ORIGINAL RESEARCH article

Front. Med., 17 July 2026

Sec. Experimental, Translational and Evidence-Based Medicine

Volume 13 - 2026 | https://doi.org/10.3389/fmed.2026.1824498

Microstructural alterations in brain tissue of ME/CFS and long COVID using diffusion tensor imaging and diffusion kurtosis imaging

  • National Center for Neuroimmunology and Emerging Diseases, Griffith University, Gold Coast, QLD, Australia

Abstract

Introduction:

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) and long COVID have overlapping symptoms, such as profound fatigue, cognitive impairment, post-exertional malaise, pain and sleep disturbances that are debilitating and reduce quality of life. While Magnetic Resonance Imaging (MRI) techniques, specifically Diffusion Tensor Imaging (DTI) and Diffusion Kurtosis Imaging (DKI), have been used to investigate brain tissue microstructure in ME/CFS or long COVID, no study has yet combined these modalities to directly compare tissue microstructural differences between people living with ME/CFS and long COVID.

Methods:

We recruited 37 ME/CFS participants (Age: 43.56 ± 12.5), 19 long COVID participants (Age: 47.92 ± 13.3), and 27 healthy controls (Age: 37.9 ± 10.4). Data were acquired using a 3 Tesla (3T) Prisma MRI scanner. DTI and DKI metrics were determined using MRtrix v3.0.7 and Designer V2.0 software, respectively. Voxel-based statistical analysis of cohort differences was performed using the Statistical Parametric Mapping (SPM12) toolbox in MATLAB. Correlation analysis was performed between DTI, DKI metrics and clinical measures such as duration of illness, fatigue severity, SF36 domains and WHODAS domains.

Results:

Compared with healthy controls, individuals with ME/CFS showed microstructural alterations in the cingulum, supplementary motor areas, and parts of the corpus callosum (all p < 0.05). Long COVID participants demonstrated microstructural alterations in regions including the fusiform and precentral gyrus and in major white matter tracks (all p < 0.05). Direct comparisons between ME/CFS and long COVID revealed difference in the left corona radiata (p = 0.001).

Conclusion:

This study identifies distinct tissue microstructural alterations in ME/CFS and long COVID and offers a vital insight into the neuropathological basis of shared symptoms in both conditions.

1 Introduction

People with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) and long COVID experience a range of shared symptoms, including severe fatigue, cognitive difficulties, post-exertional malaise, pain, and disrupted sleep, that are highly disabling and cause a significant decrease in quality of life (1, 2). Currently, there is no universal diagnostic test for ME/CFS or long COVID (3). The diagnosis of ME/CFS is based on standardized case definitions such as the Canadian Consensus Criteria (CCC) (4) and International Consensus Criteria (ICC) (5). Similarly, the diagnosis of long COVID uses the World Health Organization (WHO) case definition (6), a detailed medical history, and the patient's physical examination to rule out other conditions (7, 8).

Under the WHO International Classification of Diseases, ME/CFS is classified as a disease of the nervous system (9). Long COVID also presents with various neurological symptoms (10). Cognitive impairment is a prominent central nervous system (CNS) related symptom in both conditions, often presenting as difficulties with memory, concentration, or “brain fog” (11, 12). Other CNS-related symptoms such as fatigue, sleep disturbances, headache and pain are also present in these conditions (11). Therefore, neuroimaging techniques such as MRI have been used to study brain dysfunction in ME/CFS and long COVID (13, 38). Diffusion weighted MRI (DWI), which measures the diffusion of water molecules, is a non-invasive technique to study brain microstructure, such as axons, glial cells and myelin (14).

Diffusion Tensor Imaging (DTI) is a specific type of modeling of the DWI datasets, which quantifies tissue water diffusion and enables the indirect measurement of the degree of diffusion anisotropy and its structural orientation (15). DTI is based on the Brownian motion of water molecule diffusion and is anisotropic along the axonal pathway, as perpendicular diffusion to the fiber is hindered by myelin sheaths, axonal cellular membranes and the neurofibrils (16, 17). This helps to estimate the axonal organization in the brain and provides image contrast based on structural orientation thereby indirectly assessing neuroanatomy structure at a microscopic level (17). The diffusion tensor has three perpendicular eigenvectors and three positive eigenvalues (18). Fractional Anisotropy (FA), the most widely used scalar, is basically a normalized variance of the eigenvalues (18). Mean Diffusion (MD) is the mean diffusion of each direction, which is the average of three eigenvalues (19). Axial Diffusion (AD) describes the diffusion rate along the primary axis, which is the largest eigenvalue of diffusion, while Radial Diffusion (RD) reflects the average diffusivity along the two minor axes (19).

Diffusion Kurtosis Imaging (DKI), an extension of DTI, enables the estimation of the diffusion kurtosis tensor to characterize additional non-Gaussian diffusion properties within complex biological tissues by estimating the excess kurtosis of the displacement distribution (20, 21). DKI has the ability to capture non-Gaussian diffusion and is more sensitive to microstructural complexity, although it requires a multishell acquisition protocol and has comparatively lower spatial resolution (22, 23). Similar to DTI, DKI possesses corresponding scalars for measuring diffusion kurtosis, including kurtosis fractional anisotropy (KFA), mean kurtosis (MK), axial kurtosis (AK) and radial kurtosis (RK) (24). KFA summarizes the directional variation in the degree of non-Gaussian diffusion (25). Similarly, MK quantifies the magnitude of diffusion kurtosis along the three spatial axes (26). Likewise, AK quantifies the magnitude of diffusion kurtosis along the main axis, and RK quantifies the magnitude of diffusion kurtosis perpendicular to the main axis (26).

DTI and DKI parameters have been widely used to investigate microstructural changes in the brain across various neurological conditions (27). People living with ME/CFS and long COVID also experience overlapping neurological symptoms, including brain fog, fatigue, cognitive impairment, and sleep disturbances (11, 12). Given these shared clinical features, examining both the similarities and differences in brain tissue microstructural alterations is important to improve our understanding whether these two conditions are similar or distinct. Thapaliya et al. (28) has demonstrated the microstructural changes in the brainstem region using DTI where AD and MD were significantly decreased whereas RD was increased in people with ME/CFS compared to healthy controls (28). Yu et al. reported ME/CFS individuals with post-infectious onset showed increased AD, whereas those with gradual onset exhibited reduced AD (29). Zeineh et al. (30) reported increased FA in the right anterior arcuate, while Kimura et al. (31) observed significantly lower FA in the genu of the corpus callosum and in the anterior limb of the internal capsule of ME/CFS compared to healthy controls. Furthermore, in people living with long COVID, Liang et al. (32) reported increases in FA in several white matter regions and in MD in the left amygdala. Studies have also reported significantly decreased FA, MD, RD and AD in long COVID compared to healthy controls (33, 34). Additionally, Thapaliya et al. demonstrated significantly higher FA value in the right superior longitudinal fasciculus in long COVID compared to COVID-recovered healthy controls and significantly lower AD and MD in the left caudate in COVID-recovered healthy control compared to non-COVID-healthy controls (6). Using the DKI method, Kimura et al. (31) found a significant decreases in the DKI metric MK in the right frontal area, anterior cingulate cortex, superior longitudinal fasciculus and left parietal areas. Yuan et al. (35) also found reduced MK and RK in the right inferior fronto-occipital fasciculus in recovered COVID-19 patients. Overall, inconsistencies in ME/CFS DTI findings were driven by varying diagnostic criteria, heterogeneous symptoms, sample size, data processing pipelines and the scarcity of DKI studies in ME/CFS and long COVID, limits the robustness of the current conclusions.

