MRI Studies Research Articles

Behind the curtain lies the truth: a case of arrhythmogenic cardiomyopathy mistaken for COVID-19 vaccine-associated myocarditis.

A 15-year-old male presented with vasovagal syncope and troponin leak 4 days after his second COVID-19 vaccine. Based on initial diagnostic work-up, he was thought to have COVID-19 vaccine-associated myocarditis. His cardiac dysfunction persisted and further work-up including genetic evaluation and serial MRI studies later confirmed a diagnosis of arrhythmogenic cardiomyopathy. This is a unique case of an incorrect diagnosis based on timing and context of vaccine-related myocarditis. Reports of mild and self-limited myocarditis post-COVID-19 vaccination may cause vaccine hesitancy among the public, and so case reports such as this one show the importance of discerning underlying conditions amongst rare COVID-19 vaccination complications.

Disrupted functional connectivity of the emotion regulation network in major depressive disorder and its association with symptom improvement: A multisite resting-state functional MRI study.

The emotion regulation network (ERN) in the brain provides a framework for understanding the neuropathology of affective disorders. Although previous neuroimaging studies have investigated the neurobiological correlates of the ERN in major depressive disorder (MDD), whether patients with MDD exhibit abnormal functional connectivity (FC) patterns in the ERN and whether the abnormal FC in the ERN can serve as a therapeutic response signature remain unclear. A large functional magnetic resonance imaging dataset comprising 709 patients with MDD and 725 healthy controls (HCs) recruited across five sites was analyzed. Using a seed-based FC approach, we first investigated the group differences in whole-brain resting-state FC of the 14 ERN seeds between participants with and without MDD. Furthermore, an independent sample (45 MDD patients) was used to evaluate the relationship between the aforementioned abnormal FC in the ERN and symptom improvement after 8 weeks of antidepressant monotherapy. Compared to the HCs, patients with MDD exhibited aberrant FC between 7 ERN seeds and several cortical and subcortical areas, including the bilateral middle temporal gyrus, bilateral occipital gyrus, right thalamus, calcarine cortex, middle frontal gyrus, and the bilateral superior temporal gyrus. In an independent sample, these aberrant FCs in the ERN were negatively correlated with the reduction rate of the HAMD17 score among MDD patients. These results might extend our understanding of the neurobiological underpinnings underlying unadaptable or inflexible emotional processing in MDD patients and help to elucidate the mechanisms of therapeutic response.

Reduced volume of the mediodorsal and anteroventral thalamus is associated with anxiety in Parkinson's disease: A cross-sectional 7-tesla MRI study

Background Parkinson's disease (PD)-related anxiety occurs frequently and may be associated with imbalance between anxiety-related circuits. While the thalamus is a shared region of these circuits, its role in PD-related anxiety has not been explored so far. Objective To identify changes in volume of the thalamus and its subnuclei in patients with PD-related anxiety. Methods Cognitively intact PD patients (n = 105) were divided into two groups based on their score on the Parkinson anxiety scale (PAS): 31 PD patients had anxiety (Anx-PD) and 74 did not have anxiety (non-Anx-PD). Forty-five healthy control subjects were included. Participants underwent 7-Tesla MRI scanning. Using automatic segmentation, the volumes of the thalamus and its subnuclei were measured, compared between the groups and regressed on the PAS. Results The volumes of the thalamus and its subnuclei did not significantly differ between the groups. However, in anxious PD patients, more severe anxiety was strongly associated with a smaller volume of the right medial thalamic subregion, specifically the right mediodorsal magnocellular nucleus and the right mediodorsal parvocellular nucleus (R = 0.63, ßPAS = −0.546, p-valuemodel = 0.007 and R = 0.60, ßPAS = −0.547, p-valuemodel = 0.016, respectively), and of the left anteroventral thalamus (R = 0.73, FDR p-valuemodel = 0.002, ßPAS = −0.407, p-valuePAS = 0.01). Conclusions A reduced volume of the mediodorsal and anteroventral thalamus, overlapping structures between the anxiety related circuits, are associated with more severe PD-related anxiety and may explain its high prevalence in the disease.

