Search Results (64)

Diagnosing digital flexor tendon sheath (DFTS) pathologies, particularly manica flexoria (MF) tears, can be challenging with standard imaging modalities. Standing low-field MRI tenography (MRIt) may improve the detection rate of MF tears. This study aimed to compare ultrasonography, contrast radiography, pre-contrast MRI, and MRIt to detect naturally occurring MF lesions in horses undergoing tenoscopy. Ten horses with a positive DFTS block, which underwent contrast radiography, ultrasonography, MRI, MRIt, and tenoscopy were included. Two radiologists evaluated the images and recorded whether an MF lesion was present and determined the lesion side. Sensitivity and specificity were calculated for each modality using tenoscopy as a reference. MRIt and contrast radiography detected MF lesions with the same frequency, both showing 71% sensitivity and 100% specificity. Pre-contrast MRI and ultrasonography detected MF lesions with a lower sensitivity (57%); however, the MRI (100%) demonstrated a higher specificity than ultrasonography (33%). Adding contrast in MRI changed the sensitivity from (4/7 lesions) 57% to (5/7 lesions) 71%, with a constant high specificity (100%). MRIt diagnoses MF tears with a similar sensitivity to contrast radiography, with the same specificity, but with the added benefit of lesion laterality detection. The combined advantages of the anatomical detail of the T1 sequence and the post-contrast hyperintense appearance of the fluid may help diagnose MF tears and identify intact MFs. However, this needs to be substantiated in a larger number of cases. Full article

Background/Objectives: Assessment of a novel deep-learning (DL)-based T1w volumetric interpolated breath-hold (VIBE_DL) sequence in breast MRI in comparison with standard VIBE (VIBE_Std) for image quality evaluation. Methods: Prospective study of 52 breast cancer patients examined at 1.5T breast MRI with T1w VIBE_Std and T1 VIBE_DL sequence. T1w VIBE_DL was integrated as an additional early non-contrast and a delayed post-contrast scan. Two radiologists independently scored T1w VIBE _Std/DL sequences both pre- and post-contrast and their calculated subtractions (SUBs) for image quality, sharpness, (motion)–artifacts, perceived signal-to-noise and diagnostic confidence with a Likert-scale from 1: Non-diagnostic to 5: Excellent. Lesion diameter was evaluated on the SUB for T1w VIBE_Std/_DL. All lesions were visually evaluated in T1w VIBE_Std/_DL pre- and post-contrast and their subtractions. Statistics included correlation analyses and paired t-tests. Results: Significantly higher Likert scale values were detected in the pre-contrast T1w VIBE_DL compared to the T1w VIBE_Std for image quality (each p < 0.001), image sharpness (p < 0.001), SNR (p < 0.001), and diagnostic confidence (p < 0.010). Significantly higher values for image quality (p < 0.001 in each case), image sharpness (p < 0.001), SNR (p < 0.001), and artifacts (p < 0.001) were detected in the post-contrast T1w VIBE_DL and in the SUB. SUB_DL provided superior diagnostic certainty compared to SUB_Std in one reader (p = 0.083 or p = 0.004). Conclusions: Deep learning-enhanced T1w VIBE_DL at 1.5T breast MRI offers superior image quality compared to T1w VIBE_Std. Full article

This study aimed to develop and evaluate an AI-driven platform, the Adaptive RAG Assistant MRI Platform (ARAMP), for assisting in the diagnosis and reporting of brain metastases using post-contrast axial T1-weighted (AX_T1+C) MRI. In this retrospective study, 2447 cancer patients who underwent MRI between 2010 and 2022 were screened. A subset of 100 randomized patients with confirmed brain metastases and 100 matched non-cancer controls were selected for evaluation. ARAMP integrates quantitative radiomic feature extraction with an adaptive Retrieval-Augmented Generation (RAG) framework based on a large language model (LLM, GPT-4o), incorporating five authoritative medical references. Three board-certified neuroradiologists and an independent LLM (Gemini 2.0 Pro) assessed ARAMP performance. Metrics of the assessment included Pre-/Post-Trained Inference Difference, Inter-Inference Agreement, and Sensitivity. Post-training, ARAMP achieved a mean Inference Similarity score of 67.45%. Inter-Inference Agreement among radiologists averaged 30.20% (p = 0.01). Sensitivity for brain metastasis detection improved from 0.84 (pre-training) to 0.98 (post-training). ARAMP also showed improved reliability in identifying brain metastases as the primary diagnosis post-RAG integration. This adaptive RAG-based framework may improve diagnostic efficiency and standardization in radiological workflows. Full article

