Magnetic Resonance Imaging Data Research Articles

Contrastive machine learning reveals Parkinson’s disease specific features associated with disease severity and progression

Parkinson’s disease (PD) exhibits heterogeneity in terms of symptoms and prognosis, likely due to diverse neuroanatomical alterations. This study employs a contrastive deep learning approach to analyze Magnetic Resonance Imaging (MRI) data from 932 PD patients and 366 controls, aiming to disentangle PD-specific neuroanatomical alterations. The results reveal that these neuroanatomical alterations in PD are correlated with individual differences in dopamine transporter binding deficit, neurodegeneration biomarkers, and clinical severity and progression. The correlation with clinical severity is verified in an external cohort. Notably, certain proteins in the cerebrospinal fluid are strongly associated with PD-specific features, particularly those involved in the immune function. The most notable neuroanatomical alterations are observed in both subcortical and temporal regions. Our findings provide deeper insights into the patterns of brain atrophy in PD and potential underlying molecular mechanisms, paving the way for earlier patient stratification and the development of treatments to slow down neurodegeneration.

Efficient MRI image enhancement by improved denoising techniques for better skull stripping using attention module-based convolution neural network

ABSTRACT Anatomical structure preservation throughout the denoising process is a challenge in the domain of medical imaging. The Rician noise introduced through the acquisition procedure by the Magnetic Resonance Imaging (MRI) scanner distorts the images. In this study, denoising using Wavelet-based Non-Local Median Filter (WBNLMF) and a novel contrast-enhancement method termed Improved Minimum Intensity Error Intuitionistic Fuzzy Contrast Enhancement (IMIEIFCET) is suggested. This methodology gives superior results while maintaining the edges and the brightness of the original image. An Attention Module-based Convolution Neural Network (AM-CNN) is suggested in the research as a methodology for skull stripping from MRI data. With a mean Dice coefficient of 0.998, a Sensitivity of 0.9975, and a Specificity of 0.9985, the proposed network exhibits result that are comparable to those of the specified Deep Learning (DL)-based technique.

Greater ventral striatal functional connectivity in cigarette smokers relative to non-smokers across a spectrum of alcohol consumption.

Cigarette smoking is associated with elevated risk of disease and mortality and contributes to heavy healthcare-related economic burdens. The nucleus accumbens is implicated in numerous reward-related behaviors, including reinforcement learning and incentive salience. The established functional connectivity of the accumbens includes regions associated with motivation, valuation, and affective processing. Although the high comorbidity of cigarette smoking with drinking behaviors may collectively affect brain activity, there could be independent effects of smoking in alcohol use disorder that impact brain function and behavior. We hypothesized that smoking status, independent of alcohol use, would be associated with aberrations of nucleus accumbens functional connectivity to brain regions that facilitate reward processing, salience attribution, and inhibitory control. Resting state functional magnetic resonance imaging data from thirty-one nonsmokers and nineteen smoking individuals were analyzed using seed-based correlations of the bilateral accumbens with all other brain voxels. Statistical models accounted for drinks consumed per week. The smoking group demonstrated significantly higher functional connectivity between the left accumbens and the bilateral insula and anterior cingulate cortex, as well as hyperconnectivity between the right accumbens and the insula. Confirmatory analyses using the insula and cingulate clusters generated from the original analysis as seed regions reproduced the hyperconnectivity in smokers between the bilateral insular regions and the accumbens. In conclusion, smoking status had distinct effects on neural activity; hyperconnectivity between the accumbens and insula in smokers may reflect enhanced encoding of the reinforcing effects of smoking and greater orientation toward smoking-associated stimuli.