Localizing music’s “language of emotion” in the human brain: A functional MRI study of scale and emotion processing.

Localizing music’s “language of emotion” in the human brain: A functional MRI study of scale and emotion processing.

Phenotype differentiation and white matter preservation by early hematopoietic cell transplantation – Free-water diffusion tensor MRI study of the brain in mucopolysaccharidosis type I patients

Phenotype differentiation and white matter preservation by early hematopoietic cell transplantation – Free-water diffusion tensor MRI study of the brain in mucopolysaccharidosis type I patients

Mapping Activity and Functional Organisation of the Motor and Visual Pathways Using ADC-fMRI in the Human Brain.

In contrast to blood-oxygenation level-dependent (BOLD) functional MRI (fMRI), which relies on changes in blood flow and oxygenation levels to infer brain activity, diffusion fMRI (DfMRI) investigates brain dynamics by monitoring alterations in the apparent diffusion coefficient (ADC) of water. These ADC changes may arise from fluctuations in neuronal morphology, providing a distinctive perspective on neural activity. The potential of ADC as an fMRI contrast (ADC-fMRI) lies in its capacity to reveal neural activity independently of neurovascular coupling, thus yielding complementary insights into brain function. To demonstrate the specificity and value of ADC-fMRI, both ADC- and BOLD-fMRI data were collected at 3 T in human subjects during visual stimulation and motor tasks. The first aim of this study was to identify an acquisition design for ADC that minimises BOLD contributions. By examining the timings in responses, we report that ADC 0/1 timeseries (acquired with b values of 0 and 1 ms/ ) exhibit residual vascular contamination, while ADC 0.2/1 timeseries (with b values of 0.2 and 1 ms/ ) show minimal BOLD influence and higher sensitivity to neuromorphological coupling. Second, a general linear model was employed to identify activation clusters for ADC 0.2/1 and BOLD, from which the average ADC and BOLD responses were calculated. The negative ADC response exhibited a significantly reduced delay relative to the task onset and offset as compared to BOLD. This early onset further supports the notion that ADC is sensitive to neuromorphological rather than neurovascular coupling. Remarkably, in the group-level analysis, positive BOLD activation clusters were detected in the visual and motor cortices, while the negative ADC clusters mainly highlighted pathways in white matter connected to the motor cortex. In the averaged individual level analysis, negative ADC activation clusters were also present in the visual cortex. This finding confirmed the reliability of negative ADC as an indicator of brain function, even in regions with lower vascularisation such as white matter. Finally, we established that ADC-fMRI time courses yield the expected functional organisation of the visual system, including both grey and white matter regions of interest. Functional connectivity matrices were used to perform hierarchical clustering of brain regions, where ADC-fMRI successfully reproduced the expected structure of the dorsal and ventral visual pathways. This organisation was not replicated with the b = 0.2 ms/ diffusion-weighted time courses, which can be seen as a proxy for BOLD (via T2-weighting). These findings underscore the robustness of ADC time courses in functional MRI studies, offering complementary insights into BOLD-fMRI regarding brain function and connectivity patterns.

Key Pathophysiological Role of Skeletal Muscle Disturbance in Post COVID and Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS): Accumulated Evidence.

