搜索结果：Magnetic resonance imaging

Magnetic Resonance Imaging (MRI) is a non-invasive imaging method that can give detailed visualization of the cardiac structures and blood flow, which is effective in diagnosis of cardiovascular diseases (CVDs). It has been proposed that the combination of deep learning (DL) with MRI has an improved ability to automatically identify cardiovascular anomalies by identifying intricate patterns in large-scale imaging data. In this study, an Automated Cardiovascular Disease Detection framework (ACVD-RDODL) is proposed, which combines deep learning with the Red Deer Optimiser (RDO). After image enhancement methods like Wiener Filtering (WF) and Dynamic Histogram Equalization (DHE), features are extracted using radiomics. An Attention-Based Convolutional Gated Recurrent Unit (ACGRU) network is considered to ensure proper classification and RDO is used to optimize the hyperparameters and improve the performance of the progress model. As experimental testing of a benchmark cardiac MRI dataset shows, the proposed method is greater to the existing approaches in terms of classification accuracy and computational efficiency.

Large poly(ethylene) glycol (PEG) chains are often conjugated to proteins or biomolecules to inhibit proteolytic degradation, mask immunogenic response, reduce clearance rates, and improve biodistribution of therapeutics, vaccines, drug delivery systems, and gene therapy formulations. The PEG macromolecular chain can also be used as a noninvasive reporter to track biologics in vivo by magnetic resonance spectroscopy (MRS). Rapid internal dynamics of PEG render the transverse 1H spin relaxation time to be comparable to water (~0.5 s) and amenable to imaging through traditional pulsed field gradient techniques. While water signal grossly exceeds that of PEG it is possible to filter 1H MRS signal of PEGylated conjugates through one of two ways-(1) stimulated echo acquisition mode (STEAM) MRS, which leverages huge differences in the diffusion of water versus PEGylated constructs, and (2) 13C-edited 1H MRS of fully 13C-enriched PEGylated constructs. Here, we compare both approaches. A 15 kDa 13C-enriched PEG chain was prepared alone, conjugated to bovine serum-albumin (BSA), and incorporated into a 52-nm-diameter PEG-poly(lactic acid) (PLA) nanoparticle. These three PEG constructs were then separately monitored in real time by 13C-edited 1H MRS, after introducing them into rat animal models intravenously. A 13C-editing scheme was employed to monitor 1H MRS PEG signal in the vasculature via a radiofrequency coil placed around the tail. An observed two-component decay of the PEG signal is attributed to perfusion and early equilibration (alpha phase) and slow clearance (beta phase). 13C-PEG alone, 13C-PEG-BSA, and 13C-PEG-PLA nanoparticles exhibited half-lives of 38.6 min, 23.4 h, and 11.9 h, respectively. The relatively rapid clearance rates associated with the PEG-PLA nanoparticles is expected to arise from enzymatic degradation of the PLA chain. Using STEAM-based editing schemes, we then evaluated sensitivity and water suppression in diffusion-edited 1H MRS for (12C)-PEGylated BSA contrasting 2-, 20-, and 40-kDa PEG chains, in imaging phantom samples. Larger molecular weight PEG chains (i.e., 40 kDa) proved far superior to smaller PEG chain reporters due to reduced inhomogeneities and longer T2, upon employing either a 13C-HQMC filter or a STEAM-based diffusion filter.

This work aims to investigate the benefits of incorporating fluid dynamic models into ultrasound vector flow imaging through a novel data assimilation framework using tensor product B-splines for model-based regularization. A variational data assimilation method was developed using tensor product B-splines with the goal of solving high-dimensional regularization problems governed by the Navier-Stokes equations. The method was implemented in an open-source library and validated across three experimental setups: in silico using a computational fluid dynamics phantom, in vitro using a pulsatile flow phantom with particle imaging velocimetry and in vivo using 4-D ultrasound compared with magnetic resonance imaging. The proposed method outperformed conventional smoothing techniques and matched the performance of state-of-the-art regularization approaches. In silico tests showed improved noise suppression and lower root mean squared error. In vitro experiments demonstrated accurate reconstruction of flow features and gradient-based metrics. In vivo comparisons revealed good agreement with magnetic resonance imaging in high-velocity regions and successful reconstruction in dropout zones. The data assimilation approach using B-splines and fluid dynamic constraints enables efficient and accurate reconstruction of 4-D flow fields in ultrasound vector flow imaging. It offers a promising solution for bedside clinical applications, balancing noise suppression and resolution while leveraging physical models for robust flow estimation.

