Contrast-enhanced Magnetic Resonance Imaging Research Articles

A simplified fluid-sensitive MRI protocol for the hands to detect inflammation without contrast administration: a large study of symptom-free subjects from the general population as a reference for normality.

MRI of the hands is valuable for risk-stratification in patients with arthralgia at-risk for developing rheumatoid arthritis (RA). Contrast-enhanced MRI is considered standard for assessment of RA, but has practical disadvantages. It also shows inflammation-like features in the general population, especially at older age, which should be considered in image interpretation. The modified-Dixon(mDixon) technique is reliable compared to contrast-enhanced sequences. Moreover, this short protocol without contrast-enhancement is patient-friendly. Whether it also shows inflammation-like features in the general population is unknown. We studied this to support accurate use in the clinic. Two hundred twenty symptom-free volunteers from different age-categories were recruited from the general population and underwent mDixon MRI of both hands. Two readers independently scored MRIs for synovitis, tenosynovitis, and bone marrow edema (BME) in the metacarpophalangeal-joints (MCP) and wrists according to the RAMRIS. Features were considered present if scored by both readers; frequencies > 5% were considered relevant in terms of specificity and determined per age-category (< 40/40- < 60/ ≥ 60-years). Higher age correlated with higher BME-scores (p-value < 0.005), but not with synovitis and tenosynovitis-scores. BME (grade 1) occurred in some bones in people aged ≥ 60, 14% had BME in the lunate, 7% in metacarpal-1, and 6% in the trapezium. Synovitis and tenosynovitis did not occur in > 5%, except for grade-1 synovitis in the right distal radio-ulnar-joint in people aged ≥ 60 (11%). On mDixon MRI, inflammatory features in the hands of the general population are rare. This facilitates image interpretation. To prevent overinterpretation, only several locations should be considered when evaluating people aged ≥ 60-years.

Enhancing brain tumor classification by integrating radiomics and deep learning features: A comprehensive study utilizing ensemble methods on MRI scans

Background and objective This study aims to assess the effectiveness of combining radiomics features (RFs) with deep learning features (DFs) for classifying brain tumors—specifically Glioma, Meningioma, and Pituitary Tumor—using MRI scans and advanced ensemble learning techniques. Methods A total of 3064 T1-weighted contrast-enhanced brain MRI scans were analyzed. RFs were extracted using Pyradiomics, while DFs were obtained from a 3D convolutional neural network (CNN). These features were used both individually and together to train a range of machine learning models, including Support Vector Machines (SVM), Decision Trees (DT), Random Forests (RF), AdaBoost, Bagging, k-Nearest Neighbors (KNN), and Multi-Layer Perceptrons (MLP). To enhance the accuracy of these models, ensemble approaches such as Stacking, Voting, and Boosting were employed. LASSO feature selection and five-fold cross-validation were utilized to ensure the models’ robustness. Results The results demonstrated that combining RFs and DFs significantly improved the model's performance compared to using either feature set alone. The best performance was achieved using the combined RF + DF approach with ensemble methods, particularly Boosting, which resulted in an accuracy of 95.0%, an AUC of 0.92, a sensitivity of 88%, and a specificity of 90%. Conversely, models utilizing only RFs or DFs showed lower performance, with RFs reaching an AUC of 0.82 and DFs achieving an AUC of 0.85. Conclusion The integration of RFs and DFs, along with advanced ensemble methods, significantly improves the accuracy and reliability of brain tumor classification using MRI. This approach shows strong clinical potential, with opportunities for further enhancing generalizability and precision through additional MRI sequences and advanced machine learning techniques.

Magnetic resonance imaging findings of cerebral venous malformations.

Our study aimed to evaluate the incidence and anatomical locations of cerebral venous malformations (CVMs), their alterations over time, and magnetic resonance imaging (MRI) findings of their relationships with each other in patients who underwent magnetic susceptibility-weighted imaging (SWI) and contrast-enhanced MRI. The drainage pattern of developmental venous anomalies (DVAs), the number of collecting veins, DVA thrombosis, signal-intensity abnormalities, and hemosiderin deposits (nodular, diffuse, or both) related to the DVA, were examined. The alterations over time in cavernomas (type and size) were assessed. The study evaluated 7,826 patients who had 8,957 magnetic SWIand contrast-enhanced brain MRIs. A total of 643 CVMs were detected in 520 patients. The frequencies of DVAs, cavernomas, and capillary telangiectasias were 5.26%, 1.28%, and 0.62%, respectively. Isolated DVAs were detected in 430 (71.5%), cavernomas in 77 (12.8%), and capillary telangiectasias in 52 (8.7%) cases of CVMs. The coexistence of DVA with cavernoma and DVA with capillary telangiectasia was found in 38 (6.3%) and 4 (0.7%) cases, respectively. DVA drainage was largely provided by a single collecting vein. Signal-intensity abnormalities associated with the DVA were observed in 7, and two patients had DVA thrombosis. Hemosiderin deposits were nodular in 14 cases, diffuse in three, and both nodular and diffuse in two. In a follow-up of 11 (7.8%) cavernomas, alterations in the type and size were detected. It is not uncommon for cavernomas to become symptomatic owing to alterations in their nature, size, and type over time, while DVAs rarely become symptomatic. Capillary telangiectasias are vascular malformations that tend to remain stable in terms of clinical and imaging features, and those with supratentorial localization are detected more frequently in the frontal lobe. On magnetic susceptibility sequences, hemosiderin deposits can be seen in the GVA region, not rarely but more frequently in nodular form, as well as in diffuse or nodular or diffuse and nodular forms.