Recent studies provide strong evidence for a key role of skeletal muscle pathophysiology in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). In a 2021 review article on the pathophysiology of ME/CFS, we postulated that hypoperfusion and ischemia can result in excessive sodium and calcium overload in skeletal muscles of ME/CFS patients to cause mitochondrial damage. Since then, experimental evidence has been provided that supports this concept. We collect, summarize and discuss the current state of knowledge for the key role of skeletal muscle pathophysiology. We try to explain which risk factors and mechanisms are responsible for a subgroup of patients with post COVID syndrome (PCS) to develop ME/CFS (PC-ME/CFS). Mitochondrial dysfunction is a long-held assumption to explain cardinal symptoms of ME/CFS. However, mitochondrial dysfunction could not be convincingly shown in leukocytes. By contrast, recent studies provide strong evidence for mitochondrial dysfunction in skeletal muscle tissue in ME/CFS. An electron microscopy study could directly show damage of mitochondria in skeletal muscle of ME/CFS patients with a preferential subsarcolemmal localization but not in PCS. Another study shows signs of skeletal muscle damage and regeneration in biopsies taken one day after exercise in PC-ME/CFS. The simultaneous presence of necroses and signs of regeneration supports the concept of repeated damage. Other studies correlated diminished hand grip strength (HGS) with symptom severity and prognosis. A MRI study showed that intracellular sodium in muscles of ME/CFS patients is elevated and that levels correlate inversely with HGS. This finding corroborates our concept of sodium and consecutive calcium overload as cause of muscular and mitochondrial damage caused by enhanced proton-sodium exchange due to anaerobic metabolism and diminished activity of the sodium-potassium-ATPase. The histological investigations in ME/CFS exclude ischemia by microvascular obstruction, viral presence or immune myositis. The only known exercise-induced mechanism of damage left is sodium induced calcium overload. If ionic disturbance and mitochondrial dysfunction is severe enough the patient may be captured in a vicious circle. This energy deficit is the most likely cause of exertional intolerance and post exertional malaise and is further aggravated by exertion. Based on this pathomechanism, future treatment approaches should focus on normalizing the cause of ionic disbalance. Current treatment strategies targeting hypoperfusion have the potential to improve the dysfunction of ion transporters.

Abstract WP203: Ipsilateral Hemodynamic Failure of Extracranial Internal Carotid Artery is Associated with Severe Middle Cerebral Artery Atherosclerosis——a 4D Flow MRI Study

Abstract WP203: Ipsilateral Hemodynamic Failure of Extracranial Internal Carotid Artery is Associated with Severe Middle Cerebral Artery Atherosclerosis——a 4D Flow MRI Study

The role of sleep quality in mediating the relationship between habenula volume and resilience.

Our human volumetric MRI study (Dai et al., 2024) demonstrated that habenula (Hb) volume is associated with psychological resilience, a key protective factor against depression. However, the biological mechanisms underpinning this relationship remain unclear. A recent animal study highlighted that neuronal activity in the Hb modulates rapid eye movement (REM) sleep, influencing depressive behaviors during wakefulness. Based on this, we hypothesized that sleep quality mediates the relationship between Hb volume and psychological resilience in humans. We utilized a deep learning-based automated segmentation model to estimate Hb volume from 3T-MRI T1-weighted images of 84 healthy participants. Correlation analyses were performed to examine the relationship between Hb volume and questionnaire-based assessments of sleep quality. Mediation analysis was then conducted with Hb volume as the independent variable, psychological resilience as the dependent variable, and sleep quality as the mediator. Hb volume was found to be negatively correlated with sleep disturbance, indicating that individuals with larger Hb volumes experienced better sleep quality. A lateralization effect was also observed, where greater leftward asymmetry (larger left Hb volume compared to right) was associated with more severe sleep disturbances. Moreover, sleep quality was identified as a mediator in the relationship between Hb volume and psychological resilience. This study provides preliminary evidence of the association among Hb volume, sleep quality, and resilience. Sleep quality appears to be a critical mediator in the biological processes linking smaller Hb volumes to decreased psychological resilience. Enhancing sleep quality may be a promising approach for bolstering psychological resilience and reducing the risk of depression.