The purpose of this study was to assess abdominal imaging utilization across a large healthcare system over a 10-year period to compare growth rates and relative distribution of exams among patient classes (emergency department [ED] patients, inpatients [IP], and outpatients [OP]) and across cross-sectional imaging modalities. Data from all abdominal computed tomography (CT), ultrasound (US), and magnetic resonance imaging (MRI) examinations performed in ED patients, inpatients, and outpatients were collected from a large healthcare system over a 10-year period (January 2014 through December 2023). Poisson regression analysis was used to model the number of examinations over time. Additional models were constructed to assess trends over time for each modality and for each patient class. Over the study period, 3,408,471 imaging studies were performed in 1,150,921 patients. ED volumes showed the largest average increase, 15.9% per year which was significantly greater than the 14.6% average annual increase for inpatients and 10.5% for outpatients. The average annual increases by modality were 14.5% for MRI, 14.4% for CT, and 10.7% for US. The percentage of acute care cases (IP and ED) relative to all exams significantly increased from 41.7% in 2014 to 52% in 2023. This study demonstrated a pattern of near-continuous annual growth in abdominal imaging examinations across all patient classes and major modalities over a 10-year period, with an increasing contribution of acute care cases. These findings highlight areas of need for resources.

Fistulas represent abnormal communications between two epithelialized surfaces, which can involve organs, vessels, or the skin. While common fistulas are frequently encountered, uncommon fistulas in the thoracic and abdominal cavities pose significant diagnostic challenges due to their varied etiologies and presentations. Imaging plays a pivotal role in the detection, characterization, and management planning of these complex entities. This pictorial review will explore the spectrum of uncommon thoracic and abdominal fistulas, emphasizing the crucial contributions of various imaging modalities. We will discuss etiological factors, characteristic imaging findings, and specific techniques that aid in accurate diagnosis, including illustrative cases. Keywords: Fistula, Digestive System Fistula, Respiratory Tract Fistula, Vascular Fistula, Diagnostic Imaging,  Magnetic Resonance Imaging, Radiography, Thoracic Diseas.

Multifunctional theranostic nanoplatforms enabling precise targeting and controlled deep tissue therapy are vital. Herein, we present a self-assembled, defect-engineered, biomimetic nanoplatform (CDMBBTO), in which piezoelectric material and downconversion nanoparticles (DCNPs) are co-assembled within a polymer matrix, preventing interactions between both components, thereby allowing each to retain its intrinsic properties while collectively enabling dual second near infrared fluorescence (NIR-II FL)/magnetic resonance (MR) imaging and synergistic piezo/chemodynamic therapy (PZDT/CDT). Mn/Bi codoped piezoelectric BaTiO3 (MBBTO) NPs exhibit a significantly enhanced piezoelectric coefficient (~5 fold higher than pristine BTO) owing to bandgap modulation induced by defect engineering. The incorporated Mn2+ not only catalyzes Fenton like reactions for sustained tumor suppression but also provides strong T2 weighted MR contrast. Self-assembled NIR-II emissive DCNPs enable deep tissue optical imaging, forming an integrated theranostic system. Cloaking of U87 glioma cell membranes imparts homologous tumor targeting capability. In vivo dual modal imaging reveals efficient tumor accumulation, with peak NIR-II FL and MR signal intensities observed 6 h post-injection. Upon ultrasound activation, CDMBBTO elicits potent tumor ablation through synergistic reactive oxygen species generation and immune modulation, characterized by macrophage polarization (M2 → M1) and upregulation of proinflammatory cytokines (TNF-α and IL-6). This work establishes CDMBBTO as a powerful nanoplatform for dual modal imaging-guided, immune potentiated PZDT/CDT toward effective glioblastoma treatment.

