, least absolute shrinkage and selection operator (LASSO), dice similarity coefficient (DSC), decision curve analysis (DCA), area under the curve (AUC), and DeLong's test. The threshold for statistical significance was set at P < 0.05. The DKI_MD achieved the best segmentation performance (DSC, 0.925 ± 0.011). A combined radiomics model (DTI_FA, DTI_MD, DKI_FA, DKI_MD, and DKI_RD) showed the best performance (AUC, 0.918; 95% confidence interval [CI]: 0.820-0.991). When the threshold probability was greater than 20%, the combined model provided the greatest net benefit. Among the single parameter maps, the DTI_FA exhibited superior diagnostic performance (AUC, 887; 95% CI: 0.779-0.972). The radiomics signature constructed based on DKI and DTI may be used as an accurate and non-invasive tool to identify T2DM and DKD. 2 TECHNICAL EFFICACY: Stage 2.

Diffusion tensor imaging (DTI) is commonly used to assess the integrity of gray and white matter (WM) structures in progressive supranuclear palsy (PSP). Beyond DTI, nontraditional diffusion techniques such as diffusion kurtosis imaging (DKI) have been shown to characterize brain tissue further. In this work, we aim to determine the utility of DKI in the differential diagnosis of PSP-Richardson syndrome (PSP-RS) and PSP with predominant parkinsonism (PSP-P) from Parkinson's disease (PD) and controls. A multishell diffusion-weighted sequence was acquired at 3Tesla on a Siemens system in 22 patients with PSP-RS, 23 with PSP-P, 19 with PD, and 19 controls. Fractional anisotropy, mean diffusivity, kurtosis fractional anisotropy (KFA), and mean kurtosis (Kmean) were calculated for nine deep gray matter regions and six different WM tracts. DKI identified differences (not found by DTI) between control and PSP groups in the globus pallidum externus, subthalamic region, and putamen, with Kmean in the putamen able to differentiate PSP-RS and PD. DKI WM measurements in the body of the corpus callosum and dentatorubrothalamic tract differentiated PSP-RS from PD, and the corticostriatal tract differentiated PSP-P from PD. KFA in the body of the corpus callosum identified worse microstructural anomalies in PSP-RS compared to PSP-P. DKI metrics correlated with the severity of ocular motor impairment and parkinsonism scores. DKI measurements could differentiate PSP-RS, PSP-P, and PD and, hence, may be a promising imaging tool for studying structural neuropathological changes in PSP.