Effects of post-acute COVID-19 syndrome on cerebral white matter and emotional health among non-hospitalized individuals

IntroductionPost-acute COVID syndrome (PACS) is a growing concern, given its impact on mental health and quality of life. However, its effects on cerebral white matter remain poorly understood, particularly in non-hospitalized cohorts. The goals of this cross-sectional, observational study were to examine (1) whether PACS was associated with distinct alterations in white matter microstructure, compared to symptom-matched non-COVID viral infection; and (2) whether microstructural alterations correlated with indices of post-COVID emotional health.MethodsData were collected for 54 symptomatic individuals who tested positive for COVID-19 (mean age 41 ± 12 yrs., 36 female) and 14 controls who tested negative for COVID-19 (mean age 41 ± 14 yrs., 8 female), with both groups assessed an average of 4–5 months after COVID testing. Diffusion magnetic resonance imaging data were collected, and emotional health was assessed via the NIH emotion toolbox, with summary scores indexing social satisfaction, well-being and negative affect.ResultsDespite similar symptoms, the COVID-19 group had reduced mean and axial diffusivity, along with increased mean kurtosis and neurite dispersion, in deep white matter. After adjusting for social satisfaction, higher levels of negative affect in the COVID-19 group were also correlated with increased mean kurtosis and reduced free water in white matter.DiscussionThese results provide preliminary evidence that indices of white matter microstructure distinguish PACS from symptomatic non-COVID infection. Moreover, white matter effects seen in PACS correlate with the severity of emotional sequelae, providing novel insights into this highly prevalent disorder.

Difference in Laterality of the Dorsal Striatum in Schizoaffective Disorder

Background: Recent research has demonstrated that the dorsal striatum is directly associated with the integration of cognitive, sensory-motor, and motivational/emotional data. Disruptions in the corticostriatal circuit have been implicated in the pathophysiology of psychosis. The dorsal striatum was reported to show lateralized pathology in psychotic disorders. In this study, we aimed to analyze the laterality of the dorsal striatum with texture analysis of T2-weighted magnetic resonance imaging (MRI) images from schizoaffective disorder (SAD) patients. Methods: Twenty SAD patients, met the inclusion criteria and had available cranial MRI data were assigned as the patient group. Twenty healthy individuals were determined as the control group. Texture analysis values were obtained from striatum region of interests (ROI) generated from T2-weighted MRI images. Data are presented as mean and standard deviation. The suitability of the data for normal distribution was analyzed with the Kolmogorov-Smirnov test. Analysis of variance (ANOVA) test (Post Hoc TUKEY) was employed to compare the group data based on test findings. Results: There was no significant difference between the groups in terms of gender and age. There were differences in the values of texture analysis parameters of both caudate and putamen nuclei in comparison to controls. We identified differences in the left dorsal striatum nuclei in SAD. The differences in the putamen were more and more pronounced than in the caudate. Conclusions: Texture analyses suggest that the left dorsal striatum nuclei may be different in SAD patients. Further studies are needed to determine the pathophysiology of SAD and how it may affect disease treatment.

A diagnostic model for differentiating tuberculous spondylodiscitis from pyogenic spondylodiscitis based on pathogen-confirmed patients.

This study aimed to distinguish tuberculous spondylodiscitis (TS) from pyogenic spondylodiscitis (PS) based on laboratory, magnetic resonance imaging (MRI) and computed tomography (CT) findings. Further, a novel diagnostic model for differential diagnosis was developed. We obtained MRI, CT and laboratory data from TS and PS patients. Predictive models were built using binary logistic regression analysis. The receiver operating characteristic curve was analyzed. Both internal and external validation was performed. A total of 81 patients with PS (n = 46) or TS (n = 35) were enrolled. All patients had etiological evidence from the focal lesion. Disc signal or height preservation, skip lesion or multi segment (involved segments ≥ 3) involvement, paravertebral calcification, massive sequestra formation, subligamentous bone destruction, bone erosion with osteosclerotic margin, higher White Blood Cell Count (WBC) and positive result of tuberculosis infection T cell spot test (T-SPOT.TB) were more prevalent in the TS group. A diagnostic model was developed and included four predictors: WBC<7.265 * (10^9/L), skip lesion or involved segments ≥ 3, massive sequestra formation and subligamentous bone destruction. The model showed good sensitivity, specificity, and total accuracy (91.4%, 95.7%, and 93.8%, respectively); the area under the receiver operating characteristic curve (AUC) was 0.981, similar to the results of internal validation using bootstrap resampling (1000 replicates) and external validation set, indicating good clinical predictive ability. This study develop a good diagnostic model based on both CT and MRI, as well as laboratory findings, which may help clinicians distinguish between TS and PS.

Evaluation of Glymphatic System Development in Neonatal Brain via Diffusion Analysis along the Perivascular Space Index.