Acute sport-related avulsion fractures and osteochondral injuries are common in athletes, particularly in the pediatric population, and may present diagnostic challenges when radiographic findings are subtle. These injuries occur because of excessive tensile, shear, or impaction forces and most frequently involve apophyses, osteochondral junctions, or articular surfaces. Accurate characterization of injury location, fragment morphology, displacement, and associated soft-tissue abnormalities is essential to guide management, promote healing, and facilitate safe return to play (RTP). Radiographs are the initial imaging modality for evaluation and are useful for identifying osseous avulsion fragments and assessing alignment. Computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound (US) may serve as complementary modalities depending upon the suspected injury. Common injury sites include the shoulder, elbow, pelvis, knee, foot, and ankle. Recognition of characteristic imaging patterns, associated injuries, and potential mimics, particularly in skeletally immature patients, is critical for accurate diagnosis. Management ranges from conservative therapy to surgical fixation or cartilage restoration procedures. This review summarizes the mechanisms, imaging features, treatments, pitfalls, and RTP implications when available of acute sport-related avulsion fractures and osteochondral injuries.

Alterations in knee joint biomechanics have been observed in individuals with recurrent lateral ankle sprain, potentially affecting knee joint health. However, it remains unclear whether femoral cartilage morphology and composition are associated with recurrent lateral ankle sprain. We aimed to compare femoral cartilage cross-sectional area and T2 relaxation times using magnetic resonance imaging between individuals with and without recurrent lateral ankle sprains. In this single-blinded, case-control study conducted at a university imaging laboratory, 15 male participants with recurrent lateral ankle sprains (age = 21.3 ± 1.8 years, height = 177.2 ± 5.1 cm, body mass = 68.7 ± 5.8 kg) and 15 healthy controls (age = 21.3 ± 1.9 years, height = 173.7 ± 6.6 cm, body mass = 64.6 ± 5.7 kg) were included. Bilateral femoral articular cartilage images were acquired using T2 mapping compositional magnetic resonance scans of the tibiofemoral joint. A single reader manually segmented the medial cartilage on a sagittal image slice to evaluate the cross-sectional area and T2 relaxation time. A two-way repeated measures analysis of variance with Bonferroni post hoc analyses was used to analyze the magnetic resonance variables. Hedge g effect sizes using the pooled standard deviations were calculated, along with 95% confidence intervals (CIs) for each pairwise comparison. There were no differences in anthropometric characteristics between individuals with and without recurrent lateral ankle sprain (P > .05). Participants with recurrent lateral ankle sprains exhibited a significantly greater femoral cartilage cross-sectional area than healthy controls, regardless of limb dominance (P = .03, η2 = 0.16). In addition, T2 relaxation times were significantly longer in the dominant limb of participants with recurrent lateral ankle sprains than in healthy controls (P = .02, Hedge g = 0.88, 95% CIs: 0.14-1.61) and in their nondominant limb (P = .03, Hedge g = 0.72, 95% CIs: 0.16-1.25). No significant differences were observed between the dominant and nondominant limbs in the control groups (P = .23, Hedge g = -0.26, 95% CIs: -0.74 to 0.24) or between the recurrent lateral ankle sprain and control groups in the nondominant limb (P = .74, Hedge g = -0.12, 95% CIs: -0.82 to 0.61). These findings indicate altered femoral articular cartilage morphology and composition in participants with recurrent lateral ankle sprains. The findings of this study support the need for monitoring knee joint function and symptoms during rehabilitation in patients with recurrent lateral ankle sprains. A prospective longitudinal study with a larger and more diverse sample is needed to investigate these associations further.