Evaluation of Spinal Deformity and its Progression in Pyogenic Spondylodiscitis: A Retrospective MRI Study of 59 Cases

Evaluation of Spinal Deformity and its Progression in Pyogenic Spondylodiscitis: A Retrospective MRI Study of 59 Cases

2678 Small vessel disease contributions to acute delirium: a pilot feasibility MRI study

Abstract Background and aims Delirium carries an eightfold risk of future dementia. Small vessel disease (SVD), best seen on MRI, increases delirium risk, yet delirium is understudied in MRI research. We aimed to determine MRI feasibility, tolerability, image usability, and prevalence of acute and chronic SVD lesions in acute delirium. Methods This case–control feasibility study performed MRI (3D T1/T2-weighted, FLAIR, Susceptibility-weighted, and Diffusion-weighted imaging (DWI) on 20 medical inpatients >65 years: 10 with delirium ≥3 weeks and 10 without delirium, matched for vascular risk, Clinical Frailty Scale (CFS), and cognitive status. We excluded acute stroke, agitation necessitating sedation, assistance of >2 staff to mobilise, and MRI contraindications. We measured scan duration, tolerability, image usability, acute infarcts on DWI, and chronic SVD features. Six months later, we recorded CFS and cognitive diagnoses. Results Mean age was 83.5 years (delirium 78.7 vs non-delirium 88.4); 13/20 were female; 17/20 had premorbid cognitive decline/impairment or dementia. Acquisition took mean 26.8 minutes. MRI was well-tolerated in 16/20 (7/10 in delirium arm; 9/10 in non-delirium arm). 4/20 had early scan termination but 20/20 had clinically interpretable images. We detected DWI-hyperintense lesions in 3/10 (33.3%) with delirium (2/10 small subcortical and 1/10 cortical) and in 3/10 (33.3%) without delirium (2/10 small subcortical; 1/10 cortical). Mean SVD score was 2.4 in delirium vs 3.3 without. Conclusions MRI is feasible, usable, and tolerable in delirium, and we detected DWI hyperintense lesions in one third of patients overall. This study indicates acute vascular contributions, including SVD, to delirium, supporting the need for larger studies.

Assessment of hemo- and cerebrospinal fluid dynamics disorders in idiopathic normal pressure hydrocephalus according to MRI data: a prospective study

INTRODUCTION: Idiopathic normal pressure hydrocephalus (NPH) is a condition characterized by enlargement of the cerebral ventricles and an isolated disturbance of cerebrospinal fluid dynamics, the etiology and pathogenesis of which are still not fully determined.OBJECTIVE: To evaluate changes in hemo- and cerebrospinal fluid dynamics in idiopathic normal pressure hydrocephalus according to phase-contrast MRI data.MATERIALS AND METHODS: Three groups of subjects were formed: patients with NPH (12 people), elderly volunteers with atrophic ventriculomegaly (15 people), a group of healthy volunteers (15 people). For each subject, the data from a routine MRI study, the volumes of gray, white matter and cerebrospinal fluid were assessed, and the volume-velocity characteristics of CSF and blood flows at several levels were calculated.Statistics: For quantitative indicators of cerebrospinal fluid dynamics, the median (Me), 25% and 75% percentiles, and interquartile range were calculated. An intergroup assessment of the significance of differences was carried out using the Mann-Whitney U test and nonparametric multivariate analysis of variance MANOVA.RESULTS: Individual neuroimaging markers of IGT were determined, as well as an increase in the volume-velocity characteristics of CSF flow at the level of the cerebral aqueduct (p<0.01, with a predominance of the retrograde component) in comparison with the control group and patients with atrophic ventriculomegaly. Multidirectional changes in the volume of intracranial venous outflow were revealed in the study groups: in patients with IGT – a decrease in outflow along the straight sinus by 1.4 times (p<0.01), in patients with atrophic ventriculomegaly – a decrease in outflow along the superior sagittal sinus by 1.3 times (p<0.05).DISCUSSION: The data obtained show a significant impairment of cerebrospinal fluid dynamics in patients with NPH, in contrast to elderly volunteers with age-related atrophy and replacement expansion of the cerebrospinal fluid spaces at the level of the brain aqueduct, and also indicates the importance of the venous link in maintaining intracranial volumetric interactions.CONCLUSION: The combined use of a routine protocol and phase-contrast MRI techniques made it possible to identify a number of neuroimaging, hemo- and CSF dynamic changes in patients with NPH in comparison with healthy volunteers, and also, most importantly, with patients of comparable age with the presence of replacement ventriculomegaly against the background of atrophy.