Comparison of tracer kinetic models in differentiating malignant from normal prostate tissue using dynamic contrast-enhanced MRI

PurposeThe aim of this study was to evaluate the diagnostic value of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) derived kinetic parameters with high spatiotemporal resolution in discriminating malignant from normal prostate tissues.MethodsFifty patients with suspicious of malignant diseases in prostate were included in this study. Regions of interest (ROI) were manually delineated by experienced radiologists. Voxel-wise kinetic parameters were produced with the following tracer kinetic models (TKMs): Tofts model, extended Tofts model (ETM), Brix’s conventional two-compartment model (Brix), adiabatic tissue homogeneity model (ATH), and distributed parameter model (DP). The initial area under the signal-time curve (IAUC) with an uptake integral approach was also included. Mann–Whitney U test and receiver operating characteristic (ROC) curves were used to evaluate the capability of distinguishing tumor lesions from normal tissues. A p-value of 0.05 or less is considered statistically significant. ROI based parameters correlation analysis between DP and ETM were performed.Results624 lesions and 269 normal tissue ROIs were obtained. Thirty parameters were derived from the six kinetic models. Except for PS from Brix, statistically significant differences between lesions and normal tissues (P&amp;lt;0.05) were observed in other parameters.Ve from DP, ATH and Brix and PS from ATH have AUC values less than 0.6 in the ROC analysis. MTT, Vp and PS from DP, Ktrans from ETM and Tofts, E and PS from ATH, IAUC parameters and F from Brix have AUC values larger than 0.8. Ve and Vp from DP and ETM are correlated (r&amp;gt; 0.65). The correlation coefficient between Ktrans from ETM and PS from DP is 0.751.ConclusionMTT, Vp and PS from DP, Ktrans from ETM and Tofts, E and PS from ATH, F from Brix and IAUC parameters can be used to differentiate malignant lesions from normal tissues in the prostate.

Gadolinium Elimination in a Gadolinium Deposition Disease Population After a Single Exposure to Gadolinium-Based Contrast Agents.

This study documents the gadolinium (Gd) content in urine over time after the administration of a single dose of Gd-based contrast agent (GBCA) in patients diagnosed with Gd deposition disease. In this retrospective observational study, 45 subjects with normal renal function who had performed 1 contrast-enhanced magnetic resonance imaging and had a nonprovoked (native) 24-hour urine test for Gd quantification after the examination were evaluated. The GBCA brand and the time interval in days between the GBCA administration and 24-hour urine Gd measurements were recorded. Log-log plot visualization of time points for urine Gd content was obtained. Time points collected for urine Gd content showed that Gd was above the reference levels for 3 months postinjection. The urinary concentration of Gd was similar for all agents, including linear and macrocyclic. The urinary content decreased in a dog-leg fashion. Gd urine content was substantially elevated at 1 month and decreased to remain above the accepted normal range by 3 months. Gd is retained in the body and shows demonstrable continued spontaneous elimination in urine for at least several months after administration, including the most stable macrocyclic agents. The Gd elimination pattern shows a logarithmic decrease pattern between 1 and 3 months for all agents, regardless of their structure.

Delayed Magnetic Resonance Imaging of Alzheimer's Disease by Using Poly(2-(methacryloyloxy)ethyl phosphorylcholine)-Functionalized Nanoprobes.