Background: Hereditary immune-mediated polymyositis has been reported in the Kooiker dog breed, associated with a 39 kb deletion and low penetrance. Approximately 10–20 percent of homozygous dogs and 0.5–2 percent of heterozygous dogs develop polymyositis. This study examines whether magnetic resonance imaging (MRI) can assist in diagnosing polymyositis in this breed. Methods: All dogs in this prospective case study were purebred Kooiker dogs referred for clinical examination to assess them for polymyositis. A dataset was compiled, including sex, neuter status, and, if applicable, age of onset, clinical signs, CK activity, electromyogram, and histopathological findings. MRI was performed using a 1.5 Tesla MRI scanner, with T1-weighted, T2-weighted, T2W fat-suppressed short tau inversion recovery (STIR), and T1-weighted post-contrast sequences. Results: Five Kooiker dogs were included in the study. Four dogs exhibited clinical signs compatible with polymyositis (one heterozygous and three homozygous for the 39 kb deletion), while one dog was homozygous for the 39 kb deletion but showed no clinical signs. The clinically affected dogs exhibited T2-weighted, STIR, and T1-weighted post-contrast muscular hyperintensity, and the diagnosis was confirmed with histopathology. The asymptomatic dog displayed no MRI abnormalities. Conclusions: MRI has proven to be a valuable tool in assisting with the diagnosis of Kooiker dogs carrying the 39 kb deletion. MRI can act as a screening tool for dogs with the 39 kb deletion, eliminating the need for an initial biopsy. A muscle biopsy, following a confirmatory MRI, is still the preferred method for diagnosing polymyositis. Full article

Objectives: To assess amide proton transfer weighted (APTw) MR imaging capabilities in differentiating high-grade glial tumors across alpha-thalassemia/mental retardation X-linked (ATRX) expression, tumor-suppressor protein p53 expression (p53), O6-methylguanine-DNA methyltransferase promoter (MGMTp) methylation, isocitrate dehydrogenase (IDH) status, and proliferation marker Ki-67 (Ki-67 index) as a preoperative diagnostic aid. Material & Methods: A total of 42 high-grade glioma WHO grade 4 (HGG) patients were evaluated prospectively (30 males and 12 females). All patients were examined using conventional MRI, including the following: T1w-MPRAGE pre- and post-contrast administration, conventional T2w and 3D FLAIR, and APTw imaging with a 3T MR scanner. Receiver operating characteristic (ROC) curves were calculated for the APTw% mean, median, and max signal for the different molecular biomarkers. A logistic regression model was constructed for combined mean and median APTw% signals for p53 expression. Results: The whole-tumor max APTw% signal could significantly differentiate MGMTp from non-MGMTp HGG, p = 0.035. A cutoff of 4.28% max APTw% signal yielded AUC (area under the curve) = 0.702, with 70.6% sensitivity and 66.7% specificity. The mean/median APTw% signals differed significantly in p53 normal versus p53-overexpressed HGG s: 1.81%/1.83% vs. 1.15%/1.18%, p = 0.002/0.006, respectively. Cutoffs of 1.25%/1.33% for the mean/median APTw% signals yielded AUCs of 0.786/0.757, sensitivities of 76.9%/76.9%, and specificities of 50%/66.2%, p = 0.002/0.006, respectively. A logistic regression model with a combined mean and median APTw% signal for p53 status yielded an AUC = 0.788 and 76.9% sensitivity and 66.2% specificity. ATRX-, IDH- wild type (wt) vs. mutation (mut), and the level of Ki-67 did not differ significantly, but trends were found: IDH-wt and low Ki-67 showed higher mean/median/max APTw% signals vs. IDH-mut and high Ki-67, respectively. ATRX-wt vs. mutation showed higher mean and median APTw% signals but lower max APTw% signal. Conclusions: APTw imaging can potentially be a useful marker for the stratification of p53 expression and MGMT status in high-grade glioma in the preoperative setting and potentially aid surgical decision-making. Full article