Brain development is one of the most fascinating subjects in the field of biological sciences. Nonetheless, our scientific community still faces challenges in trying to understand the concepts that define the underlying mechanisms of neural tissue development. After all, it is a very complex subject to grasp and many of the processes that take place during central nervous system maturation are yet to be ascertained. Despite this challenge, we have come to recognize that understanding the natural course of normal brain tissue development on both microscopic and macroscopic scales is the key to deciphering the mechanisms through which these neural networks also heal and regenerate. Realizing this concept, my good friend and colleague, Dr. Sarah Milla, and I decided to take on a human study to investigate brain maturation using non-invasive imaging techniques in the pediatric population at New York University (NYU) School of Medicine (Paydar et al., 2013). Our research subjects included 59 normal infants with an age spectrum ranging from birth to approximately 5 years of age, when the brain is in its most active stage of development. We implemented a Magnetic Resonance Imaging (MRI) diffusion technique called Diffusional Kurtosis Imaging (DKI) to investigate the microstructural changes that occur in both the white matter (WM) and gray matter (GM) in the developing brain. Macrostructural changes that take place within the brain during the course of maturation have been well documented by conventional MRI techniques in both normal and pathologic states. However, these conventional techniques are limited in their ability to quantify developmental changes that occur at the microstructural level. Therefore, in vivo characterization and accurate diagnosis of microstructural abnormalities currently remains challenging. Diffusion imaging, including Diffusion-Weighted Imaging (DWI), has been utilized for evaluation of microstructural changes that are difficult to detect using conventional MRI techniques (Basser and Jones, 2002). In particular, the widely used Diffusion Tensor Imaging (DTI) has been shown to be sensitive to age-related microstructural changes during rodent and human brain development in both physiologic and pathologic states (Takeda et al., 1997; Mukherjee et al., 2002; Deipolyi et al., 2005; Lebel et al., 2008; Cheung et al., 2009; Huang et al., 2009; Treit et al., 2013; Yoshida et al., 2013). As a quantitative measuring tool, DTI has also been implemented in studies investigating the degree of neuronal damage due to both acute and chronic injury (Erjian et al., 2013; Guojie et al., 2014) as well as progressive neurodegeneration (Thomalla et al., 2004; Zhang et al., 2009). Fractional Anisotropy (FA), the DTI metric that is the primary index of diffusional directionality, can be used to evaluate the anisotropic neuroarchitectural orientation of WM fiber tracts. FA has demonstrated its sensitivity to certain processes that contribute to tissue organization and increase in anisotropic complexity, particularly myelination, which predominantly occurs in the WM (Beaulieu, 2002; Mukherjee et al., 2002; Lebel et al., 2008). Accordingly, DTI is an excellent tool for investigating the age-related increase in anisotropy that occurs within the network of WM tracks as a result of myelination. However, DTI is based on a Gaussian approximation of water diffusion, which limits its sensitivity to diffusional and microstructural properties of biological tissues (Veraart et al., 2011). Several years ago, DKI diffusion weighted technique, which exploits diffusional non-Gaussianity, was developed at NYU by our colleagues, Drs. Jens Jensen and Joseph Helpern. This technique takes into account the non-Gaussian diffusional properties of water motion in complex media and is therefore more comprehensive in evaluating brain tissue microstructural complexity (Jensen et al., 2005; Lu et al., 2006; Jensen and Helpern, 2010). The DKI method is basically a clinically feasible extension of the traditional DTI model, maintaining the ability to estimate all of DTI's standard diffusion tensor metrics, although with improved accuracy (Veraart et al., 2011). Moreover, DKI provides an additional parameter that quantifies non-Gaussian diffusion called diffusional kurtosis, K. By using the K parameter, multiple additional kurtosis metrics, most importantly, mean kurtosis (MK), can be generated. In our study, we hypothesized that, owing to its potentially higher sensitivity for detection of age-related microstructural changes, MK may provide additional information about brain maturation when compared to that obtainable with the conventional FA metric in both WM and GM. And our results were quite conclusive. We demonstrated a progressive rise in both FA and MK throughout seven WM regions (splenium and genu of corpus callosum, frontal and parietal WM, anterior and posterior limbs of the internal capsule, and external capsule) that we examined in the first 2 years of life. This finding suggested that both DTI and DKI can reflect the age-related increase in diffusional anisotropy in WM tracts, predominantly as a function of myelination in the first 2 years. However, our data also showed that MK continues to rise beyond the FA plateau at the 2-year mark in all WM regions, showing its ability to resolve the more delayed microstructural changes that occur in the WM beyond 2 years of age. In other words, compared to DTI, DKI offers additional characterization of the isotropic diffusion barriers that continue to develop in the WM even after myelination and axonal packing have already peaked. Our study also supported the hypothesis that DKI is sensitive to age-related microstructural changes that occur in the isotropic GM, for which DTI has previously shown to have limited sensitivity (Mukherjee et al., 2002; Cheung et al., 2009). Our results proved that, unlike FA, MK showed a steady rise in signal in the interrogated GM regions (putamen and thalamus) overtime, accounting for specific isotropic GM changes to which FA is not as sensitive. Therefore, when compared to FA, MK can better resolve the progression of GM organization with respect to age by accounting for other isotropic microstructural barriers that form at the cellular level. Finally, we generated three-dimentional DKI tractography images at various stages of development. These tractography images, as illustrated in Figure 1 courtesy of Paydar and colleagues (2013), qualitatively display DKI's ability to detect the progressive age-related increase in volume and coherent orientation of central WM tracts throughout development.Figure 1: Diffusional Kurtosis Imaging (DKI) tractography at various stages of development, including at birth (A), 6 months (B), 11 months (C), and 2 years 1 month (D) of age.Fiber tracking is displayed for the genu (red), splenium (cyan), anterior limb of internal capsule (yellow), posterior limb of internal capsule (pink), and external capsule (dark green).In summary, DKI is an innovative diffusion MRI technique that can provide a more comprehensive evaluation of age-related changes in the microstructural complexity of both WM and GM when compared to DTI. Indeed, both DTI and DKI can detect the anisotropic WM changes which occur predominantly during the first 2 years of life as a result of myelination. However, DKI is able to identify other isotropic WM changes that occur beyond the first 2 years. It also provides greater characterization of GM maturation. Accordingly, DKI offers sensitive and comprehensive measures for the quantitative evaluation of age-related microstructural changes in both WM and GM. So, our investigation has demonstrated the potential utility of a valuable MRI technique for detection of microstructural changes within neural tissues, particularly in the setting of normal brain development. But what is the relevance of this discovery for neural regeneration research? The answer to this question is clear. Speculatively, the diffusion barriers which may form due to the progressive increase in macromolecular reorganization during neural maturation are probably similar to ones that take shape during the course of neural regeneration. These barriers may partly result from many cytoarchitectural changes that take place at the microstructural level during both neural development and regeneration. For example, these changes may include the overall increase in the complexity of intrinsic cellular processes (e.g., proliferation of cell membranes, organelles, and extracellular matrix), axonal pruning and cell packing, myelination and functional reorganization of myelin, as well as addition of basal dendrites and transition of radial glial cells to astrocytic neuropil (Truwit, 2001; Mukherjee et al., 2002; Huppi and Dubois, 2006; Lu et al., 2006; Cheung et al., 2009; Jensen and Helpern, 2010; Veraart et al., 2011; Provenzale et al., 2012; Yoshida et al., 2013). Therefore, since the increase in tissue complexity that occurs during development may be similar to regeneration, DKI may potentially serve as a valuable measuring tool for detection of cellular processes that alter microstructural complexity of tissues in the setting of neural regeneration. We are optimistic about this great prospect and hope that our neuroscience community will effectively use this non-invasive MRI diffusion technique in both in vivo and in vitro settings for neural regeneration research in the near future.