Glymphatic system is a recently discovered macroscopic waste clearance system associated with numerous neurological diseases. However, little is known about glymphatic system development in neonates. We sought to evaluate diffusion along the perivascular space (ALPS) index, a proxy for glymphatic system function, in neonates and investigate its potential associations with maturation, sex, and preterm birth. Diffusion magnetic resonance imaging (MRI) data in 418 neonates, including 92 preterm neonates (57 males) and 326 term neonates (175 males), from the Developing Human Connectome Project were used for evaluating ALPS index. Linear regression modeling was performed to assess group differences in the ALPS index according to preterm birth and sex. Pearson's and partial correlation analysis were performed to assess the association between the ALPS index and gestational age (GA) as well as postmenstrual age (PMA) at MRI. Moderation analysis was performed to assess the moderation effect of preterm birth on the relationship between the ALPS index and PMA. Compared to term neonates, preterm neonates exhibited lower ALPS indices (p < 0.001). The ALPS index positively correlated with PMA (p = 0.004) and GA (p < 0.001). Preterm birth (p = 0.013) had a significant moderation effect on the relationship between the ALPS index and PMA. Sex had no significant direct effect (p = 0.639) or moderation effect (p = 0.333) on ALPS index. Glymphatic system development is a dynamic process in neonates, which can be moderated by preterm birth, the ALPS index could serve as a sensitive biomarker for monitoring this process. ANN NEUROL 2024.

A transformer-based unified multimodal framework for Alzheimer's disease assessment

In Alzheimer's disease (AD) assessment, traditional deep learning approaches have often employed separate methodologies to handle the diverse modalities of input data. Recognizing the critical need for a cohesive and interconnected analytical framework, we propose the AD-Transformer, a novel transformer-based unified deep learning model. This innovative framework seamlessly integrates structural magnetic resonance imaging (sMRI), clinical, and genetic data from the extensive Alzheimer's Disease Neuroimaging Initiative (ADNI) database, encompassing 1651 subjects. By employing a Patch-CNN block, the AD-Transformer efficiently transforms image data into image tokens, while a linear projection layer adeptly converts non-image data into corresponding tokens. As the core, a transformer block learns comprehensive representations of the input data, capturing the intricate interplay between modalities. The AD-Transformer sets a new benchmark in AD diagnosis and Mild Cognitive Impairment (MCI) conversion prediction, achieving remarkable average area under curve (AUC) values of 0.993 and 0.845, respectively, surpassing those of traditional image-only models and non-unified multimodal models. Our experimental results confirmed the potential of the AD-Transformer as a potent tool in AD diagnosis and MCI conversion prediction. By providing a unified framework that jointly learns holistic representations of both image and non-image data, the AD-Transformer paves the way for more effective and precise clinical assessments, offering a clinically adaptable strategy for leveraging diverse data modalities in the battle against AD.

Fast model calibration for predicting the response of breast cancer to chemotherapy using proper orthogonal decomposition

Constructing digital twins for predictive tumor treatment response models can have a high computational demand that presents a practical barrier for their clinical adoption. In this work, we demonstrate that proper orthogonal decomposition, by which a low-dimensional representation of the full model is constructed, can be used to dramatically reduce the computational time required to calibrate a partial differential equation model to magnetic resonance imaging (MRI) data for rapid predictions of tumor growth and response to chemotherapy. In the proposed formulation, the reduction basis is based on each patient’s own MRI data and controls the overall size of the “reduced order model”. Using the full model as the reference, we validate that the reduced order mathematical model can accurately predict response in 50 triple negative breast cancer patients receiving standard of care neoadjuvant chemotherapy. The concordance correlation coefficient between the full and reduced order models was 0.986 ± 0.012 (mean ± standard deviation) for predicting changes in both tumor volume and cellularity across the entire model family, with a corresponding median local error (inter-quartile range) of 4.36 % (1.22 %, 15.04 %). The total time to estimate parameters and to predict response dramatically improves with the reduced framework. Specifically, the reduced order model accelerates our calibration by a factor of (mean ± standard deviation) 378.4 ± 279.8 when compared to the full order model for a non-mechanically coupled model. This enormous reduction in computational time can directly help realize the practical construction of digital twins when the access to computational resources is limited.