Alzheimer's disease (AD) is one of the most common neurodegenerative diseases, commonly affecting the aged, with pathophysiological changes presenting 15 to 20 years before clinical symptoms. Early diagnosis and intervention are crucial in effectively slowing the progression of AD. In the current study, poly(2-(methacryloyloxy)ethyl phosphorylcholine) (PMPC)-functionalized NaGdF4 nanoparticles (NaGdF4-PMPC) were developed as magnetic resonance imaging (MRI) contrast agents for targeting alpha 7 nicotinic acetylcholine receptors (α7 nAChRs) in AD mice. NaGdF4-PMPC showed excellent biocompatibility, targeting ability, and MRI performance, with the longitudinal molar relaxivity (r1) and transverse molar relaxivity (r2) being 1.21-fold and 1.33-fold higher than those of the clinical contrast agent Gd-DTPA, respectively, resulting in higher-sensitive MR angiography. After intravenous injection, 3D dynamic contrast-enhanced (DCE) MR images with high-resolution vasculature of the mouse brain were obtained. In addition, by using NaGdF4-PMPC, susceptibility-weighted imaging (SWI) signals in AD mouse brains were greatly retained compared to those in healthy mice for 24 h, emphasizing the excellent targeting ability of NaGdF4-PMPC. Furthermore, the CD31, α7 nAChRs, and Thioflavin S staining were also utilized to investigate the relationship among vascular inflammation, α7 nAChRs, and amyloid-β (Aβ) deposition in AD mice. This work highlights a promising targeted imaging strategy for the timely diagnosis of AD.

A head-to-head comparison of breast lesion’s conspicuity at contrast-enhanced mammography and contrast-enhanced MRI

Abstract Purpose Lesion conspicuity, the relative enhancement of a lesion compared to surrounding tissue, is a new descriptor in the ACR BI-RADS 2022 CEM supplement. We compared lesion conspicuity in contrast-enhanced mammography (CEM) and contrast-enhanced MRI (CE-MRI) in patients with suspicious breast lesions. Materials and methods IRB-approved retrospective study; three blinded readers rated 462 indeterminate or suspicious breast lesions in 388 patients (54.2 ± 11 years; range 30–90) who underwent CEM and CE-MRI from 2018 to 2022. Each lesion’s conspicuity was scored from 1 to 5, with 5 indicating excellent conspicuity. Visual grading characteristics (VGC) analysis and area under the curve (AUC) were used for comparison, with sub-analyses for benign and malignant lesions. Results VGC analysis showed a significant AUC of 0.670 to 0.723 (p &lt; 0.001) favouring CE-MRI. No lesion enhancement (score 1) was observed in 16.2% of CE-MRI and 44.7% of CEM. Excellent conspicuity was seen in 29.6% of CE-MRI and 11.9% of CEM. Sub-analysis showed higher conspicuity on CE-MRI for both malignant (AUC 0.665 to 0.732, p &lt; 0.001) and benign lesions (AUC 0.734 to 0.798, p &lt; 0.001). CE-MRI showed higher lesion conspicuity compared to CEM both for non-mass lesions (0.656) and for mass lesions 0.605. Conclusion CE-MRI shows significantly higher conspicuity for benign and malignant breast lesions compared to CEM, especially for benign lesions. The low conspicuity of benign lesions on CEM may help reduce false positives in clinical practice. Key Points QuestionLesion conspicuity is a new descriptor for lesion enhancement according to the new CEM lexicon. Data correlating lesion conspicuity with malignancy likelihood are limited. FindingsLesion conspicuity is higher for contrast-enhanced-MRI than for contrast-enhanced mammography (CEM) for all lesions but significantly better for benign lesions. Clinical relevanceThe low conspicuity of benign lesions on CEM may reduce false-positive results, making it a valuable tool in breast cancer screening.

Contrast-Enhanced Digital Breast Tomosynthesis Compared With Contrast-Enhanced Mammography and Magnetic Resonance Imaging in the Assessment of Breast Lesions: A Pilot Study.

Contrast-enhanced mammography (CEM) is an accurate competitor for contrast-enhanced breast magnetic resonance imaging (CE-MRI), but the examination is limited by the lack of 3D information. Digital breast tomosynthesis (DBT) allows better lesion detection and characterization compared with mammography. The availability of quasi-3D contrast imaging could further improve the performance of CEM. The aim of our analysis was to compare the diagnostic performance of a contrast-enhanced digital breast tomosynthesis prototype (CE-DBTp) to CEM and to CE-MRI. This prospective study was approved by the ethics committee, and all patients gave written informed consent. Women who presented with suspicious findings on mammography, DBT, or ultrasound were invited to participate in the study. Participants underwent CEM and CE-DBTp of the breast with the suspicious findings as well as bilateral CE-MRI. Histology was used as the standard of reference. Four readers (R1 and R2 non-experienced; R3 and R4 experienced) evaluated the images, blinded to patients' history, previous imaging, and histology. The readers evaluated CEM, CE-DBTp, and CE-MRI in separate sessions and gave a BI-RADS score for each finding. Sensitivity, specificity, lesion conspicuity, and readers' confidence were calculated and compared. We included 84 patients (mean age, 56 years; range, 39-70) with 91 histologically verified breast lesions (27 benign, 64 malignant). The accuracy of the CE-DBTp was high, but significant differences were seen between experienced (both 86.8%) and non-experienced readers (76.9% and 78%, P = 0.021). No differences were found between CEM and CE-DBTp, whereas the accuracy of CE-MRI was higher (P = 0.002). Sensitivity with CE-DBTp varied (89.1% to 100%) between experienced and non-experienced readers (P = 0.074), and it was comparable to CEM but lower than CE-MRI (P = 0.003). Specificity was variable between readers with all modalities. Lesion conspicuity was higher for the CE-DBTp and CE-MRI than for CEM, and confidence was significantly higher with the CE-DBTp than with CEM for one of the readers (P < 0.001). A high sensitivity and good accuracy were achieved with the CE-DBTp. Lesion conspicuity and readers' confidence were higher with the CE-DBTp compared with CEM. However, CE-MRI had the highest sensitivity and accuracy.