Background: T1 mapping has become a valuable technique in cardiac magnetic resonance imaging (CMR) for evaluating myocardial tissue properties. However, its quantitative accuracy remains limited by noise-related variability. Super-resolution deep learning-based reconstruction (SR-DLR) has shown potential in enhancing image quality across various MRI applications, yet its effectiveness in myocardial T1 mapping has not been thoroughly investigated. This study aimed to evaluate the impact of SR-DLR on noise reduction and measurement consistency in myocardial T1 mapping. Methods: This single-center retrospective observational study included 36 patients who underwent CMR between July and December 2023. T1 mapping was performed using a modified Look-Locker inversion recovery (MOLLI) sequence before and after contrast administration. Images were reconstructed with and without SR-DLR using identical scan data. Phantom studies using seven homemade phantoms with different Gd-DOTA dilution ratios were also conducted. Quantitative evaluation included mean T1 values, standard deviation (SD), and coefficient of variation (CV). Intraclass correlation coefficients (ICCs) were calculated to assess inter-observer agreement. Results: SR-DLR had no significant effect on mean native or post-contrast T1 values but significantly reduced SD and CV in both patient and phantom studies. SD decreased from 44.0 to 31.8 ms (native) and 20.0 to 14.1 ms (post-contrast), and CV also improved. ICCs indicated excellent inter-observer reproducibility (native: 0.822; post-contrast: 0.955). Conclusions: SR-DLR effectively reduces measurement variability while preserving T1 accuracy, enhancing the reliability of myocardial T1 mapping in both clinical and research settings. Full article

Hippocampal segmentation is essential in neuroimaging for evaluating conditions like Alzheimer’s dementia and mesial temporal sclerosis, where small volume changes can significantly impact normative percentiles. However, inaccurate segmentation is common due to the inclusion of non-hippocampal structures such as choroid plexus and cerebrospinal fluid (CSF), leading to volumetric overestimation and confounding of functional analyses. Current methods of assessment largely rely on virtual or manual ground truth labels, which can fail to capture these inaccuracies. To address this shortcoming, this study introduces a more direct voxel intensity-based method of segmentation assessment. Using paired precontrast and postcontrast T1-weighted MRIs, hippocampal segmentations were refined by adding marginal gray matter and removing marginal CSF and enhancement to determine a total required correction volume. Six segmentation algorithms—e2dhipseg, HippMapp3r, hippodeep, AssemblyNet, FastSurfer, and QuickNat—were implemented and compared. HippMapp3r and e2dhipseg, followed closely by hippodeep, exhibited the least total correction volumes, indicating superior accuracy. Dedicated hippocampal segmentation algorithms outperformed whole-brain methods. Full article

Background: This study aimed to evaluate the diagnostic performance of the Kaiser score (KS) on the modified abbreviated breast magnetic resonance imaging (AB-MRI) protocol for characterizing breast lesions by comparing it with full-protocol MRI (FP-MRI), using the histological data as the reference standard. Methods: Breast MRIs detecting histologically verified contrast-enhancing breast lesions were evaluated retrospectively. A modified AB-MRI protocol was created from the standard FP-MRI, which comprised axial fat-suppressed T2-weighted imaging (T2WI), pre-contrast T1-weighted imaging (T1WI), and first, second, and fourth post-contrast phases. Two radiologists reviewed both protocols, recording the KS for each detected lesion. Sensitivity, specificity, and positive and negative predictive values, as well as accuracy, were calculated for each protocol. Receiver operating characteristic (ROC) analysis was performed to determine the diagnostic performance of the modified AB-MRI compared to the FP-MRI. Results: In total, 154 patients with 158 histopathologically proven lesions (107 malignant, 51 benign) were included. For the diagnostic performance of the KS for modified AB-MRI and FP-MRI, the sensitivity was 96.3% vs. 98.1%, the specificity was 78.4% vs. 74.5%, PPV was 90.4% vs. 89%, NPV was 90.9% vs. 95%, and the diagnostic accuracy was 90.5% vs. 90.5%. The area under the curve (AUC) obtained from the ROC curve analysis was 0.873 and 0.863 for modified AB-MRI and FP-MRI for reader 1, respectively, and 0.859 and 0.878 for modified AB-MRI and FP-MRI for reader 2, respectively, (p < 0.001). Conclusions: Our modified AB-MRI protocol revealed comparable results in terms of the diagnostic value of the KS in characterizing breast lesions compared to FP-MRI and reduced both scanning and interpretation time. Full article

Backgrounds: Imaging of parotid tumors is crucial for surgery planning, but it cannot distinguish malignant from benign lesions with absolute reliability. The aim of the study was to establish a diagnostic MRI algorithm to differentiate parotid tumors. Methods: A retrospective study was conducted including all patients with parotid tumors, who underwent 3T-MRI and surgery. Morphological characteristics and normalized T2 and late postcontrast T1 signal intensities (SI) were assessed. “Ghosting sign” on late postcontrast T1 sequence was defined as indistinguishability of the tumor except for a thin peripheral enhancement. Patients were divided according to histology and imaging data were compared. A diagnostic MRI algorithm was established. Results: Thirty-six patients were included. The combination of normalized late T1 postcontrast SI, normalized T2 SI and “ghosting sign” allowed for the distinguishing of malignant from benign parotid tumors with high sensitivity (100%), specificity (93%), positive predictive value (80%), negative predictive value, (100%) and accuracy (94%). Moreover, pleomorphic adenomas often showed a homogeneous T2 signal and a complete capsule (p < 0.01), Warthin tumors protein-rich cysts and calcifications (p < 0.005 and p < 0.05), and malignant tumors an inhomogeneous contrast enhancement (p < 0.01). Conclusions: High field MRI represents a promising tool in parotid tumors, allowing for an accurate differentiation of malignant and benign lesions. Full article