The brain tumors possess different causative factors and properties, making their diagnosis and treatment difficult. Growth of these cancers usually leads to compression of the adjacent nerves and obstruction of the flow of cerebrospinal fluid, thus leading to increase in intracranial pressure. This affects the working of brain in many ways; thus, the difficulty involved in its treatment. With the improvements in technology in neuroimaging, including Diffusion Tensor Imaging (DTI), Positron Emission Tomography (PET), and multiparametric Magnetic Resonance Imaging (mpMRI), the diagnosis process has become easy. The effectiveness of any form of therapy in such patients depends primarily on their prognosis. While it is a common practice that physicians determine the prognosis of the disease by considering the age of the patient, histological grade of the tumor, and resection status, now this method has become more comprehensive by adding molecular signature and genetic analyses to the list of criteria. Next-generation sequencing (NGS) allows a reliable molecular classification. It increases the level of risk stratification, facilitating the application of therapies tailored to individual patients. Thus, molecular oncology has greatly changed our views on brain tumors' pathology and prognosis while neoadjuvant treatments aim at increasing the survival rate. On the other hand, radiogenomics is a field of study that combines non-invasive imaging phenotypes and genomic information in order to find unique molecular signatures of tumors without collecting samples from tumors. Molecular biomarkers are absolutely essential in the diagnosis of cancer, treatment monitoring, and recurrence of cancer. Advances in liquid biopsy technology, particularly the methods for circulating tumor DNA (ctDNA) and Extracellular Vesicle (EV) based analysis, have enabled the possibility of non-invasive monitoring of the progression of the tumors over time. This review highlights key studies and important scientific works about imaging technologies, biomarkers, and prognostic factors of malignant brain tumors.

Cardiac complications are among the most common causes of death in patients after pediatric kidney transplantation (KTx), but defined diagnostic procedures identifying young patients at risk are not established. Cardiovascular magnetic resonance (CMR) imaging with native T1 mapping allows detection of diffuse myocardial alterations but is not routinely available for cardiovascular screening. Whether abnormalities detected by echocardiography reflect underlying myocardial structural changes remains unclear. Pediatric KTx recipients underwent comprehensive transthoracic echocardiography and CMR imaging with native T1 mapping. Associations between echocardiographic measures and T1 values were analyzed using multivariable linear regressions. Receiver operating characteristics analyses assessed the ability of septal E/e' to identify elevated T1 values, with area under the curve (AUC) and optimal cut-offs determined using positive likelihood ratios (LR +). Forty-six pediatric KTx recipients (16 ± 3.5 years old; time since KTx 7.9 ± 5.3 years) were included. Diastolic echocardiographic abnormalities were common, with 87% exhibiting at least one abnormal diastolic parameter. Septal T1 was associated with septal E/e', A-wave, and pulmonary venous atrial reversal, while lateral T1 was associated only with septal E/e'. Optimal septal E/e' cut-offs were 10.550 for detecting an elevated septal T1 (LR +  = 7.143) and 10.630 for detecting an elevated lateral T1 (LR +  = 9). Pediatric KTx recipients with structural myocardial alterations on CMR imaging exhibit detectable abnormalities in routine echocardiographic diastolic parameters. Especially a markedly elevated septal E/e' could identify patients at increased risk for underlying myocardial involvement and justify the use of CMR imaging in post-transplant follow-up.

Women carrying the apolipoprotein E4 (APOE4) allele have a greater risk of Alzheimer's disease (AD) from ages 65-75 years compared to men, yet the effects of APOE4 on cognitive and neuroimaging outcomes among midlife women remain poorly understood. We investigated APOE4-related differences in memory-based functional neuroimaging outcomes in cognitively normal, midlife postmenopausal women. We measured blood-oxygen-level-dependent activation and hippocampal functional connectivity during a functional magnetic resonance imaging verbal encoding task. Linear regression models tested APOE4 differences (carriers vs. non-carriers) and associations of neuroimaging indices with verbal memory measures and plasma AD biomarkers, adjusting for age, race, and education. Among 145 women from MsBrain, APOE4 carriers and non-carriers did not differ in verbal memory performance or AD biomarker levels. During verbal encoding, APOE4 carriers had significantly decreased activation and hippocampal functional connectivity in several regions compared to non-carriers. APOE4-related functional brain differences are present by midlife in postmenopausal women.