Given the equivocal nature of current findings, further investigation of tissue microstructural alterations in ME/CFS and long COVID is critical. Furthermore, no study has yet assessed DTI and DKI methods to directly compare tissue microstructural alterations between ME/CFS and long COVID. Therefore, this study aims to compare tissue microstructural alterations between ME/CFS, long COVID, and healthy controls using DTI and DKI metrics. Additionally, this study will explore the correlation between DTI and DKI metrics and cognitive measures in both ME/CFS and long COVID.

2 Materials and methods

2.1 Participant recruitment

The study was approved by the Griffith University Human Research Ethics Committee (Ref: 2022/666) and conducted in accordance with the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants prior to inclusion in the study.

This cross-sectional study was carried out in the National Center for Neuroimmunology and Emerging Diseases (NCNED), Gold Coast, Australia. Participants were recruited via the NCNED research registry database, as previously described in Thapaliya et al. (34). 37 ME/CFS participants were recruited after meeting the Canadian Consensus Criteria (CCC) (4) or International Consensus Criteria (ICC) (5) and had received a formal diagnosis of ME/CFS by a physician. 19 long COVID participants were recruited after meeting the WHO working case definition (6). 27 healthy controls were recruited if they reported no chronic health conditions, underlying illness and had no prior COVID-19 infection. Medical histories for diseased and control participants were reviewed to identify comorbid symptoms or exclusionary diagnoses, including mental illness, malignancies, autoimmune, neurological, or cardiovascular diseases. Female participants were excluded if they were pregnant and/or breastfeeding. For ME/CFS and long COVID, exclusionary mental illness applies only if they were present before the illness was diagnosed. Furthermore, pregnancy and breastfeeding were excluded due to safety issue of fetus during MRI scans. Additionally, pregnancy and breastfeeding are associated with substantial hormonal and metabolic changes that can influence brain microstructure, introducing potential confounding effects in diffusion-based measurements (36, 37).

2.2 Clinical measures

The National Center for Neuroimmunology and Emerging Diseases (NCNED) in conjunction with the Centers for Disease Control and Prevention (CDC), has developed the research registry questionnaire and distributed it online through Lime Survey and Redcap, as reported in a previous publication (38). The CDC 2005 Symptom Inventory is a self-report questionnaire designed by the CDC regarding physical symptoms that a ME/CFS patient may have experienced during the past month (39). One of the questionnaires captured fatigue severity from the ICC (5) and CCC (4) criteria. The inventory also asks the participant to rate his/her experience in the past month regarding unusual fatigue (39). Fatigue severity was self-reported using the CDC's Symptom Inventory, which scores symptoms on a five-level Likert scale: 1 = very mild, 2 = mild, 3 = moderate, 4 = severe, and 5 = very severe (39). Fatigue was considered present if it was reported as being at least very mild within the month prior to completing the questionnaire (40). Quality of life was measured using the 36-item Short-Form Health Survey, version 2 (SF-36v2) (98). Domains of SF-36v2 includes general health, physical functioning, role physical, role emotional, pain, mental health, vitality and social functioning. Scores for each domain fall between 0 and 100%, reflecting the individual's overall level of quality of life, with higher percentages indicating better perceived wellbeing (40). Functional capacity was evaluated using the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0), a standardized tool for assessing everyday functioning (97). Domains of WHODAS 2.0 includes cognitive, mobility, self-care, interpersonal-relation, life activity and society. Each domain produces a score from 0 to 100%, reflecting the extent of functional difficulty or disability in that area. Higher percentages indicate greater impairment, with 0% representing no limitation and 100% indicating very severe difficulty (40). Duration of illness for ME/CFS and long COVID following symptom onset was determined based on participant self-reports and formal diagnoses made by medical professionals. Table 1 shows demographic and clinical characteristics of ME/CFS, long COVID, and healthy controls.

Table 1

VariablesME/CFS (n = 37)Long COVID (n = 19)Healthy controls (n = 27)P-ValueMissing data
ME/CFSLong COVID
Age (years)43.56 ± 12.547.92 ± 13.337.9 ± 10.40.20a 0.021b 0.618cN/AN/A
Sex (F/M)30/714/519/8N/AN/AN/A
Duration of illness (years)14.53 ± 12.30.78 ± 0.69N/A<0.00161
Fatigue severity3.73 ± 0.823.39 ± 0.5N/A0.17671
SF36 general health31.6 ± 13.755.9 ± 22.580.9 ± 10.2<0.001a <0.001b 0.002c106
SF36 physical functioning37.2 ± 27.875.7 ± 7.390.9 ± 24.9<0.001a 0.028b 0.017c111
SF36 role physical13.9 ± 17.843.7 ± 20.497.6 ± 6.4<0.001a <0.001b 0.008c115
SF36 role emotional57.1 ± 38.571.4 ± 24.995.8 ± 7.4<0.001a 0.002b 0.143c113
SF36 pain44.0 ± 24.961.0 ± 29.288.1 ± 15.8<0.001a <0.001b 0.259c94
SF36 mental health58.4 ± 20.368.5 ± 17.985 ± 11.6<0.001a 0.002b 0.09c103
SF36 vitality11.6 ± 10.934.8 ± 23.676.5 ± 12.8<0.001a <0.001b 0.005c136
SF36 social functioning22.1 ± 19.255.3 ± 20.294.5 ± 9.0<0.001a <0.001b 0.007c116
WHODAS cognitive impairment49.6 ± 14.036.2 ± 22.73.0 ± 5.5<0.001a <0.001b 0.004c93
WHODAS mobility48.5 ± 21.031.1 ± 20.50.33 ± 1.2<0.001a <0.001b 0.158c96
WHODAS selfcare28.1 ± 25.513.2 ± 19.30.00 ± 0.00<0.001a 0.01b 0.482c82
WHODAS interpersonal40.8 ± 26.735.3 ± 32.61.6 ± 3.7<0.001a 0.005b 0.079c97
WHODAS life activity69.5 ± 21.042.3 ± 27.91.6 ± 4.3<0.001a <0.001b 0.007c94
WHODAS society63.8 ± 15.842.5 ± 18.61.6 ± 3.5<0.001a <0.001b <0.001c93

Demographic and clinical characteristics.

Values are given as mean ± standard deviation for ME/CFS, long COVID and Healthy controls. Multiple comparison correction was performed using the Bonferroni method. For the three P-values superscript ‘a' represents ME/CFS vs healthy controls, ‘b' represents long COVID vs healthy controls and ‘c' represents ME/CFS vs. long COVID. M: Male, F: Female, n number of subjects, ME/CFS: Myalgic Encephalomyelitis/Chronic Fatigue Syndrome.

2.3 Imaging parameters

The diffusion data were acquired using a 3Tesla Prisma (Siemens Scanner) MRI scanner with a 64-channel head-neck coil. DTI data were acquired using 2-shell acquisition protocols: 30 directions at b = 1,000 s/mm2 and 66 directions at b = 2,500 s/mm2, along with nine b = 0 scans. Other settings were repetition time/echo time = 4,100/75 ms, field of view (FOV) = 244 × 244, flip angle = 90°, matrix = 122 × 122, voxel dimension of 2.0 mm3 and 66 slices, phase encoding direction: anterior to posterior (AP).