Analytical cross-sectional study. To study the role of diffusion kurtosis imaging (DKI) in evaluating microstructural changes in patients with cervical spondylosis. Cervical spondylosis is a common progressive degenerative disorder of the spine. Conventional magnetic resonance imaging (MRI) can only detect the changes in the spinal cord once there are visual signal changes; hence, it underestimates the extent of the injury. Newer imaging techniques like Diffusion Tensor and Kurtosis Imaging can evaluate the microstructural changes in cervical spinal cord before the obvious signal changes appear. Conventional MRI, diffusion tensor imaging (DTI), and DKI scans were performed for 90 cervical spondylosis patients on 1.5-T MR Siemens Magnetom aera after obtaining informed consent. Eight patients were excluded due to poor image quality. Fractional anisotropy (FA) colour maps and diffusion kurtosis (DK) maps corresponding to spinal cord cross sections at C2-C3 intervertebral disc level (control) and at the most stenotic levels were obtained. Modified Japanese Orthopaedic Association (mJOA) scoring was used for clinical assessment of the spinal cord function. The changes in DTI and DKI parameters and their correlation with mJOA scores were analysed by SPSS 23 software. In our study, mean FA and mean kurtosis (MK) values at the stenotic level (0.54, 1.02) were significantly lower than values at the non-stenotic segment (0.70, 1.27). The mean diffusivity (MD) value at the stenotic segment (1.25) was significantly higher than in the non-stenotic segment (1.09). We also observed a strong positive correlation between mJOA score and FA and MK values and a negative correlation between mJOA score and MD values, suggesting a correlation of FA, MK, and MD with the clinical severity of the disease. Addition of DTI and DKI sequences helps in early identification of the disease without any additional cost incurred by the patient.