Idiopathic orbital inflammation (IOI), or orbital pseudotumor, is a benign, non-infectious condition representing 8–10% of orbital mass lesions. This article presents a rare case of idiopathic orbital myositis (IOM) in a 45-year-old diabetic woman, who presented with acute right eye pain, diplopia, and motility deficits. Magnetic Resonance Imaging (MRI) revealed typical signs of inflammation, including hyperintense signals in T2-weighted Turbo Spin Echo (TSE) and Short Tau Inversion Recovery (STIR) sequences, along with post-contrast enhancement. Notably, there was no muscle belly enlargement, and the patient did not respond to corticosteroid therapy. Strict glycemic control, however, led to clinical improvement, suggesting a potential link between diabetes and IOM. Additionally, a comprehensive literature review on imaging in IOM was conducted, covering articles published from 2000 to 2024. The review highlights MRI as the primary diagnostic tool for IOM, offering a high sensitivity and specificity in differentiating it from other orbital conditions. This case underscores the importance of modern imaging techniques in diagnosis and emphasizes the need for continued research in evidence-based medicine, especially in complex cases where disease boundaries are not clearly defined. Full article

Background: The combination of oral pentoxifylline (Ptx) and vitamin E (VitE) has been used to treat radiation-induced fibrosis and soft tissue injury. Here, we review outcomes and perform a radiomic analysis of treatment effects in patients prescribed Ptx + VitE at our institution for the treatment of radiation necrosis (RN). Methods: A total of 48 patients treated with stereotactic radiosurgery (SRS) had evidence of RN and had MRI before and after starting Ptx + VitE. The radiation oncologist’s impression of the imaging in the electronic medical record was used to score response to treatment. Support Vector Machine (SVM) was used to train a model of radiomics features derived from radiation necrosis on pre- and 1st post-treatment T1 post-contrast MRIs that can classify the ultimate response to treatment with Ptx + VitE. Results: A total of 43.8% of patients showed evidence of improvement, 18.8% showed no change, and 25% showed worsening RN upon imaging after starting Ptx + VitE. The median time-to-response assessment was 3.17 months. Nine patients progressed significantly and required Bevacizumab, hyperbaric oxygen therapy, or surgery. Patients who had multiple lesions treated with SRS were less likely to show improvement (p = 0.037). A total of 34 patients were also prescribed dexamethasone, either before (7), with (16), or after starting (11) treatment. The use of dexamethasone was not associated with an improved response to Ptx + VitE (p = 0.471). Three patients stopped treatment due to side effects. Finally, we were able to develop a machine learning (SVM) model of radiomic features derived from pre- and 1st post-treatment MRIs that was able to predict the ultimate treatment response to Ptx + VitE with receiver operating characteristic (ROC) area under curve (AUC) of 0.69. Conclusions: Ptx + VitE appears safe for the treatment of RN, but randomized data are needed to assess efficacy and validate radiomic models, which may assist with prognostication. Full article

Background: Left ventricular remodeling in chronic mitral regurgitation (MR) encompasses two types of myocardial fibrosis: replacement fibrosis, identified by late gadolinium enhancement (LGE), and diffuse interstitial fibrosis, assessed by pre- and postcontrast T1 mapping techniques. These may explain irreversible LV dysfunction after MR correction. We aimed to assess the presence of myocardial fibrosis in patients with moderate and severe MR with no criteria for surgery versus mild MR controls. Methods: We enrolled 137 patients with chronic primary MR and 130 controls; all underwent cardiac magnetic resonance, and were followed up in a median of 2.9 years to assess mortality and the need for mitral valve replacement. Results: Patients in the study group displayed significantly higher degrees of LGE (28.4% vs 7.69%, p < 0.05), higher native T1 values (1167 ± 58.5 versus 971 ± 51.4 (p < 0.05)), and higher extracellular volumes compared to controls (32.3% ± 3.5 versus 23.9 ± 2.2, (p < 0.05)). The composite outcome occurred in 28 patients in the study group (20.4%), and significantly higher with LGE+ (78.5%). Replacement fibrosis (HR = 1.83, 95% CI, p < 0.01) and interstitial fibrosis (HR = 1.61, 95% CI, p < 0.01) were independent predictors for the composite outcome. Conclusions: Patients with moderate and severe MR with no criteria for surgery still exhibit a significant degree of both replacement and interstitial fibrosis, with prognostic implications. Full article