FMRI data acquisition and analysis for task-free, anesthetized rats.

Templates for the acquisition of large datasets such as the Human Connectome Project guide the neuroimaging community to reproducible data acquisition and scientific rigor. By contrast, small animal neuroimaging often relies on laboratory-specific protocols, which limit cross-study comparisons. The establishment of broadly validated protocols may facilitate the acquisition of large datasets, which are essential for uncovering potentially small effects often seen in functional MRI (fMRI) studies. Here, we outline a procedure for the acquisition of fMRI datasets in rats and describe animal handling, MRI sequence parameters, data conversion, preprocessing, quality control and data analysis. The procedure is designed to be generalizable across laboratories, has been validated by using datasets across 20 research centers with different scanners and field strengths ranging from 4.7 to 17.2 T and can be used in studies in which it is useful to compare functional connectivity measures across an extensive range of datasets. The MRI procedure requires 1 h per rat to complete and can be carried out by users with limited expertise in rat handling, MRI and data processing.

Remote neurodegeneration in the lumbosacral cord one month after spinal cord injury: a cross-sectional MRI study.

To characterize structural integrity of the lumbosacral enlargement and conus medullaris within one month after spinal cord injury (SCI). Lumbosacral cord MRI data were acquired in patients with sudden onset (<7 days) SCI at the cervical or thoracic level approximately one month after injury and in healthy controls. Tissue integrity and loss were evaluated through diffusion tensor (DTI) and T2*-weighted imaging (cross-sectional area [CSA] measurements). Associations with the degree of neurological impairment were assessed using linear mixed-effects models. Twenty-one patients with SCI showed lower white matter (WM) fractional anisotropy (FA) (≤-13.3%) and higher WM radial diffusivity (≤14.6%) compared to 27 healthy controls. Differences were most pronounced in the lateral columns of WM. CSA measurements revealed no group differences. For the lateral columns, lower FA values were associated with lower motor scores and lower amplitudes of motor evoked potentials. For the dorsal columns, lower FA values were associated with lower amplitudes of somatosensory evoked potentials from the lower extremities. One month after SCI, first signs of WM degeneration were apparent, without indication of tissue loss. The more pronounced differences observed in the lateral column could be attributed to anterograde degeneration of the motor tracts. The variability among DTI measurements remote from the lesion site can be partially explained by the degree of the SCI-induced neurological impairment. Together with previous studies, our findings indicate that impaired tissue integrity precedes tissue loss. The presented techniques have potential applications in monitoring the progression of various neurological diseases.

Peri-anal Hyperpigmentation in a Patient with Rectal Cavernous Haemangioma

Introduction: Cavernous hemangioma of rectum is a slow-flow vascular malformation characterized by multiple dilated intramural blood vessels which shows T2 hyperintense signal, mainly on fat-suppressed sequences and may cause bleeding. Case Details: The report describes a case of a 16 Years boy who presented with recurrent episodes of rectal bleeding with hyperpigmentation in peri-anal region. MRI study suggested the diagnosis of rectal cavernous hemangioma. Conclusion: The illustrated case images highlight the role of clinical examination and MRI in diagnosis of rectal cavernous hemangioma. Peri-anal hyperpigmentation in a young patient with rectal bleeding should raise suspicion of rectal hemangioma.