Interval Changes in 4D Flow-Derived in vivo Hemodynamics Stratify Aortic Growth in Type B Aortic Dissection Patients

BackgroundAortic diameter growth in type B aortic dissection (TBAD) is associated with progressive aortic dilation, resulting in increased mortality in patients with both de novo TBAD (dnTBAD) and residual dissection after type A dissection repair (rTAAD). Preemptive thoracic endovascular aortic repair may improve mortality in patients with TBAD, although it is unclear which patients may benefit most from early intervention. In vivo hemodynamic assessment using four-dimensional (4D) flow magnetic resonance imaging (MRI) has been used to characterize TBAD patients with growing aortas. In this longitudinal study, we investigated whether changes over time in 4D flow derived true and false lumen (TL, FL) hemodynamic parameters correlate with aortic growth rate which is a marker of increased risk. MethodsWe retrospectively identified TBAD patients with baseline and follow-up 4D flow MRI at least 120 days apart. Patients with TBAD intervention before baseline or between scans were excluded. 4D flow MRI data analysis included segmentation of the TL and FL, followed by voxel-wise calculation of TL and FL total kinetic energy (KE), maximum velocity (MV), mean forward flow (FF), and mean reverse flow (RF). Changes over time (Δ) were calculated for all hemodynamic parameters. Maximal diameter in the descending aorta was measured from MR angiogram images acquired at the time of 4D flow. Aortic growth rate was defined as the change in diameter divided by baseline diameter and standardized to scan interval. Results32 patients met inclusion criteria (age: 56.9±14.1 years, Female: 13, n=19 rTAAD, n=13 dnTBAD). Mean follow up time was 538 days (range: 135-1689). Baseline aortic diameter did not correlate with growth rate. In the entire cohort, Δ FL MV (rho=0.37, p=.04) and Δ FL RF (rho=0.45, p=0.01) correlated with growth rate. In rTAAD only, Δ FL MV (rho=0.48, p=.04) and Δ FL RF (rho=0.51, p=0.03) correlated with growth rate, while in dnTBAD only, Δ TL KE (rho=0.63, p=.02) and Δ TL MV (rho=0.69, p=.01) correlated with growth rate. Conclusions4D flow derived longitudinal hemodynamic changes correlate with aortic growth rate in TBAD and may provide additional prognostic value for risk stratification. 4D flow MRI could be integrated into existing imaging protocols to allow for identification of TBAD patients who would benefit from preemptive surgical or endovascular intervention.

In vivo and dosimetric investigation on electrical vestibular stimulation with frequency- and amplitude-modulated currents

Objective. Normal function of the vestibular system can be disturbed using a noninvasive technique called electrical vestibular stimulation (EVS), which alters a person's sense of balance and causes false sensations of movement. EVS has been widely used to study the function of the vestibular system, and it has recently gained interest as a therapeutic tool to improve postural stability and help those suffering from vestibular dysfunction. Yet, understanding of how EVS stimulates the vestibular system, the current intensity needed to produce an effect and the frequencies at which it occurs have remained unclear.Approach. The effect of EVS on postural sway was examined in five participants using sinusoidal alternating current with time-varying amplitude from 0 to 1.5 mA and frequency from 0.1 to 10 Hz for three electrode configurations. Dosimetry of the current flow inside the head was conducted using anatomically realistic computational models created individually for each subject based on magnetic resonance imaging data. An estimate for the minimal field strength capable of affecting the vestibular system was calculated with the finite element method.Main results. Bipolar EVS at frequencies up to 10 Hz caused harmonic full-body swaying, and the frequency of the sway was the same as that of the stimulation current. The size of the sway was amplified by increasing the current intensity. Dosimetry modeling indicated that, for 0.2 mA current, the average electric field strength in the vestibular system was approximately 10-30 mV m-1, depending on the electrode montage. The size of the measured postural sway was proportional to the montage-specific electric field strength in the vestibular system.Significance. The results provide insight to EVS's working mechanisms and improve its potential as a tool to study the sense of balance.