The cerebellum contributes to sensory, cognitive, emotional, and pain-modulatory processes beyond motor coordination. In temporomandibular disorders (TMD), alterations in spontaneous cerebellar neural activity and strengthened functional connectivity with limbic structures have been reported, suggesting involvement of central neuroplastic mechanisms. Mandibular displacement and occlusal imbalance may further modulate brain networks related to motor regulation, sensory integration, and pain-related emotional processing.In this study, magnetic resonance imaging(MRI) and diffusion tensor imaging(DTI) was used to examine cerebellar connectivity across different mandibular positions in a patient with painful TMD. Functional MRI and diffusion tensor imaging were acquired in three conditions: centric occlusion, centric relation, and stabilization splint. Cerebellar regions of interest were defined using SUIT parcellation, and position-dependent structural-functional connectivity was assessed using diffusion-based tractography and Δ-metrics computed as the difference between the two tractography-derived matrices (Δ = MNI_NORM - COUNT_NORM); scale-invariant similarity metrics were used for between-condition comparisons.Under centric occlusion (CO), the Δ-derived connectivity pattern was sparse and included ROIs with zero-variance profiles, yielding mathematically undefined correlation entries; these were treated as missing values. In contrast, centric relation and stabilization splint conditions demonstrated heterogeneous Δ patterns and structured ROI-wise correlation profiles.In contrast, both centric relation and stabilization splint conditions showed moderate effect sizes in summary metrics. Pairwise comparison of vectorized Δ matrices (upper triangle; N = 378 ROI pairs) demonstrated a moderate association between CR and splint (Pearson r = 0.452, 95% CI 0.368-0.527; p = 2.0 × 10⁻²⁰; cosine similarity = 0.428). The Euclidean distance between Δ vectors was 1554.5, corresponding to a normalized per-element RMS difference of 81.1, and the mean absolute difference was 42.2.These single-subject findings suggest that mandibular position may be associated with measurable differences in Cerebellum-SUIT connectivity profiles in painful TMD. The results should be interpreted as exploratory and hypothesis-generating, and they require confirmation in larger prospective studies before clinical or therapeutic implications can be inferred.

The Ovarian-Adnexal Reporting and Data System (O-RADS), developed by the American College of Radiology (ACR), provides a standardized, evidence-based framework for risk stratification and management of ovarian and adnexal lesions. The O-RADS system is comprised of two complementary imaging arms: ultrasound (US) and magnetic resonance imaging (MRI). O-RADS US serves as the primary imaging modality for adnexal lesion assessment, given its wide availability in outpatient and inpatient settings, relatively low cost, lack of ionizing radiation and the ability for real-time dynamic assessment. Using defined lexicon terms, lesions are stratified into five risk categories corresponding to an estimated malignancy risk with associated management recommendations. Validation studies have demonstrated high sensitivities for malignancy detection, good specificities, and strong interobserver agreement. The 2022 update refined selected descriptors to improve specificity while otherwise maintaining diagnostic performance and reproducibility, reinforcing the system's clinical utility. O-RADS MRI functions as a secondary, problem-solving tool for lesions that are technically limited or when MRI evaluation may add value. Validation studies have consistently shown excellent performance of O-RADS MRI with studies showing high specificities approaching 90-95% related to better characterization of solid tissue based on signal and dynamic enhancement characteristics. Together, O-RADS US and MRI establish a comprehensive, standardized approach that improves diagnostic confidence and optimizes patient management. Ongoing efforts focus on improving congruence between these two arms by addressing discrepancies, understanding and lowering barriers for adoption, and providing management recommendations for O-RADS MRI. Emerging evidence and evolving technologies, such as contrast-enhanced ultrasound (CEUS), quantitative imaging biomarkers and artificial intelligence (AI), may further advance the system. This review summarizes the background and validation of O-RADS US and O-RADS MRI, discusses their complementary roles, current challenges and barriers to adoption, and outlines future priorities and research directions.