2.4 Data processing

In this study, for DTI processing, we used the b-value of 1,000 s/mm2 because the increased b-value leads to a decrease in signal-to-noise ratio (SNR) (41) that might affect the estimation of DTI metrics. For DKI processing, we used both the shells (b = 1,000 s/mm2 and b = 2,500 s/mm2 in addition to b = 0). As with all higher-order diffusion models, DKI requires multishell acquisition in order to increase the biological specificity of diffusion imaging (42, 43). The b-value optimisation for the DTI study is based on Soares et al. (15) and for the DKI study on Yan et al. (44).

Preprocessing of diffusion data was carried out using “topup” (45) and “eddy” (46) commands in FSL (FMRIB's Diffusion Toolbox). Susceptibility-induced distortions were first estimated using topup, based on pairs of reverse phase-encoded b = 0 images. The resulting susceptibility field was then incorporated into eddy, which performed simultaneous correction for eddy current–induced distortions and subject motion, while applying the previously estimated field to correct for susceptibility effects (47). Following these corrections, a brain mask was generated from the distortion-corrected diffusion data using the bet2 command in FSL, with a fractional intensity threshold of 0.25 (48, 49). DTI metrics, FA, MD, AD and RD were obtained using dwi2tensor and tensor2metric commands in Mrtrix3 version 3.0.7 software (50). DKI metrics KFA, MK, AK and RK were obtained using the “tmi” command in Designer-v2 software (51). Each participant's FA map was then non-linearly registered to Montreal Neurological Institute (MNI) standard space using the tract-based spatial statistics (TBSS) toolkit of FSL (52). The FSL_HCP1065Fa 1 × 1 × 1 mm3 standard space image was used as a target. The rest of the DTI metrics (MD, AD, and RD) and DKI metrics (KFA, MK, AK, and RK) were normalized and warped by applying the deformations obtained from the FA co-registration.

2.5 Statistics

Voxel-based statistical analysis of FA, MD, AD, RD, KFA, MK, AK, and RK comparing ME/CFS, long COVID, and healthy controls was performed using the Statistical Parametric Mapping (SPM12) toolbox (53). Statistical inference was measured with the false discovery rate corrected cluster p-value (cluster p-FDR <0.05). Significant regions were overlaid on the T1-weighted image (mni_icbm152_t1_tal_nlin_sym_09a). Cluster locations were identified with the xjview toolbox (xjView | A viewing program for SPM) (54) and FSL software (55).

All data were assessed for normality using Shapiro-Wilk test. Only age was found to be normally distributed while all other clinical measures showed non-normal distribution (all p < 0.05). Age differences between ME/CFS, long COVID and healthy controls were analyzed using one-way ANOVA. Differences in clinical measures across three groups were examined using a generalized linear model. Multiple comparisons correction was adjusted using the Bonferroni correction method (56).

Correlation analysis between significant DTI and DKI metrics and clinical measures were performed using partial Spearman's rank correlation. All the statistical analyses were conducted using SPSS Statistics version 29.0, and p-values were adjusted for multiple comparisons using false discovery rate (57), with statistical significance defined as pFDR <0.05. Age and gender were included as covariates to control for their potential effects in all the analysis. Illness duration was included as an additional covariate when examining group differences between ME/CFS and long COVID.

DTI comparisons between long COVID participants and healthy controls were excluded to avoid redundancy and reduce the risk of type I error, as these data were already reported in a previous publication (34).

3 Results

3.1 Demographic and clinical characteristics

Group comparisons revealed significant differences across demographic and clinical measures (Table 1). Long COVID participants were older than controls (47.9 ± 13.3 vs. 37.9 ± 10.4 years, p = 0.021), and illness duration was significantly longer in ME/CFS than long COVID (14.5 ± 12.3 vs. 0.8 ± 0.7 years, p < 0.001).

Across SF-36 domains, ME/CFS and long COVID patients scored significantly lower than controls on all subscales (general health, physical functioning, role physical, role emotional, pain, mental health, vitality and social functioning) (all p0.05) and ME/CFS showed significant impairment compared to long COVID on general health, physical functioning, role physical, vitality and social functioning (all p0.05).

Across WHODAS domains, ME/CFS and long COVID patients scored significantly lower than controls on all subscales (cognitive impairment, mobility, selfcare, interpersonal, life activity and society) (all p0.01), and ME/CFS showed significant impairment compared to long COVID on cognitive impairment, life activity and societal participation (all p0.025).

3.2 DTI group comparison

3.2.1 ME/CFS vs. long COVID

No significant differences in the DTI metrics were observed between ME/CFS and long COVID.

3.2.2 ME/CFS vs. healthy controls

Voxel-based analysis of DTI metrics revealed (A) significantly increased FA (p-FDR = 0.024, Peak T = 4.42, Cluster size = 99, X = −6, Y = 16, and Z = 28; see Figure 1A, Table 2) in the left cingulum of ME/CFS compared to healthy controls, and (B) significantly decreased MD (p-FDR = 0.019, Cluster size = 109, Peak T = 4.26), AD (p-FDR = 0.034, Cluster size = 67, Peak T = 4.08) and RD (p-FDR = 0.032, Cluster size = 105, Peak T = 3.94) at the same location (X = −1, Y = 2, and Z = 50, see Figure 1B, Table 2) in the left supplementary motor area (SMA-L) of the medial frontal gyrus in ME/CFS compared to healthy controls.

Figure 1

Table 2

DTI MetricIncreased (↑)/ Decreased (↓)AreaMNI coordinates x y z (mm)Cluster size (voxels)Peak T-valueCluster p-FDR
FALeft cingulum−6 16 28994.420.024
MDSMA-L−1 2 501094.260.019
ADSMA-L−1 2 50674.080.034
RDSMA-L−1 2 501053.940.032

Significant DTI metrics cluster: ME/CFS vs. Healthy controls.

FA, represent Fractional Anisotropy; MD, Mean Diffusivity; AD, Axial Diffusivity; RD, Radial Diffusivity; Cingulum-L, Cingulum-Left; SMA-L, Supplementary motor area of left side, ↑: increased, ↓: decreased, FDR: False discovery rate.

3.3 DKI group comparison

3.3.1 ME/CFS vs. long COVID

AK was significantly increased in the left corona radiata of ME/CFS (p-FDR = 0.001, Cluster size = 107, Peak T = 4.65, X = −22, Y = 27 and Z = 26; seeFigure 2) compared to long COVID.

Figure 2

3.3.2 ME/CFS vs. healthy controls

Voxel-based analysis of DKI metrics revealed significantly increased KFA (p-FDR = 0.021, Peak T = 4.71, Cluster size = 98, X = −7, Y = 16, and Z = 29; see Figure 3A, Table 3) in the left cingulum of ME/CFS compared to healthy controls. Also, AK was significantly decreased in the body (p-FDR <0.001, Peak T = 5.42, Cluster size = 247, X = 1, Y = −3, and Z = 24; see Figure 3B, Table 3) and genu (p-FDR <0.001, Peak T = 5.02, Cluster size = 113, X = 2, Y = 12 and Z = 21; seeFigure 3C, Table 3) of corpus callosum in ME/CFS compared to healthy controls. There were no significant differences in the MK and RK metrics.

Figure 3

Table 3

DKI metricIncreased (↑)/ Decreased (↓)AreaMNI coordinates x y z (mm)Cluster size (voxels)Peak T-valueCluster p-FDR
KFALeft Cingulum−7 16 29984.710.021
MKN/AN/AN/AN/AN/AN/A
AKGenu of CC2 12 211135.020.000
Body of CC1 −3 242475.420.000
RKN/AN/AN/AN/AN/AN/A

Significant DKI metrics cluster: ME/CFS vs. Healthy Controls.