BackgroundThis monocentric, cross-sectional study explored the use of diffusion kurtosis imaging (DKI) as a non-invasive means to diagnose and monitor diabetic nephropathy (DN).MethodsPatients with diabetes mellitus (DM, n = 11), mild DN (N = 14), and severe DN (n = 29) were recruited. Eight DKI metrics (MK, MD, Da, Dr, Ka, Kr, FA, FAk) were determined from the imaging results, and their correlations with routine laboratory results were analyzed. The receiver operating characteristic (ROC) curves were plotted, and the diagnostic value of the DKI metrics was analyzed. In addition, renal biopsy was carried out for ten DN patients who had appropriate indications. Their interstitial fibrosis and tubular atrophy (IFTA) score and the fibrosis ratio of cortical area (F%) were analyzed in combination with the DKI metrics.ResultsThe progression of DN, reflected by the estimated glomerular filtration rate (eGFR), was accompanied by rising mean kurtosis (MK) and axial kurtosis (Ka) along with decreasing mean diffusivity (MD), axial diffusivity (Da), and radial diffusivity (Dr). Whereas MK was correlated negatively with hemoglobin (Hb) and eGFR and positively with neutrophil gelatinase-associated lipocalin (NGAL), cystatin C (CysC), and serum creatinine (Scr), MD, Da, and Dr were positively correlated with Hb and eGFR and negatively correlated with CysC and Scr. For the biopsied patients, MK was positively correlated with IFTA, and fractional anisotropy of kurtosis (FAk) was negatively correlated with F% and IFTA. Among the DKI indicators, MK had the highest AUC (0.922, 95% CI: 0.843–1.000).ConclusionThe noninvasive monitoring of DN was feasible with DKI, and MK could indicate the renal function and fibrosis of DN patients. Changes in MK may also serve as a biomarker to assess treatment response (eg, microstructural improvement) after therapeutic interventions (eg, drug therapy for diabetic nephropathy, anti-fibrotic therapy).

To evaluate the use of diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI) for detection of microstructural changes in the trigeminal nerves of trigeminal neuralgia (TN) patients. Forty TN patients and 40 healthy controls were examined using 3T magnetic resonance imaging (MRI) to evaluate DTI and DKI parameters in trigeminal nerves. One-way ANOVA was used to test the differences in age, sex, and DTI and DKI parameters between the TN-affected sides, TN-unaffected sides, and controls. For parameters with inter-group differences, pairwise comparisons were performed. Then, the difference ratios (DRs) of the parameters with statistical differences were calculated and used for the receiver operating characteristic (ROC) analysis to assess their diagnostic performance. In addition, for the DTI and DKI parameter values with differences, we used pure DTI and DKI values to perform the ROC analysis. Compared to the unaffected sides and controls, the FA, MK, and Kr of the affected sides of TN patients were significantly reduced, while ADC was significantly increased (p < 0.05). The diagnostic efficiency of the FA DRs (AUC: 0.974; cutoff value: 0.038; sensitivity: 100%; specificity: 95.0%) was the highest among all DTI and DKI parameters. The DRs of FA and MK more efficiently diagnosed TN than pure FA and MK values. DTI and DKI allowed detection of microstructural changes in TN-affected trigeminal nerves. FA DR was the best independent predictor of microstructural changes in TN. Both DTI and DKI can be used for noninvasive quantitative evaluation of the changes in the microstructure of the cisternal segment of the cranial nerves in clinical practice. • Diffusion tensor imaging (DTI) can be used to evaluate the in vivo integrity of white matter and nerve fiber pathway. • Diffusion kurtosis imaging (DKI) has been shown to be considerable sensitive to microstructural changes. • DTI combined with DKI can comprehensively evaluate the status of the TN-affected trigeminal nerve.