This study aimed to develop a dynamic nomogram in which clinical indicators are integrated with magnetic resonance imaging (MRI) radiomics to predict axillary lymph node metastasis (ALNM) in patients with breast cancer. Our retrospective study included 221 pathologically confirmed patients with breast cancer. Radiomic features were extracted from dynamic contrast-enhanced MRI (DCE-MRI) and fat-suppressed T2-weighted imaging (FS-T2WI) datasets. After feature screening, a support vector machine (SVM) algorithm was employed to establish radiomic models and calculate radiomic scores. Clinical independent predictors were identified through univariate and multivariate logistic regression analyses. A nomogram was established on the basis of the radiomic scores obtained with the optimal SVM model and clinically independent predictors, and it was subsequently transformed into a dynamic nomogram. Model performance was evaluated by the receiver operating characteristic curve and area under the curve (AUC). Shapley additive explanation (SHAP) was applied to interpret the contribution of clinical predictors. Calibration and decision curves were employed to assess nomogram performance. The platelet-to-lymphocyte ratio (PLR), breast imaging reporting and data system (BI-RADS) classification, and ALN status on MRI were identified as independent predictors of ALNM (all p&#x2009;<&#x2009;0.05). SHAP analysis identified PLR as the top contributor. The clinical model developed with the three predictors achieved AUCs of 0.845 and 0.706 in the training and validation cohorts, respectively. Among the SVM models, the DCE-MRI and FS-T2WI fusion sequence model outperformed the single-sequence models, with AUCs of 0.868 and 0.875, respectively, in the training and validation cohorts. When clinical predictors were incorporated into the fusion sequence model, the nomogram achieved AUC values of 0.930 and 0.928 in the training and validation cohorts. Decision curve analysis demonstrated that the nomogram has significant clinical value. A nomogram model integrating MRI-derived radiomic scores, BI-RADS classification, ALN status, and PLR demonstrated good performance in predicting ALNM in patients with breast cancer.

Cyclops syndrome remains a clinically relevant cause of extension deficit after anterior cruciate ligament reconstruction (ACLR), yet its prevention, diagnosis, and management are not standardized. The purpose of this study was to characterize practice patterns, identify areas of consensus and controversy, and derive a clinically applicable management framework based on an international survey of knee surgeons. A 69-item online survey was distributed to orthopaedic surgeons specialized in knee surgery through multiple international societies. The questionnaire explored five domains: surgeon characteristics, surgical techniques, preventive strategies, diagnostic approaches, and management of cyclops syndrome. Descriptive statistics were used to summarize responses. A total of 250 surgeons from 25 countries completed the survey. Most respondents estimated the incidence of cyclops syndrome between 1% and 5%, with symptomatic presentation typically occurring within 3-6 months after ACLR. Loss of extension greater than 5&#xb0; was consistently identified as the key clinical finding. Magnetic resonance imaging was the preferred first-line diagnostic modality. Despite heterogeneity in several aspects of care, consistent patterns emerged. Accurate tibial tunnel positioning and restoration of full preoperative knee extension were considered the most critical preventive factors. Initial management was predominantly conservative, with most surgeons advocating an extension-focused rehabilitation trial before surgical intervention. Arthroscopic excision was generally reserved for persistent extension deficits and was associated with favorable outcomes and low recurrence rates. However, substantial variability persisted regarding the role of remnant preservation, timing of surgery, and indications for imaging, highlighting unresolved clinical controversies. While substantial variability exists in the management of cyclops syndrome after ACLR, consistent clinical patterns can be identified. Based on these findings, a stepwise management approach can be proposed, emphasizing early recognition of extension deficit, selective use of imaging, and a staged treatment strategy. These results provide a foundation for future efforts toward consensus guidelines and standardized care pathways. Level V, cross-sectional survey study.