Hormonal Influences on ADC Values in Breast Tissues: A Scoping Review of DWI in Pre- and Post-menopausal Women

Background Breast cancer remains a significant global health concern, with early diagnosis and risk factor identification crucial for improving outcomes. Diffusion-Weighted Imaging (DWI) and Apparent Diffusion Coefficient (ADC) measurements have emerged as promising tools in breast cancer diagnostics. However, the influence of hormonal status on these measurements remains unclear. Objective This scoping review aims to synthesize current evidence on how hormonal changes in pre- and post-menopausal women influence ADC values of benign, malignant, and fibroglandular breast tissues. Method Following the Arksey and O’Malley framework, we conducted a comprehensive search of Scopus, Embase, and PubMed databases for relevant studies published between January 2000 and 2021. Inclusion criteria encompassed 1.5 Tesla MRI studies reporting ADC values in female subjects, considering menopausal status. Results Six studies meeting the inclusion criteria, involving 612 patients, were analyzed. Findings suggest that menopausal status may influence ADC values, with postmenopausal women generally showing lower ADC values in both normal fibroglandular tissue and breast lesions. The impact of menstrual cycle phases on ADC values was less consistent across studies. Conclusions This review highlights the potential influence of hormonal status on ADC values in breast tissues. While DWI with ADC mapping shows promise as a reliable diagnostic tool across different hormonal states, further research is needed to fully understand and account for hormonal influences on ADC measurements. Future studies should focus on longitudinal designs, standardization of DWI protocols, and integration of hormonal status information into breast cancer risk assessment models.

The impact of resective epilepsy surgery on the brain network: evidence from post-surgical imaging.

Resective epilepsy surgery can be an effective treatment for patients with medication-resistant focal epilepsy. Epilepsy resection consists of the surgical removal of an epileptic focus to stop seizure generation and disrupt the epileptic network. However, even focal surgical resections for epilepsy lead to widespread brain network changes. Understanding the impact of epilepsy surgery on the brain is crucial to improve surgical outcomes for patients. Here we provide a summary of studies imaging the postsurgical effects of epilepsy resection on the brain. We focus on MRI and PET studies of temporal lobe and pediatric epilepsy, reflecting the current literature. We discuss three potential mechanisms for surgery-induced brain changes: damage and degeneration, recovery, and reorganization. We additionally review the postsurgical brain correlates of surgical outcomes as well as the potential to predict the impact of surgery on an individual patient's brain. A comprehensive characterization of the impact of surgery on the brain and precise methods to predict these brain network changes could lead to more personalized surgeries that improve seizure outcomes and reduce neuropsychological deficits after surgery.

Mapping the aging brain: Insights into microstructural changes from free water-corrected fractional anisotropy.

Aging has a significant impact on brain structure, demonstrated by numerous MRI studies using diffusion tensor imaging (DTI). While these studies reveal changes in fractional anisotropy (FA) across different brain regions, they tend to focus on white matter tracts and cognitive regions, often overlooking gray matter and motor areas. Additionally, traditional DTI metrics can be affected by partial volume effects. To address these limitations and gain a better understanding of microstructural changes across the whole brain, we utilized free water-corrected fractional anisotropy (FAt) to examine aging-related microstructural changes in a group of 20 young adults (YA) and 24 older adults (OA). A voxel-wise analysis revealed that YA had higher FAt values predominantly in white matter tracts associated with both motor and non-motor functions. In contrast, OA showed higher levels of FAt primarily in gray matter regions, including both subcortical and cortical motor areas, and occipital and temporal cortices. Complementing these cross-sectional results, correlation analyses within the OA group showed that many of these changes are further exacerbated with increasing age, underscoring the progressive nature of these microstructural alterations. In summary, the distinct patterns of FAt changes in gray versus white matter with aging suggest different underlying mechanisms. While white matter FAt values decrease, likely due to axonal degeneration, the increase in gray matter FAt could reflect either compensatory processes or pathological changes. Including behavioral data in future studies will be crucial for understanding the functional implications of these microstructural gray matter changes and their effects on cognitive and motor functions.