Dermatomyositis (DM) and muscular dystrophy are clinically difficult to differentiate. To confirm the feasibility and assess the accuracy of conventional magnetic resonance imaging (MRI), T2 map, diffusion tensor imaging (DTI), and diffusion kurtosis imaging (DKI) in the differentiation of DM from muscular dystrophy. Forty-two patients with DM proven by diagnostic criteria were enrolled in the study along with 23 patients with muscular dystrophy. Conventional MR, T2 map, DTI, and DKI images were obtained in the thigh musculature for all patients. Intramuscular T2 value, apparent diffusion coefficient (ADC), fractional anisotropy (FA), mean diffusivity (MD), and mean kurtosis (MK) values were compared between the patients with DM and muscular dystrophy. Student's t-tests and receiver operating characteristic (ROC) curve analyses were performed for all parameters. P values < 0.05 were considered statistically significant. The intramuscular T2, ADC, FA, MD, and MK values within muscles were statistically significantly different between the DM and muscular dystrophy groups (P<0.01). The MK value was statistically significantly different between the groups in comparison with T2 and FA value. As a supplement to conventional MRI, the parameters of MD and MK differentiated DM and muscular dystrophy may be valuable. The optimal cut-off value of ADC and MD values (with respective AUC, sensitivity, and specificity) between DM and muscular dystrophy were 1.698 ×10-3mm2/s (0.723, 54.1%, and 78.1%) and 1.80 ×10-3mm2/s (61.9% and 70.2%), respectively. Thigh muscle ADC and MD parameters may be useful in differentiating patients with DM from those with muscular dystrophy.

To compare the performance of 3.0T magnetic resonance diffusion kurtosis imaging (DKI) and diffusion tensor imaging (DTI) in evaluation of the degree of tubulointerstitial damage and renal function in Immunoglobulin A Nephropathy (IgAN) patients. Both DKI and DTI were performed in 40 IgAN patients and 17 healthy volunteers. IgAN patients were divided into two groups according to tubulointerstitial lesion score: Mild injury group, n = 24; Moderate-severe injury group, n = 16. DKI characteristic parameters [mean kurtosis (MK), axial kurtosis (Ka), radial kurtosis (Kr)] and DTI parameters [fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (Da), radial diffusivity (Dr)] of renal cortex and medulla were measured and compared among different groups. Correlations between DKI, DTI parameters and clinicopathological characteristics were assessed. Diagnostic performance of DKI and DTI to evaluate tubulointerstitial damage of IgAN was compared. Cortical MK, Kr, Da and parenchymal Ka significantly differed among three groups (P < 0.05). Cortical MK, Kr, Ka were negatively correlated with estimated glomerular filtration rate (eGFR) (MK: r = - 0.613; Kr: r = - 0.539; Ka: r = - 0.664) and positively correlated with tubulointerstitial lesion score (MK: r = 0.655; Kr: r = 0.577; Ka: r = 0.661) (all P < 0.001). Lower correlation coefficient was found among cortical FA, MD, Dr and eGFR, tubulointerstitial lesion score (all|r|< 0.350). The AUCs of DKI and DTI parameters for differentiating Mild injury group from control group were (cortical MK 0.822, cortical Ka 0.816; cortical FA 0.515, cortical MD 0.714) and for differentiating Mild injury group from Moderate-severe injury group were (cortical MK 0.813, cortical Ka 0.831; medulla FA 0.784, medulla MD 0.586). Compared with DTI, DKI was more sensitive and accurate to probe the renal function and the tubulointerstitial damage of IgAN, especially the mild tubulointerstitial damage.

BackgroundMany biomarkers show high diagnostic values for diabetic kidney disease (DKD), but fewer studies focus on the predictive assessment of DKD progression by blood and urinary biomarkers.AimThis study aims to find powerful risk predictors and identifying biomarkers in blood and urine for DKD progression.MethodsA total of 117 patients with type 2 DKD including early and advanced stages and their laboratory parameters were statistically assessed. A receiver operating characteristic (ROC) curve analysis was performed to evaluate the significance of discriminating between early and advanced DKD, and the predictive power for advanced DKD was analyzed by regression analysis and trisector grouping.ResultsN‐acetyl‐β‐d‐glucosaminidase‐creatine (NAG/CR) level in advanced DKD was statistically higher than that in early DKD (p < 0.05), and there was a higher incidence of advanced DKD (72% vs. 56%) and high odds ratio (OR: 3.917, 95% CI: 1.579–10.011) of NAG/CR with ≥2.79 U/mmol compared with <2.79 U/mmol (p < 0.05). NAG/CR ratio also showed a higher area under the ROC curve of 0.727 (95% CI: 0.616–0.828, p = 0.010) with a high sensitivity (0.75) and a moderate specificity (0.66) when 1.93 U/mmol was set as the optimal cutoff value. The adjusted‐multivariable analysis revealed that NAG/CR had an OR of 1.021 (95% CI: 1.024–1.038) and 2.223 (95% CI: 1.231–4.463) based on a continuous and categorical variable, respectively, for risk of advanced DKD. Moreover, the prevalence of advanced DKD exhibited an increasing tendency by an increment of the trisector of NAG/CR.ConclusionsThis study suggests that NAG/CR ratio is an independent predictor for advanced DKD, and it also can be used as a powerful identifying marker between early and advanced DKD.