Fusing Multi-Scale Functional Connectivity Patterns via Multi-Branch Vision Transformer (MB-ViT) for Macaque Brain Age Prediction

Brain age (BA) is defined as a measure of brain maturity and could help characterize both the typical brain development and neuropsychiatric disorders in mammals. Various biological phenotypes have been successfully applied to predict BA of human using chronological age (CA) as label. However, whether the BA of macaque, one of the most important animal models, can also be reliably predicted is largely unknown. To address this question, we propose a novel deep learning model called Multi-Branch Vision Transformer (MB-ViT) to fuse multi-scale (i.e., from coarse-grained to fine-grained) brain functional connectivity (FC) patterns derived from resting state functional magnetic resonance imaging (rs-fMRI) data to predict BA of macaques. The discriminative functional connections and the related brain regions contributing to the prediction are further identified based on Gradient-weighted Class Activation Mapping (Grad-CAM) method. Our proposed model successfully predicts BA of 450 normal rhesus macaques from the publicly available PRIMatE Data Exchange (PRIME-DE) dataset with lower mean absolute error (MAE) and mean square error (MSE) as well as higher Pearson's correlation coefficient (PCC) and coefficient of determination (R2) compared to other baseline models. The correlation between the predicted BA and CA reaches as high as 0.82 of our proposed method. Furthermore, our analysis reveals that the functional connections predominantly contributing to the prediction results are situated in the primary motor cortex (M1), visual cortex, area v23 in the posterior cingulate cortex, and dysgranular temporal pole. In summary, our proposed deep learning model provides an effective tool to accurately predict BA of primates (macaque in this study), and lays a solid foundation for future studies of age-related brain diseases in those animal models.

Deep Learning Reconstruction of Accelerated MRI: False-Positive Cartilage Delamination Inserted in MRI Arthrography Under Traction.

The radiological imaging industry is developing and starting to offer a range of novel artificial intelligence software solutions for clinical radiology. Deep learning reconstruction of magnetic resonance imaging data seems to allow for the acceleration and undersampling of imaging data. Resulting reduced acquisition times would lead to greater machine utility and to greater cost-efficiency of machine operations. Our case shows images from magnetic resonance arthrography under traction of the right hip joint from a 30-year-old, otherwise healthy, male patient. The undersampled image data when reconstructed by a deep learning tool can contain false-positive cartilage delamination and false-positive diffuse cartilage defects. In the future, precision of this novel technology will have to be put to thorough testing. Bias of systems, in particular created by the choice of training data, will have to be part of those assessments.

Unique versus shared neural correlates of externalizing psychopathology in late childhood.

Childhood externalizing psychopathology is heterogeneous. Symptom variability in conduct disorder (CD), oppositional defiant disorder (ODD), attention-deficit/hyperactivity disorder (ADHD), and callous-unemotional (CU) traits designate different subgroups of children with externalizing problems who have specific treatment needs. However, CD, ODD, ADHD, and CU traits are highly comorbid. Studies need to generate insights into shared versus unique risk mechanisms, including through the use of functional magnetic resonance imaging (fMRI). In this study, we tested whether symptoms of CD, ODD, ADHD, and CU traits were best represented within a bifactor framework, simultaneously modeling shared (i.e., general externalizing problems) and unique (i.e., symptom-specific) variance, or through a four-correlated factor or second-order factor model. Participants (N = 11,878, age, M = 9 years) were from the Adolescent Brain and Cognitive Development Study. We used questionnaire and functional magnetic resonance imaging data (emotional N-back task) from the baseline assessment. A bifactor model specifying a general externalizing and specific CD, ODD, ADHD, and CU traits factors demonstrated the best fit. The four-correlated and second-order factor models both fit the data well and were retained for analyses. Across models, reduced right amygdala activity to fearful faces was associated with more general externalizing problems and reduced dorsolateral prefrontal cortex activity to fearful faces was associated with higher CU traits. ADHD scores were related to greater right nucleus accumbens activation to fearful and happy faces. Results give insights into risk mechanisms underlying comorbidity and heterogeneity within externalizing psychopathology. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

P-based referencing for correcting tissue artifacts in laser ablation-inductively coupled plasma-mass spectrometry imaging of cancer samples.