KFA, Kurtosis Fractional Anisotropy; MK, Mean Kurtosis; AK, Axial Kurtosis; RK, Radial Kurtosis; CC, Corpus Callosum; ↑: increased, ↓: decreased, FDR: False discovery rate.

3.3.3 Long COVID vs. healthy controls

In DKI of long COVID compared to the healthy controls, we observed significantly increased MK (p-FDR = 0.024, Peak T = 4.93, Cluster size = 85, X = 27, Y = −78 and Z = −5; see Figure 4A, Table 4) and RK (p-FDR = 0.006, Peak T = 4.66 Cluster size = 87, X = 26, Y = −78 and Z = −5; see Figure 4A, Table 4) in the right fusiform gyrus. RK was increased in the genu of the corpus callosum (p-FDR = 0.001, Peak T = 4.76, Cluster size = 125, X = 10, Y = 32 and Z = 10; see Figure 4B, Table 4), right superior corona radiata (p-FDR = 0.006, Peak T = 4.61, Cluster size = 79, X = 28, Y = −1 and Z = 25; seeFigure 4C, Table 4), left superior corona radiata (p-FDR = 0.012, Peak T = 4.53, Cluster size = 69, X = −15, Y = 25 and Z = 43; see Figure 4D, Table 4) and in right precentral gyrus (p-FDR = 0.006, Peak T = 4.28, Cluster size = 87, X = 48, Y = 2 and Z = 37; see Figure 4E, Table 4) in long COVID compared to healthy controls.

Figure 4

Table 4

DKI metricIncreased (↑)/ Decreased (↓)AreaMNI coordinates x y z (mm)Cluster size (voxels)Peak T-valueCluster p-FDR
KFAN/AN/AN/AN/AN/AN/A
MKRight fusiform gyrus27 −78 −5854.930.024
AKN/AN/AN/AN/AN/AN/A
RKGenu of CC/ forceps minor10 32 101254.760.001
Right superior corona radiata28 −1 25794.610.006
Left superior corona radiata−15 25 43694.540.012
Right precentral gyrus48 2 37664.280.013
Right fusiform gyrus26 −78 −5874.660.006

Significant DKI metrics cluster: long COVID vs. Healthy controls.

KFA, Kurtosis Fractional Anisotropy; MK, Mean Kurtosis; AK, Axial Kurtosis; RK, Radial Kurtosis; CC, Corpus Callosum; ↑: increased, ↓: decreased, FDR: False discovery rate, N/A: Not applicable.

3.4 Correlation analysis with clinical measures

We found significant uncorrected correlations between DTI, DKI metrics and clinical measures in ME/CFS and long COVID. However, we did not find any significant differences between these metrics after adjusting for multiple comparisons. All the correlations analysis between DTI, DKI and clinical measures are provided as a Supplementary Table 110.

4 Discussion

This is the first study to characterize brain tissue microstructural alterations in people living with ME/CFS and long COVID using both DTI and DKI methods. We identified distinct microstructural alterations between ME/CFS, long COVID and healthy controls. DKI offered valuable insights that complemented DTI findings.

4.1 Group Comparison: ME/CFS vs. Long COVID

We found significant differences in the DKI metric between ME/CFS and long COVID. There was a significant increase in AK in the left corona radiata, which is the major projection fiber and includes corticospinal, corticopontine and thalamocortical fibers (58). Yu et al. (59) in their DTI based study also reported higher AD in the major association fiber in ME/CFS. Elevated AK typically reflects greater diffusion complexity or restriction along the axonal axis, which is consistent with increased cellular packing, a more constrained microenvironment, or heightened structural complexity within axons (60). The microstructural alterations observed in the left corona radiata may therefore represent compensatory increases in cellular density or neuroinflammatory processes (61, 62). In addition, the corticospinal, corticobulbar, and thalamocortical pathways, which traverse the corona radiata are essential for voluntary motor control, fine and gross motor coordination, and sensory gating (6365). Consequently, disruptions in the microstructure of the corona radiata could plausibly contribute to the motor difficulties, altered sensory processing, and cognitive impairment observed in these conditions.

Notably, our DTI analysis revealed no significant differences between ME/CFS and long COVID. This underscores the ability of DKI to overcome critical DTI limitations such as the inability to resolve crossing fibers (66) and, higher test-retest variability (67).

4.2 Group comparison: ME/CFS vs. healthy controls

Our DTI study found significantly decreased MD, AD and RD in the left SMA and increased FA in the left cingulum of ME/CFS compared to healthy controls. Our findings align with previous findings where decreased AD and MD and also increased FA were reported in people living with ME/CFS compared to healthy controls (28). Similarly, Wu et al. demonstrated microstructural changes in the left cingulum in ME/CFS (68). The SMA is located in the medial aspect of the superior frontal gyrus, involved in higher-order cognition, including planning, initiation and execution of complex voluntary actions and higher-order processes such as speech and motor sequencing (69). Similarly, the cingulum is a large association fiber pathway and also a part of the limbic system, which connects cingulate gyrus to the frontal, parietal, temporal and occipital cortices, allowing for integration of memory, emotion and executive function (70, 71). The decrease of MD, AD, and RD in the SMA could indicate an alteration in myelination and cellular density of the membrane, probably indicating cellular swelling or cellular proliferation (28, 72, 73). Furthermore, increased in FA in the cingulum could reflect increased myelination or cellular edema (74). Therefore, alterations in these metrics in the SMA and cingulum could cause cognitive deficits in ME/CFS.

The KFA parameter was also increased in the left cingulum of ME/CFS compared to healthy controls. This could be because KFA is a natural extension of the FA concept in the kurtosis tensor, which is mathematically analogous (75).

The AK parameter was significantly decreased in the genu and body of the corpus callosum in ME/CFS compared to healthy controls. The corpus callosum is the largest commissural white matter tract, connecting the two cerebral hemispheres (7678). The genu of the corpus callosum primarily links the left and right prefrontal cortices involved in higher cognitive functions and decision making, while the body connects the frontal and parietal lobes, coordinating motor and sensory function (7678). The decreased AK in the genu and body of the corpus callosum probably means there is axonal injury or loss of integrity causing alteration in axonal pathways (79). Therefore, ME/CFS presentation with cognitive dysfunction, motor slowing and hypersensitivity to noise and light (80, 81), is consistent with involvement of the corpus callosum.

4.3 Group comparison: long COVID vs. healthy controls

We found a significant increase in the RK values in the forceps minor and right and left superior corona radiata in long COVID compared to healthy controls. Forceps minor is an interhemispheric fiber tract that forms the genu of the corpus callosum and links both orbitofrontal cortices (82) while the right and left superior corona radiata are projection fibers that connect cortex to brainstem and thalamus with both afferent and efferent fibers (83). Involvement of commissural and projection fibers has been described in the previous post COVID-19 study (84). Two studies have implicated forceps minor in relation to cognitive dysfunction in the post COVID condition (85, 86). Additionally, a study also showed a relationship between the corona radiata, fatigue and cognitive complaints (87). Increased RK in commissural and projection fibers indicates enhanced restriction of water diffusion perpendicular to the principal fiber orientation, suggesting increased microstructural complexity in the radial direction. This pattern may arise from myelin compaction, reduced extracellular space, or glial-related alterations rather than axonal loss (8890). As forceps minor and corona radiata are tracts central to cognition, attention, fatigue and emotional regulation (6365, 86), impairment in these regions may contribute to the cognitive decline and fatigue in long COVID.