To describe a self-limited presentation of pediatric orbital myositis and discuss its implications for management. A previously healthy 9-year-old boy presented with acute painful diplopia and esotropia following a recent febrile upper respiratory tract infection and presumed viral conjunctivitis. Best-corrected visual acuity was 20/40 OD and 20/40 OS. Examination demonstrated an esotropia of 30 prism diopters (PD) at corrected distance and 35 PD at corrected near, bilateral abduction limitation (-&#x2009;1), and localized temporal conjunctival injection over the lateral rectus insertions, with clinical features also suggestive of lacrimal gland involvement. Magnetic resonance imaging (MRI) of the orbits revealed bilateral enlargement and post-contrast enhancement of the lateral rectus muscles with mild tendon involvement and surrounding inflammatory changes, consistent with orbital myositis within the spectrum of idiopathic orbital inflammation (IOI). Dacryoadenitis was also suggested by imaging findings. Clear clinical improvement in pain, diplopia, and conjunctival injection was observed within 24&#xa0;h of presentation, prior to initiation of any medical therapy. Although oral antibiotics and systemic corticosteroids were prescribed, corticosteroid therapy was not initiated due to intolerance. At three-month follow-up, the patient was orthophoric with full restoration of extraocular motility. This case illustrates a self-limited presentation of presumed post-infectious pediatric orbital myositis manifesting as painful acute esotropia, with bilateral lateral rectus muscle involvement and spontaneous recovery without corticosteroid therapy. Careful clinical and imaging assessment is essential to distinguish orbital myositis from neurogenic, accommodative, and infectious causes of acute esotropia in children. The reduced best-corrected visual acuity may reflect underlying refractive amblyopia and/or significant astigmatism and warrants continued monitoring. This report highlights an unusual but recognizable phenotype within the spectrum of pediatric idiopathic orbital inflammation and underscores the importance of comprehensive diagnostic evaluation and longer-term follow-up.

The prognostic value of spontaneous brain activity in patients with Disorders of Consciousness (DoC) remains questionable due to methodological heterogeneity, small sample sizes, and the difficulty in conducting longitudinal studies on this clinical population. We performed a coordinate-based meta-analysis of the studies adopting resting state brain imaging techniques to identify whether the spontaneous activity of specific brain areas has a prognostic value for DoC patients. We included studies published until 2025 providing the peak coordinates deriving from contrasting brain activations of patients showing full consciousness recovery (Good Outcome;GO) and patients showing either no consciousness recovery or death (Bad Outcome; BO) with voxel-wise whole-brain analyses. Twelve studies were included in the meta-analysis containing data from 332 DoC patients (n= 192 BO) in the post-acute phase, assessed through resting state functional Magnetic Resonance Imaging. Compared to BO, GO patients showed increased spontaneous activity in sensory and associative areas, including visual areas, precuneus, temporo-parietal, and, marginally, premotor regions. Taken together, the results suggest preservation of posterior brain areas is pivotal in assisting prognosis, with a marginal role played by the premotor areas. However, both low-level sensory and high-level associative areas contribute to outcome prediction of DoC patients, possibly due to the need to integrate low-level sensory information to enable functional interaction with the environment.

Early and accurate detection of brain tumors is clinically valuable for improving prognosis and guiding treatment. Existing deep-learning methods for magnetic resonance imaging (MRI) brain tumor detection face three difficulties: weak texture at lesion boundaries impairs localization; heterogeneous lesion scales degrade multi-scale detection; and non-maximum suppression (NMS) post-processing limits end-to-end inference. We propose a topology-decoupled end-to-end detection framework based on boundary-preserving feature flow and inter-channel correlation (ICC) distillation. A high-capacity teacher combines a multi-gradient-flow backbone with a gather-and-distribute global fusion mechanism, capturing both pathological boundary textures and anatomical context; a lightweight student is then derived by removing the global-fusion neck while retaining the isomorphic backbone. After comparing five feature distillation methods, we adopt ICC distillation, which aligns Gram matrices of intermediate features and mitigates the background-dominated bias common in medical imaging. Across three random seeds, the ICC-distilled student reaches mAP@0.5 = [Formula: see text], surpassing the plain student ([Formula: see text]) and matching or exceeding the teacher ([Formula: see text]). On the BraTS small-lesion stratum it attains 98.7% lesion recall with a false-positive-per-image rate of 0.014. The student achieves this at low cost (6.09M parameters, 11.7 GFLOPs, 168 FPS), suiting resource-constrained clinical deployment.