A referencing strategy based on the element P is presented to compensate for cryosectioning tissue artifacts in laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) data. The study examines how the gadolinium-based contrast agent Gadofosveset is distributed in murine cancer tissue, and illustrates how referenced images can compensate for tissue artifacts like folds, overlaps, and density variations. Compared to non-referenced images that provide information on the absolute distribution of the analyte, referenced images allow for the representation of the analyte distribution relative to the amount of material introduced into the instrument, which in this case is correlated to the P signal. Tissue artifacts were corrected in referenced images for both Gadofosveset and endogenous elements, such as Fe and Zn. Additionally, the referencing approach provides valuable information on the Gd uptake relative to the tissue density in necrotic compared to vital tumor areas, which is not obtained from in vivo magnetic resonance imaging (MRI) data. However, validation of in vivo MRI and ex vivo LA-ICP-MS methods was possible by establishing a mean ratio of necrotic to vital tumor areas in the T1-weighted image post Gadofosveset injection and the non-referenced LA-ICP-MS image of Gd. In summary, P-based correction of LA-ICP-MS imaging data allows for a more accurate spatial representation of certain elements, including endogenous and exogenous elements such as injected contrast agents.

Computer-aided prognosis of tuberculous meningitis combining imaging and non-imaging data

Tuberculous meningitis (TBM) is the most lethal form of tuberculosis. Clinical features, such as coma, can predict death, but they are insufficient for the accurate prognosis of other outcomes, especially when impacted by co-morbidities such as HIV infection. Brain magnetic resonance imaging (MRI) characterises the extent and severity of disease and may enable more accurate prediction of complications and poor outcomes. We analysed clinical and brain MRI data from a prospective longitudinal study of 216 adults with TBM; 73 (34%) were HIV-positive, a factor highly correlated with mortality. We implemented an end-to-end framework to model clinical and imaging features to predict disease progression. Our model used state-of-the-art machine learning models for automatic imaging feature encoding, and time-series models for forecasting, to predict TBM progression. The proposed approach is designed to be robust to missing data via a novel tailored model optimisation framework. Our model achieved a 60% balanced accuracy in predicting the prognosis of TBM patients over the six different classes. HIV status did not alter the performance of the models. Furthermore, our approach identified brain morphological lesions caused by TBM in both HIV and non-HIV-infected, associating lesions to the disease staging with an overall accuracy of 96%. These results suggest that the lesions caused by TBM are analogous in both populations, regardless of the severity of the disease. Lastly, our models correctly identified changes in disease symptomatology and severity in 80% of the cases. Our approach is the first attempt at predicting the prognosis of TBM by combining imaging and clinical data, via a machine learning model. The approach has the potential to accurately predict disease progression and enable timely clinical intervention.

Dimensions of early life adversity are differentially associated with patterns of delayed and accelerated brain maturation

BackgroundDifferent types of early-life adversity have been associated with children’s brain structure and function. However, understanding the disparate influence of distinct adversity exposures on the developing brain remains a major challenge. MethodsThis study investigates the neural correlates of 10 robust dimensions of early-life adversity identified through exploratory factor analysis in a large community sample of youth from the Adolescent Brain Cognitive Development (ABCD) Study. Brain age models were trained, validated, and tested separately on T1-weighted (T1; N = 9524), diffusion tensor (DTI; N = 8834), and resting-state functional (rs-fMRI; N = 8233) magnetic resonance imaging (MRI) data from two time points (mean age = 10.7 years, SD = 1.2, range = 8.9-13.8 years). ResultsBayesian multilevel modelling supported distinct associations between different types of early-life adversity exposures and younger- and older-looking brains. Dimensions generally related to emotional neglect, such as lack of primary and secondary caregiver support, and lack of caregiver supervision, were associated with lower brain age gaps (BAGs), i.e., younger-looking brains. In contrast, dimensions generally related to caregiver psychopathology, trauma exposure, family aggression, substance use and separation from biological parent, and socio-economic disadvantage and neighbourhood safety were associated with higher BAGs, i.e., older-looking brains. ConclusionsThe findings suggest that dimensions of early-life adversity are differentially associated with distinct neurodevelopmental patterns, indicative of dimension-specific delayed and accelerated brain maturation.

Deep Learning-Based Techniques in Glioma Brain Tumor Segmentation Using Multi-Parametric MRI: A Review on Clinical Applications and Future Outlooks.