We also found increased RK in the right precentral gyrus and in the right fusiform gyrus. The precentral gyrus is the anatomical location of the primary motor cortex, which is responsible for controlling voluntary motor movement of the contralateral side of the body (91). Diez-Cirarda et al. (87) showed a relationship between reduction in the gray matter volume in the precentral gyrus and cognitive decline which is one of the hallmark symptoms of long COVID. Troll et al. (92) using fMRI found increased functional connectivity between the right caudate nucleus and both left and right precentral gyrus in long COVID compared to healthy controls. Additionally, increased T1/T2 signal intensity was reported in the precentral gyrus in long COVID compared to healthy controls, which is related to the higher myelin signal (34) consistent with our RK findings (8890).

We observed significantly increased MK in the right fusiform gyrus, a region critical for visual processing and face recognition. Prior studies showed atrophy in the left fusiform gyrus (93), and right-sided involvement was associated with higher levels of COVID-related post-traumatic stress and social anxiety (94) and a prosopagnosia in long COVID (95). The elevated MK likely reflects greater tissue complexity, increased cellularity, or more restricted water diffusion, potentially due to neuroinflammation or reactive gliosis (62, 96). As no prior long COVID studies have utilized DKI, these findings represent a novel discovery of tissue microstructural alterations in people living with ME/CFS and long COVID.

5 Limitations

Despite novel findings in ME/CFS and long COVID, this study has some limitations. We did not stratify ME/CFS and healthy controls based on prior COVID infection. Additionally, as this is a cross-sectional study, future longitudinal research will be important to clarify whether the observed changes in DTI and DKI measures evolve or remain stable. Additionally, the long COVID group was relatively small (n = 19), which limits statistical power and the confidence of group comparisons. Therefore, these findings should be viewed as preliminary and will require confirmation in larger, well-powered samples. Furthermore, although age and sex were included as covariates, the significant age difference between groups may introduce residual confounding, particularly given the known sensitivity of diffusion metrics to age-related microstructural changes. Also, we were unable to formally compare the sensitivity of DTI and DKI, so differences between the methods may reflect acquisition or processing factors rather than true biological effects. DKI should therefore be viewed as providing complementary information, and dedicated studies are needed to assess comparative sensitivity. Additionally, we also excluded DTI comparisons between the long COVID and control groups due to prior publication (34), which may limit the direct comparisons across all groups.

6 Conclusions

This study identifies distinct tissue microstructural alterations in people living with ME/CFS and long COVID using DTI and DKI. These findings offer vital insights into the neuropathological basis of shared symptoms in both conditions. Additionally, our results suggest that DKI enables a more nuanced characterization of the complex neuropathological processes underlying these conditions. Complementary DTI and DKI insights contribute to a more comprehensive assessment of the associated microstructural changes.

Statements

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by the Griffith University Human Research Ethics Committee (Ref: 2022/666) and conducted in accordance with the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants prior to inclusion in the study.. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

TS: Conceptualization, Software, Data curation, Formal analysis, Writing – original draft, Methodology. SM-G: Supervision, Funding acquisition, Writing – review & editing. LB: Supervision, Writing – review & editing, Funding acquisition. NE-F: Writing – review & editing. TH: Data curation, Writing – review & editing. MI: Writing – review & editing, Data curation. KT: Data curation, Methodology, Writing – review & editing, Conceptualization, Supervision, Funding acquisition.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was funded by ME Research UK (SCIO Charity Number SC036942) with the financial support of The Fred and Joan Davies Bequest and The Stafford Fox Medical Research Foundation (489798). The authors declare that the funders had no role in the study design, data collection, data analysis, decision to publish or preparation of the present manuscript.

Acknowledgments

We are thankful to Ms. Tania Manning and Kay Schwarz for recruiting participants for this study, radiographers at the University of Queensland, and the patients and healthy controls who donated their time and effort to participate in this study.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmed.2026.1824498/full#supplementary-material

References

  • 1.

    DehliaAGuthridgeMA. The persistence of myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) after SARS-CoV-2 infection: a systematic review and meta-analysis. J Infect. (2024) 89:106297. doi: 10.1016/j.jinf.2024.106297

  • 2.

    KomaroffALDantzerR. Causes of symptoms and symptom persistence in long COVID and myalgic encephalomyelitis/chronic fatigue syndrome. Cell Rep Med. (2025) 6:102259. doi: 10.1016/j.xcrm.2025.102259

  • 3.

    ThapaliyaKMarshall-GradisnikSEaton-FitchNEftekhariZInderyasMBarndenL. Imbalanced brain neurochemicals in long COVID and ME/CFS: a preliminary study using MRI. Am J Med. (2025) 138:56774.e1. doi: 10.1016/j.amjmed.2024.04.007

  • 4.

    CarruthersBMJainAKDe MeirleirKLDanielLPNancyGKMartinLAet al. Myalgic encephalomyelitis/chronic fatigue syndrome: clinical working case definition, diagnostic and treatment protocols. J Chronic Fatigue Syndr. (2003) 11:7115. doi: 10.1300/J092v11n01_02

  • 5.

    CarruthersBMvan de SandeMIDe MeirleirKLKlimasNGBroderickGMitchellTet al. Myalgic encephalomyelitis: international consensus criteria. J Intern Med. (2011) 270:32738. doi: 10.1111/j.1365-2796.2011.02428.x

  • 6.

    SorianoJBMurthySMarshallJCRelanPDiazJV. A clinical case definition of post-COVID-19 condition by a Delphi consensus. Lancet Infect Dis. (2022) 22:e1027. doi: 10.1016/S1473-3099(21)00703-9

  • 7.

    ErlandsonKMGengLNSelvaggiCAThaweethaiTChenPErdmannNBet al. Differentiation of prior SARS-CoV-2 infection and postacute sequelae by standard clinical laboratory measurements in the RECOVER cohort. Ann Intern Med. (2024) 177:120921. doi: 10.7326/M24-0737

  • 8.

    GreenhalghTSivanMPerlowskiANikolich. Long COVID: a clinical update. Lancet. (2024) 404:70724. doi: 10.1016/S0140-6736(24)01136-X

  • 9.

    Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: the biology of a neglected disease-PMC n.d. Available online at: https://pmc.ncbi.nlm.nih.gov/articles/PMC11180809/ (accessed February 19, 2026).

  • 10.

    LeonelJWCiurleoGCVFormigaAMVasconcelos deTMFAndrade deMHFeitosaWLQet al. Long COVID: neurological manifestations-an updated narrative review. Dement Neuropsychol. (2024) 18:e20230076. doi: 10.1590/1980-5764-dn-2023-0076

  • 11.

    KomaroffALLipkinWI. ME/CFS and Long COVID share similar symptoms and biological abnormalities: road map to the literature. Front Med. (2023) 10:1187163. doi: 10.3389/fmed.2023.1187163

  • 12.

    MaksoudRMagawaCEaton-FitchNThapaliyaKMarshall-GradisnikS. Biomarkers for myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS): a systematic review. BMC Med. (2023) 21:189. doi: 10.1186/s12916-023-02893-9

  • 13.

    RudroffTKlénRRainioOTuulariJ.The untapped potential of dimension reduction in neuroimaging: artificial intelligence-driven multimodal analysis of long COVID fatigue. Brain Sci. (2024) 14:1209. doi: 10.3390/brainsci14121209

  • 14.