Primary hepatic angiosarcoma (PHA) is a rare liver malignancy with few studies describing its radiological characteristics. This article aims to assess the imaging features of each of the multiple delayed contrast-enhanced magnetic resonance imaging (MRI) scans, in addition to the conventional MRI protocol, in a patient with PHA. Standard MRI sequences and a liver protocol were used in the examination of a 71 year-old male with pathologically proven PHA after current imaging evaluation. In addition, the patient underwent transversal and coronal MRI T1-weighted scans at 10 min, 20 min and 90 min after intravenous (IV) administration of gadobenatedimeglumine (Gd-BOPTA). The PHA revealed a variable appearance on MRI, with classic imaging being insufficient in making a reliable diagnosis. Lesions have increased vascularity, which translates into increased IV contrast uptake in the MRI arterial phase, showing progressive and globular enhancement in the portal and parenchymatous phases. On delayed scans, at 10 min after IV administration, the lesions maintained no washout, but slightly began to washout at 20 min post-contrast. However, in the hepatobiliary phase (90 min post-contrast injection), on an MRI T1-weighted sequence, PHA lesions were hypointense, suggesting the absence of hepatocytes, thus indicating high-grade malignancy. This approach proved the conclusion that in a patient with PHA, an extra MRI T1-weighted scan at 90 min post-gadobenatedimeglumine injection can provide helpful information in differential diagnosis. Full article

Background: Head and neck cancer treatment does not yield desired outcomes for all patients. This investigation aimed to explore the feasibility of predicting treatment outcomes from routine pre-treatment magnetic resonance images (MRIs). Radiomics features were “mined” and used to train machine learning (ML) classifiers to predict treatment outcomes. Moreover, iterative deep texture analysis (DTA) was explored to boost model performances. Methods: Radiomics features were determined from T1-weighted post-contrast MRIs of pathologically involved lymph node (LN) segmentations for n = 63 patients. SVM, k-NN, and FLD classifier models were trained, selecting for 1–10 features. The model with the top balanced accuracy was chosen for an iteration of DTA. New feature sets were used to retrain and test the ML. Radiomics features were explored for a total of three layers through two iterations of DTA. Results: Models proved useful in predicting treatment outcomes. The best model was a nine-feature multivariable k-NN model with a sensitivity ($\%S_n$) of 93%, specificity ($\%S_p$) of 74%, 86% accuracy ($\%Acc$), and 86% precision ($\%Per$). The best model for two of the three classifiers (k-NN and FLD) was trained using features from three layers. The performance of the average k-NN and FLD models trained with features was boosted significantly with the inclusion of deeper-layer features. Conclusions: Pre-treatment LN MRIs contain quantifiable texture information that can be used to train ML models to predict cancer treatment outcomes. Furthermore, DTA proved useful to boosting predictive models. Full article

Patients diagnosed with glioblastoma multiforme (GBM) continue to face a dire prognosis. Developing accurate and efficient contouring methods is crucial, as they can significantly advance both clinical practice and research. This study evaluates the AI models developed by MRIMath© for GBM T1c and fluid attenuation inversion recovery (FLAIR) images by comparing their contours to those of three neuro-radiologists using a smart manual contouring platform. The mean overall Sørensen–Dice Similarity Coefficient metric score (DSC) for the post-contrast T1 (T1c) AI was 95%, with a 95% confidence interval (CI) of 93% to 96%, closely aligning with the radiologists’ scores. For true positive T1c images, AI segmentation achieved a mean DSC of 81% compared to radiologists’ ranging from 80% to 86%. Sensitivity and specificity for T1c AI were 91.6% and 97.5%, respectively. The FLAIR AI exhibited a mean DSC of 90% with a 95% CI interval of 87% to 92%, comparable to the radiologists’ scores. It also achieved a mean DSC of 78% for true positive FLAIR slices versus radiologists’ scores of 75% to 83% and recorded a median sensitivity and specificity of 92.1% and 96.1%, respectively. The T1C and FLAIR AI models produced mean Hausdorff distances (<5 mm), volume measurements, kappa scores, and Bland–Altman differences that align closely with those measured by radiologists. Moreover, the inter-user variability between radiologists using the smart manual contouring platform was under 5% for T1c and under 10% for FLAIR images. These results underscore the MRIMath© platform’s low inter-user variability and the high accuracy of its T1c and FLAIR AI models. Full article