Functional MRI study of neurovascular coupling in patients with non-lesional epilepsy.

The diagnosis of patients with non-lesional epilepsy (NLE) is relatively challenging because of the absence of a clear focus on imaging, and the underlying pathological mechanism remains unclear. The neuronal activity and functional connectivity of NLE patients are significantly abnormal, and the neuronal activity of epilepsy patients is closely related to cerebral blood flow (CBF). Neurovascular coupling (NVC) offers insights into the relationship between neuronal activity and CBF. Hence, we intend to explore the alterations of NVC in NLE patients and their influences on cognitive function. Clinical data of 24 patients with NLE (15 female; age range 19-40 years; median age 30.5 years) and 39 healthy controls (27 female; age range 19-40 years; median age 30 years) were collected, and resting-state functional magnetic resonance imaging (rs-fMRI) and 3D arterial spin labeling (ASL) were performed. The imaging indexes of amplitude of low-frequency fluctuation (ALFF) and CBF were calculated, respectively, by post-processing analysis. The differences in CBF, ALFF and CBF/ALFF ratio between the two groups were analyzed, along with correlation with clinical data of NLE patients. Compared with the healthy controls, the CBF of the right parahippocampal gyrus was significantly decreased, and the CBF/ALFF ratio of the right inferior parietal, but supramarginal and angular gyri was significantly increased in NLE patients (p < 0.001). Moreover, the CBF/ALFF ratio was positively correlated with epilepsy depression score (r = 0.546, p = 0.006). NLE patients showed abnormal local NVC, which was associated with the severity of depression. The combined application of rs-fMRI and ASL can comprehensively evaluate the neuronal activity and cerebral blood perfusion in patients with NLE. The abnormal NVC is of great significance for us to explore the central mechanism of the occurrence and development of NLE.

Redefining diagnostic lesional status in temporal lobe epilepsy with artificial intelligence.

Despite decades of advancements in diagnostic MRI, 30-50% of temporal lobe epilepsy (TLE) patients remain categorized as "non-lesional" (i.e., MRI negative or MRI-) based on visual assessment by human experts. MRI- patients face diagnostic uncertainty and significant delays in treatment planning. Quantitative MRI studies have demonstrated that MRI- patients often exhibit a TLE-specific pattern of temporal and limbic atrophy that may be too subtle for the human eye to detect. This signature pattern could be successfully translated into clinical use via artificial intelligence (AI) advances in computer-aided MRI interpretation, thereby improving the detection of brain "lesional" patterns associated with TLE. Here, we tested this hypothesis by employing a three-dimensional convolutional neural network (3D CNN) applied to a dataset of 1,178 scans from 12 different centers. 3D CNN was able to differentiate TLE from healthy controls with high accuracy (85.9% ± 2.8), significantly outperforming support vector machines based on hippocampal (74.4% ± 2.6) and whole-brain (78.3% ± 3.3) volumes. Our analysis subsequently focused on a subset of patients who achieved sustained seizure freedom post-surgery as a gold standard for confirming TLE. Importantly, MRI- patients from this cohort were accurately identified as TLE 82.7% ± 0.9 of the time, an encouraging finding since clinically these were all patients considered to be MRI- (i.e., not radiographically different than controls). The saliency maps from the CNN revealed that limbic structures, particularly medial temporal, cingulate, and orbitofrontal areas, were most influential in classification, confirming the importance of the well-established TLE signature atrophy pattern for diagnosis. Indeed, the saliency maps were similar in MRI+ and MRI- TLE groups, suggesting that even when humans cannot distinguish more subtle levels of atrophy, these MRI- patients are on the same continuum common across all TLE patients. As such, AI can identify TLE lesional patterns and AI-aided diagnosis has the potential to greatly enhance the neuroimaging diagnosis of TLE and redefine the concept of "lesional" TLE.