    AfzaliMPieciakTNewmanSGaryfallidisEÖzarslanEChengHet al. The sensitivity of diffusion MRI to microstructural properties and experimental factors. J Neurosci Methods. (2021) 347:108951. doi: 10.1016/j.jneumeth.2020.108951

  • 15.

    SoaresJMMarquesPAlvesVSousaN.A hitchhiker's guide to diffusion tensor imaging. Front Neurosci. (2013) 7:31. doi: 10.3389/fnins.2013.00031

  • 16.

    MädlerBDrabyczSAKolindSHWhittallKPMacKayAL.Is diffusion anisotropy an accurate monitor of myelination? Correlation of multicomponent T2 relaxation and diffusion tensor anisotropy in human brain. Magn Reson Imaging. (2008) 26:87488. doi: 10.1016/j.mri.2008.01.047

  • 17.

    MoriSZhangJ. Principles of diffusion tensor imaging and its applications to basic neuroscience research. Neuron. (2006) 51:52739.

  • 18.

    O'DonnellLJWestinC-F. An introduction to diffusion tensor image analysis. Neurosurg Clin N Am. (2011) 22:1858. doi: 10.1016/j.nec.2010.12.004

  • 19.

    WinklewskiPJSabiszANaumczykPJodzioKSzurowskaESzarmachA. Understanding the physiopathology behind axial and radial diffusivity changes—what do we know?Front Neurol. (2018) 9:92. doi: 10.3389/fneur.2018.00092

  • 20.

    BoppMHAEmdeJCarlBNimskyCSaßB. Diffusion kurtosis imaging fiber tractography of major white matter tracts in neurosurgery. Brain Sci. (2021) 11:381. doi: 10.3390/brainsci11030381

  • 21.

    ZhuJZhuoCQinWWangDMaXZhouYet al. Performances of diffusion kurtosis imaging and diffusion tensor imaging in detecting white matter abnormality in schizophrenia. NeuroImage Clin. (2015) 7:1706. doi: 10.1016/j.nicl.2014.12.008

  • 22.

    LanzafameSGiannelliMGaraciFFlorisRDuggentoAGuerrisiMet al. Differences in Gaussian diffusion tensor imaging and non-Gaussian diffusion kurtosis imaging model-based estimates of diffusion tensor invariants in the human brain. Med Phys. (2016) 43:246475. doi: 10.1118/1.4946819

  • 23.

    KornaropoulosENWinzeckSRumetshoferTWikstromAKnutssonLCorreiaMMet al. Sensitivity of diffusion MRI to white matter pathology: influence of diffusion protocol, magnetic field strength, and processing pipeline in systemic lupus erythematosus. Front Neurol. (2022) 13:837385. doi: 10.3389/fneur.2022.837385

  • 24.

    GuoYLLiSJZhangZPShenZWZhangGSYanGet al. Parameters of diffusional kurtosis imaging for the diagnosis of acute cerebral infarction in different brain regions. Exp Ther Med. (2016) 12:9338. doi: 10.3892/etm.2016.3390

  • 25.

    HansenB. An introduction to kurtosis fractional anisotropy. AJNR Am J Neuroradiol. (2019) 40:163841. doi: 10.3174/ajnr.A6235

  • 26.

    KoerteIKWiegandTLTBonkeEMKochsiekJShentonME. Diffusion imaging of sport-related repetitive head impacts—a systematic review. Neuropsychol Rev. (2023) 33:12243. doi: 10.1007/s11065-022-09566-z

  • 27.

    GhaderiSMohammadiSFatehiFA. systematic review of diffusion microstructure imaging (DMI): current and future applications in neurology research. Brain Disord. (2025) 19:100238. doi: 10.1016/j.dscb.2025.100238

  • 28.

    ThapaliyaKMarshall-GradisnikSStainesDBarndenL. Diffusion tensor imaging reveals neuronal microstructural changes in myalgic encephalomyelitis/chronic fatigue syndrome. Eur J Neurosci. (2021) 54:621428. doi: 10.1111/ejn.15413

  • 29.

    Distinct white matter alteration patterns in post-infectious and gradual onset chronic fatigue syndrome revealed by diffusion MRI | Scientific Reports n.d. Available online at: https://www.nature.com/articles/s41598-025-09379-z (accessed October 31, 2025).

  • 30.

    ZeinehMMKangJAtlasSWRamanMMReissALNorrisJLet al. Right arcuate fasciculus abnormality in chronic fatigue syndrome. Radiology. (2015) 274:51726. doi: 10.1148/radiol.14141079

  • 31.

    KimuraYSatoNOtaMShigemotoYMorimotoEEnokizonoMet al. Brain abnormalities in myalgic encephalomyelitis/chronic fatigue syndrome: evaluation by diffusional kurtosis imaging and neurite orientation dispersion and density imaging. J Magn Reson Imaging. (2019) 49:81824. doi: 10.1002/jmri.26247

  • 32.

    KimuraYSatoNOtaMShigemotoYMorimotoEEnokizonoMet al. Abnormal brain diffusivity in participants with persistent neuropsychiatric symptoms after COVID-19. NeuroImmune Pharmacol Ther. (2023) 2:3748. doi: 10.1515/nipt-2022-0016

  • 33.

    Díez-CirardaMYusMGómez-RuizNPoliduraCGil-MartínezLDelgado-AlonsoCet al. Multimodal neuroimaging in post-COVID syndrome and correlation with cognition. Brain. (2023) 146:214252. doi: 10.1093/brain/awac384

  • 34.

    ThapaliyaKMarshall-GradisnikSInderyasMBarndenL. Altered brain tissue microstructure and neurochemical profiles in long COVID and recovered COVID-19 individuals: a multimodal MRI study. Brain Behav Immun Health. (2025) 50:101142. doi: 10.1016/j.bbih.2025.101142

  • 35.

    YuanMLuRLiuYZhuHWangHWangJet al. White matter changes in recovered COVID-19 patients: insights from DTI, DKI, and NODDI metrics. Front Neurol. (2025) 16: 1580262. doi: 10.3389/fneur.2025.1580262

  • 36.

    PritschetLTaylorCMCossioDFaskowitzJSantanderTHandwerkerDAet al. Neuroanatomical changes observed over the course of a human pregnancy. Nat Neurosci. (2024) 27:225360. doi: 10.1038/s41593-024-01741-0

  • 37.

    BridgesRS. Long-term alterations in neural and endocrine processes induced by motherhood. Horm Behav. (2016) 77:193203. doi: 10.1016/j.yhbeh.2015.09.001

  • 38.

    ThapaliyaKMarshall-GradisnikSEaton-FitchNBarthMInderyasMBarndenL. Hippocampal subfield volume alterations and associations with severity measures in long COVID and ME/CFS: a 7T MRI study. PLoS ONE. (2025) 20:e0316625. doi: 10.1371/journal.pone.0316625

  • 39.

    WagnerDNisenbaumRHeimCJonesJFUngerERReevesWC. Psychometric properties of the CDC symptom inventory for assessment of chronic fatigue syndrome. Popul Health Metr. (2005) 3:8. doi: 10.1186/1478-7954-3-8

  • 40.

    WeigelBEaton-FitchNThapaliyaKMarshall-GradisnikS. Sustained illness burden over time among Australians with myalgic encephalomyelitis/chronic fatigue syndrome. PLoS ONE. (2025) 20:e0338433. doi: 10.1371/journal.pone.0338433

  • 41.

    BaoSSZhaoCBaoXXRaoJS. Effect of b value on imaging quality for diffusion tensor imaging of the spinal cord at ultrahigh field strength. BioMed Res Int. (2021) 2021:4836804. doi: 10.1155/2021/4836804

  • 42.