Differentiation of high-grade-astrocytomas from solitary-brain-metastases: Comparing diffusion kurtosis imaging and diffusion tensor imaging

In this work, we aim to assess the significance of diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI) parameters in grading gliomas. Retrospective studies were performed on 53 subjects with gliomas belonging to WHO grade II (n=19), grade III (n=20) and grade IV (n=14). Expert marked regions of interest (ROIs) covering the tumour on T2-weighted images. Statistical texture measures such as entropy and busyness calculated over ROIs on diffusion parametric maps were used to assess the tumour heterogeneity. Additionally, we propose a volume heterogeneity index derived from cross correlation (CC) analysis as a tool for grading gliomas. The texture measures were compared between grades by performing the Mann-Whitney test followed by receiver operating characteristic (ROC) analysis for evaluating diagnostic accuracy. Entropy, busyness and volume heterogeneity index for all diffusion parameters except fractional anisotropy and anisotropy of kurtosis showed significant differences between grades. The Mann-Whitney test on mean diffusivity (MD), among DTI parameters, resulted in the highest discriminability with values of P=0.029 (0.0421) for grade II vs. III and P=0.0312 (0.0415) for III vs. IV for entropy (busyness). In DKI, mean kurtosis (MK) showed the highest discriminability, P=0.018 (0.038) for grade II vs. III and P=0.022 (0.04) for III vs. IV for entropy (busyness). Results of CC analysis illustrate the existence of homogeneity in volume (uniformity across slices) for lower grades, as compared to higher grades. Hypothesis testing performed on volume heterogeneity index showed P values of 0.0002 (0.0001) and 0.0003 (0.0003) between grades II vs. III and III vs. IV, respectively, for MD (MK). In summary, the studies demonstrated great potential towards automating grading gliomas by employing tumour heterogeneity measures on DTI and DKI parameters.

Comparing the value of DKI and DTI in detecting isocitrate dehydrogenase genotype of astrocytomas

Performances of diffusion kurtosis imaging and diffusion tensor imaging in detecting white matter abnormality in schizophrenia

Conditional generative diffusion deep learning for accelerated diffusion tensor and kurtosis imaging.

In this preliminary study, we aimed to investigate the abnormalities of water diffusion in children with temporal lobe epilepsy (TLE). Eight children with unilateral TLE (according to electroencephalography, EEG) and eight age- and sex-matched controls were recruited. Diffusion tensor imaging (DTI)/diffusional kurtosis imaging (DKI) acquisitions were performed. Radial diffusivity (λ(⊥)), axial diffusivity (λ(∥)), mean diffusivity (MD) and fractional anisotropy (FA) maps were calculated for both DTI and DKI, and radial kurtosis (K(⊥)), axial kurtosis (K(∥)) and mean kurtosis (MK) maps were calculated for DKI only. Mann-Whitney test showed that, for white matter in the temporal lobe, DKI-derived λ(∥) , MD and K(∥) were significantly different in bilateral temporal lobes and EEG-abnormal and EEG-normal sides of the temporal lobe between patients and controls, whereas DTI showed no abnormalities. For gray matter, DKI detected significantly higher MD and MK in the same three comparisons, whereas DTI detected abnormalities only in the comparison between bilateral temporal lobes and between EEG-normal sides in cases and left-right matched sides in controls. No significant difference was observed between EEG-abnormal and EEG-normal sides in cases. These preliminary results indicate that DKI is more sensitive than DTI for the detection of diffusion abnormalities in the temporal lobes of children with TLE, even when EEG signals are normal. These findings pave the way for the application of DKI for in-depth studies on TLE in children.