This comprehensive review explores the role of deep learning (DL) in glioma segmentation using multiparametric magnetic resonance imaging (MRI) data. The study surveys advanced techniques such as multiparametric MRI for capturing the complex nature of gliomas. It delves into the integration of DL with MRI, focusing on convolutional neural networks (CNNs) and their remarkable capabilities in tumor segmentation. Clinical applications of DL-based segmentation are highlighted, including treatment planning, monitoring treatment response, and distinguishing between tumor progression and pseudo-progression. Furthermore, the review examines the evolution of DL-based segmentation studies, from early CNN models to recent advancements such as attention mechanisms and transformer models. Challenges in data quality, gradient vanishing, and model interpretability are discussed. The review concludes with insights into future research directions, emphasizing the importance of addressing tumor heterogeneity, integrating genomic data, and ensuring responsible deployment of DL-driven healthcare technologies. EVIDENCE LEVEL: N/A TECHNICAL EFFICACY: Stage 2.

Structure-function coupling reveals the brain hierarchical structure dysfunction in Alzheimer's disease: A multicenter study.

Alzheimer's disease (AD) is a neurodegenerative condition characterized by cognitive decline. To date, the specific dysfunction in the brain's hierarchical structure in AD remains unclear. We introduced the structural decoupling index (SDI), based on a multi-site data set comprising functional and diffusion-weighted magnetic resonance imaging data from 793 subjects, to assess their brain hierarchy. Compared to normal controls (NCs), individuals with AD exhibited increased SDI within the posterior superior temporal sulcus, insular gyrus, precuneus, hippocampus, amygdala, postcentral gyrus, and cingulate gyrus; meanwhile, the patients with AD demonstrated decreased SDI in the frontal lobe. The SDI in those regions also showed a significant correlation with cognitive ability. Moreover, the SDI was a robust AD neuroimaging biomarker capable of accurately distinguishing diagnostic status (area under the curve [AUC]=0.86). Our findings revealed the dysfunction of the brain's hierarchical structure in AD. Furthermore, the SDI could serve as a promising neuroimaging biomarker for AD. This study utilized multi-center, multi-modal data from East Asian populations. We found an increased spatial gradient of the structure decoupling index (SDI) from sensory-motor to higher-order cognitive regions. Changes in SDI are associated with energy metabolism and mitochondria. SDI can identify Alzheimer's disease (AD) and further uncover the disease mechanisms of AD.

Identify Potential Causal Relationships Between Cortical Thickness, Mismatch Negativity, Neurocognition, and Psychosocial Functioning in Drug-Naïve First-Episode Psychosis Patients.

Cortical thickness (CT) alterations, mismatch negativity (MMN) reductions, and cognitive deficits are robust findings in first-episode psychosis (FEP). However, most studies focused on medicated patients, leaving gaps in our understanding of the interrelationships between CT, MMN, neurocognition, and psychosocial functioning in unmedicated FEP. This study aimed to employ multiple mediation analysis to investigate potential pathways among these variables in unmedicated drug-naïve FEP. We enrolled 28 drug-naïve FEP and 34 age and sex-matched healthy controls. Clinical symptoms, neurocognition, psychosocial functioning, auditory duration MMN, and T1 structural magnetic resonance imaging data were collected. We measured CT in the superior temporal gyrus (STG), a primary MMN-generating region. We found a significant negative correlation between MMN amplitude and bilateral CT of STG (CT_STG) in FEP (left: r = -.709, P < .001; right: r = -.612, P = .008). Multiple mediation models revealed that a thinner left STG cortex affected functioning through both direct (24.66%) and indirect effects (75.34%). In contrast, the effects of the right CT_STG on functioning were mainly mediated through MMN and neurocognitive pathways. Bilateral CT_STG showed significant association with MMN, and MMN plays a mediating role between CT and cognition. Both MMN alone and its interaction with cognition mediated the effects of structural alterations on psychosocial function. The decline in overall function in FEP may stem from decreased CT_STG, leading to subsequent MMN deficits and neurocognitive dysfunction. These findings underline the crucial role of MMN in elucidating how subtle structural alterations can impact neurocognition and psychosocial function in FEP.