    Martinez-HerasEGrussuFPradosFSolanaELlufriuS. Diffusion-weighted imaging: recent advances and applications. Semin Ultrasound CT MRI. (2021) 42:490506. doi: 10.1053/j.sult.2021.07.006

  • 43.

    TròRRoascioMTortoraDSeverinoMRossiACohen-AdadJet al. Diffusion kurtosis imaging of neonatal spinal cord in clinical routine. Front Radiol. (2022) 2:794981. doi: 10.3389/fradi.2022.794981

  • 44.

    YanXZhouMYingLYinDFanMYangGet al. Evaluation of optimized b-value sampling schemas for diffusion kurtosis imaging with an application to stroke patient data. Comput Med Imaging Graph Off J Comput Med Imaging Soc. (2013) 37:27280. doi: 10.1016/j.compmedimag.2013.04.007

  • 45.

    AnderssonJLRSkareSAshburnerJ. How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging. Neuroimage. (2003) 20:87088. doi: 10.1016/S1053-8119(03)00336-7

  • 46.

    AnderssonJLRSotiropoulosSN. An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging. Neuroimage. (2016) 125:106378. doi: 10.1016/j.neuroimage.2015.10.019

  • 47.

    YamadaHAbeOShizukuishiTKikutaJShinozakiTDezawaKet al. Efficacy of distortion correction on diffusion imaging: comparison of FSL eddy and eddy_correct using 30 and 60 directions diffusion encoding. PLoS ONE. (2014) 9:e112411. doi: 10.1371/journal.pone.0112411

  • 48.

    SmithSM. Fast robust automated brain extraction. Hum Brain Mapp. (2002) 17:14355. doi: 10.1002/hbm.10062

  • 49.

    SmithSMJenkinsonMWoolrichMWBeckmannCFBehrensTEJohansen-BergHet al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage. (2004) 23:S20819. doi: 10.1016/j.neuroimage.2004.07.051

  • 50.

    TournierJDSmithRRaffeltDTabbaraRDhollanderTPietschMet al. MRtrix3: a fast, flexible and open software framework for medical image processing and visualisation. Neuroimage. (2019) 202:116137. doi: 10.1016/j.neuroimage.2019.116137

  • 51.

    Ades-AronBVeraartJKochunovPMcGuireSShermanPKellnerEet al. Evaluation of the accuracy and precision of the diffusion parameter EStImation with Gibbs and NoisE removal pipeline. Neuroimage. (2018) 183:53243. doi: 10.1016/j.neuroimage.2018.07.066

  • 52.

    SmithSMJenkinsonMJohansen-BergHRueckertDNicholsTEMackayCEet al. Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroimage. (2006) 31:1487505. doi: 10.1016/j.neuroimage.2006.02.024

  • 53.

    Statistical Parametric Mapping. ScienceDirect n.d. Available online at:https://www.sciencedirect.com/book/edited-volume/9780123725608/statistical-parametric-mapping (accessed January 6, 2026).

  • 54.

    xjView | A viewing program for SPM n.d. Available online at: https://www.alivelearn.net/xjview/ (accessed January 6, 2026).

  • 55.

    JenkinsonMBeckmannCFBehrensTEWoolrichMWSmithSM. FSL. Neuroimage. (2012) 62:78290. doi: 10.1016/j.neuroimage.2011.09.015

  • 56.

    BonferroniCE. Teoria statistica delle classi e calcolo delle probabilità. Pubblicazioni R Ist Super Sci Econ E Commericiali Firenze. (1936) 8:362.

  • 57.

    BenjaminiYHochbergY. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B Methodol. (1995) 57:289300. doi: 10.1111/j.2517-6161.1995.tb02031.x

  • 58.

    JellisonBJFieldASMedowJLazarMSalamatMSAlexanderAL. Diffusion tensor imaging of cerebral white matter: a pictorial review of physics, fiber tract anatomy, and tumor imaging patterns. Am J Neuroradiol. (2004) 25:35669.

  • 59.

    YuQKwiatekRADel FantePBonnerACalhounVDBatemanGAet al. Distinct white matter alteration patterns in post-infectious and gradual onset chronic fatigue syndrome revealed by diffusion MRI. Sci Rep. (2025) 15:24256. doi: 10.1038/s41598-025-09379-z

  • 60.

    LinLBhawanaRXueYDuanQJiangRChen H„ etal. Comparative analysis of diffusional kurtosis imaging, diffusion tensor imaging, and diffusion-weighted imaging in grading and assessing cellular proliferation of meningiomas. AJNR Am J Neuroradiol. (2018) 39:10328. doi: 10.3174/ajnr.A5662

  • 61.

    KarlsenRHEinarsenCMoeHKHåbergAKVikASkandsenTet al. Diffusion kurtosis imaging in mild traumatic brain injury and postconcussional syndrome. J Neurosci Res. (2019) 97:56881. doi: 10.1002/jnr.24383

  • 62.

    PlankJRMorganCADell'AcquaFSundramFHoehNRMuthukumaraswamySet al. Mapping neuroinflammation with diffusion-weighted magnetic resonance imaging: a randomized crossover study. Biol Psychiatry Cogn Neurosci Neuroimaging. (2025) 10:94453. doi: 10.1016/j.bpsc.2025.05.002

  • 63.

    NataliALReddyVBordoniB. “Neuroanatomy, corticospinal cord tract,” in StatPearls (Treasure Island: StatPearls Publishing) (2026).

  • 64.

    BhardwajNYadalaS. “Neuroanatomy, corticobulbar tract,”StatPearls (Treasure Island: StatPearls Publishing) (2026).

  • 65.

    GeorgeKDasJM. “Neuroanatomy, thalamocortical radiations,”StatPearls (Treasure Island: StatPearls Publishing) (2026).

  • 66.

    LouçãoRNunesRGNeto-HenriquesRCorreiaMFerreiraH. Human brain tractography: a DTI vs. DKI comparison analysis. 2015 IEEE 4th Portuguese Meeting on Bioengineering (ENBENG) New York, NY: IEEE. (2015). p. 12.

  • 67.

    ShahimPHolleranLKimJHBrodyDL. Test-retest reliability of high spatial resolution diffusion tensor and diffusion kurtosis imaging. Sci Rep. (2017) 7:11141. doi: 10.1038/s41598-017-11747-3

  • 68.

    WuKWuZFengSZhouTNingYLiKet al. Microstructural white matter impairments in chronic fatigue syndrome: evidence of segmental injury in the cingulum bundle. Brain Res Bull. (2026) 234:111671. doi: 10.1016/j.brainresbull.2025.111671

  • 69.

    NachevPKennardCHusainM. Functional role of the supplementary and pre-supplementary motor areas. Nat Rev Neurosci. (2008) 9:85669. doi: 10.1038/nrn2478

  • 70.

    RollsET. The cingulate cortex and limbic systems for emotion, action, and memory. Brain Struct Funct. (2019) 224:300118. doi: 10.1007/s00429-019-01945-2

  • 71.

    BubbEJMetzler-BaddeleyCAggletonJP. The cingulum bundle: anatomy, function, and dysfunction. Neurosci Biobehav Rev. (2018) 92:10427. doi: 10.1016/j.neubiorev.2018.05.008

  • 72.

    KumarRNguyenHDMaceyPMWooMAHarperRM. Regional brain axial and radial diffusivity changes during development. J Neurosci Res. (2012) 90:34655. doi: 10.1002/jnr.22757

  • 73.

    HorsfieldMAJonesDK. Applications of diffusion-weighted and diffusion tensor MRI to white matter diseases-a review. NMR Biomed. (2002) 15:5707. doi: 10.1002/nbm.787

  • 74.

    RanzenbergerLRDasJMSnyderT. “Diffusion tensor imaging,” in StatPearls (Treasure Island: StatPearls Publishing) (2025).

  • 75.

    GlennGRHelpernJATabeshAJensenJH. Quantitative assessment of diffusional kurtosis anisotropy. NMR Biomed. (2015) 28:44859. doi: 10.1002/nbm.3271

  • 76.

    GoldsteinACovingtonBPMahabadiNMesfinFB. Neuroanatomy, corpus callosum. StatPearls (Treasure Island: StatPearls Publishing) (2025).

  • 77.

    FitsioriANguyenDKarentzosADelavelleJVargasMI. The corpus callosum: white matter or terra incognita. Br J Radiol. (2011) 84:518. doi: 10.1259/bjr/21946513

  • 78.

    BhattRRGadewarSPShettyAGariIBHaddadEJavidSet al. The genetic architecture of the human corpus callosum and its subregions. bioRxiv[Preprint]. bioRxiv: 2024.07.22.603147 (2024).

  • 79.

    KumarSDe LucaALeemansASaffariSEHartonoSZailanFZet al. Topology of diffusion changes in corpus callosum in Alzheimer's disease: an exploratory case-control study. Front Neurol. (2022) 13:1005406. doi: 10.3389/fneur.2022.1005406

  • 80.

    WirthKJScheibenbogenCPaulF. An attempt to explain the neurological symptoms of myalgic encephalomyelitis/chronic fatigue syndrome. J Transl Med. (2021) 19:471. doi: 10.1186/s12967-021-03143-3

  • 81.

    Committee Committee on the Diagnostic Criteria for Myalgic Encephalomyelitis/ChronicFatigue Syndrome Board Board on the Health of Select Populations Institute Institute of Medicine. Beyond Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: Redefining an Illness. Washington, DC: National Academies Press (US) (2015).

  • 82.

    StollerFHindsEIonescuTKhatamsazEMarstonHMHengererB. Forceps minor control of social behaviour. Sci Rep. (2024) 14:30492. doi: 10.1038/s41598-024-81930-w

  • 83.

    HuangSZhouZYangDZhaoWZengMXieXet al. Persistent white matter changes in recovered COVID-19 patients at the 1-year follow-up. Brain. (2021) 145:18308. doi: 10.1093/brain/awab435

  • 84.

    BispoDDCBrandãoPRPPereiraDAMalufFBDiasBAParanhosHRet al. Brain microstructural changes and fatigue after COVID-19. Front Neurol. (2022) 13:1029302.

  • 85.

    ArrigoniAPrevitaliMBosticardoSPezzettiGPoloniSCapelliSet al. Brain microstructure and connectivity in COVID-19 patients with olfactory or cognitive impairment. NeuroImage Clin. (2024) 43:103631. doi: 10.1016/j.nicl.2024.103631

  • 86.

    MamiyaPCRichardsTLKuhlPK. Right forceps minor and anterior thalamic radiation predict executive function skills in young bilingual adults. Front Psychol. (2018) 9:118. doi: 10.3389/fpsyg.2018.00118

  • 87.

    Diez-CirardaMYus-FuertesMPoliduraCGil-MartinezLDelgado-AlonsoCDelgado-ÁlvarezAet al. Neural basis of fatigue in post-COVID syndrome and relationships with cognitive complaints and cognition. Psychiatry Res. (2024) 340:116113. doi: 10.1016/j.psychres.2024.116113

  • 88.

    TummalaSRoyBVigRParkBKangDWWooMAet al. Non-gaussian diffusion imaging shows brain myelin and axonal changes in obstructive sleep apnea. J Comput Assist Tomogr. (2017) 41:1819. doi: 10.1097/RCT.0000000000000537

  • 89.

    WangSPetersonDJWangYWangQGrabowskiTJLiWet al. Empirical comparison of diffusion kurtosis imaging and diffusion basis spectrum imaging using the same acquisition in healthy young adults. Front Neurol. (2017) 8:118. doi: 10.3389/fneur.2017.00118

  • 90.

    MaiterARiemerFAllinsonKZaccagnaFCrispin-OrtuzarMGehrungMet al. Investigating the relationship between diffusion kurtosis tensor imaging (DKTI) and histology within the normal human brain. Sci Rep. (2021) 11:8857. doi: 10.1038/s41598-021-87857-w

  • 91.

    BankerLTadiP. “Neuroanatomy, precentral gyrus,” in StatPearls (Treasure Island: StatPearls Publishing) (2025).

  • 92.

    TrollMLiMChandTMachnikMRocktäschelTToepfferAet al. Altered corticostriatal connectivity in long-COVID patients is associated with cognitive impairment. Psychol Med. (2025) 55:e49. doi: 10.1017/S0033291725000054

  • 93.

    SamanciBAyUGezegenHYörükSSMedetalibeyo?luAKurtEet al. Persistent neurocognitive deficits in long COVID: evidence of structural changes and network abnormalities following mild infection. Cortex. (2025) 187:98110. doi: 10.1016/j.cortex.2025.04.004

  • 94.

    GuoYPanNZouYLongYZhangXLiQet al. Neuroimaging insights into the psychosocial impact of the COVID-19 pandemic: a systematic review. Transl Psychiatry. (2025) 15:236. doi: 10.1038/s41398-025-03423-2

  • 95.

    KieselerM-LDuchaineB. Persistent prosopagnosia following COVID-19. Cortex J Devoted Study Nerv Syst Behav. (2023) 162:5664. doi: 10.1016/j.cortex.2023.01.012

  • 96.

    ZhuoJXuSProctorJLMullinsRJSimonJZFiskumGet al. Diffusion kurtosis as an in vivo imaging marker for reactive astrogliosis in traumatic brain injury. Neuroimage. (2012) 59:46777. doi: 10.1016/j.neuroimage.2011.07.050

  • 97.

    UstünTBChatterjiSKostanjsekNRehmJKennedyCEpping-JordanJet al. Developing the World Health Organization Disability Assessment Schedule 2.0. Bull World Health Organ. (2010) 88:81523. doi: 10.2471/BLT.09.067231

  • 98.

    WareJE. SF-36 health survey update. Spine. (2000) 25:31309. doi: 10.1097/00007632-200012150-00008

Summary

Keywords

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, ME/CFS, diffusion tensor imaging, DTI, diffusion kurtosis imaging, DKI, long COVID, MRI

Citation

Singh TB, Marshall-Gradisnik S, Barnden L, Eaton-Fitch N, Huynh TH, Inderyas M and Thapaliya K (2026) Microstructural alterations in brain tissue of ME/CFS and long COVID using diffusion tensor imaging and diffusion kurtosis imaging. Front. Med. 13:1824498. doi: 10.3389/fmed.2026.1824498

Received

06 March 2026

Revised

10 June 2026

Accepted

29 June 2026

Published

17 July 2026

Volume

13 - 2026

Edited by

Jiandi Wan, University of California, Davis, CA, United States

Reviewed by

Nicola Manocchio, University of Rome Tor Vergata, Italy

Marine Tanashyan, Research Center of Neurology, Russia

Updates

Copyright

*Correspondence: Tanoj Bahadur Singh,

Disclaimer

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Outline

Figures

Cite article

Copy to clipboard


Export citation file


Share article

Article metrics