Sturge–Weber syndrome: updates in translational neurology

Sturge–Weber syndrome (SWS) is a rare congenital neurovascular disorder that initially presents with a facial port-wine birthmark (PWB) and most commonly associated with a R183Q somatic mosaic mutation in the gene GNAQ. This mutation is enriched in endothelial cells. Contrast-enhanced magnetic resonance imaging (MRI) diagnoses brain abnormalities including leptomeningeal vascular malformation, an enlarged choroid plexus, and abnormal cortical and subcortical blood vessels. Mouse SWS models identify dysregulated proteins important for abnormal vasculogenesis and blood brain barrier permeability. Recent clinical research has focused on early diagnosis, biomarker development, presymptomatic treatment, and development of novel treatment strategies. Prospective pilot clinical drug trials with cannabidiol (Epidiolex) or with sirolimus, an mTOR inhibitor, indicate possible reductions in seizure frequency and improved cognitive outcome. This review connects the most recent molecular research in SWS cell culture and animal models to developing new treatment methods and identifies future areas of research.

Prediction model for cesarean scar disorder disappearance following transvaginal uterine diverticulum repair

Cesarean scar disorder (CSD) is a recently defined condition resulting from cesarean delivery (CD) surgery, and transvaginal uterine diverticulum repair (TVUDR) is generally chosen for treatment of CSD. This study constructed a prediction model for CSD disappearance following TVUDR using imaging data for CSD measured by transvaginal ultrasound (TVU), magnetic resonance imaging (MRI), and contrast-enhanced MRI. The data of 283 women with prior CD treated with TVUDR were retrospectively collected between January 2014 and February 2016. The imaging data for the CSD parameters were measured, including length (L), width (W), depth (D), residual myometrium thickness (RMT), residual myometrium thickness/depth (RMT/D), and residual myometrium thickness/(residual myometrium thickness + depth) RMT/(RMT+D). Of the included patients, 129 women presented with disappearance of CSD. We noted potential differences between CSD disappearance and existence following TVUDR for W, RMT, RMT/D, and RMT/(RMT+D) measured by MRI, and D, RMT/D, and RMT/(RMT+D) measured by contrast-enhanced MRI. After adjusting for potential confounding factors, W measured using MRI was found to be associated with the disappearance of CSD (odds ratio, 1.134; 95% CI, 1.050-1.224; P = 0.001). Subsequently, the W measured by MRI was selected in the prediction model, for which the C index was 0.624. The area under the receiver operating characteristic curve in the least absolute shrinkage and selection model was 62.40% (95% CI, 54.96-69.83%). MRI and contrast-enhanced MRI were found to be relatively accurate methods for detecting CSD. Moreover, W measured using MRI was significantly associated with the disappearance of CSD following TVUDR.

Prognosis of asymptomatic endolymphatic hydrops in healthy volunteers: A five-year cohort study.

This study aimed to clarify the prognosis of asymptomatic endolymphatic hydrops (EH) in healthy volunteers via five-year follow-ups with inner ear magnetic resonance imaging (MRI). Inner ear MRI was performed on 115 participants recruited as controls in a previous study on Meniere's disease. The endolymphatic space was visualized using Naganawa's method of contrast-enhanced MRI with intravenous gadolinium injection and evaluated using Nakashima's method of 2D imaging analysis. Cochlear or vestibular EH was present in 7.0% of participants (n = 8), with all cases being unilateral (laterality), moderate (severity), and asymptomatic (onset). Only cochlear-localized EH, only vestibular-localized EH, and both EH were present in 1.7% (n = 2) (C group), 4.3% (n = 5) (V group), and 0.9% (n = 1) (CV group) of participants, respectively. Conducting inner ear MRI after 5 years showed that EH had almost disappeared in two participants in the C and V groups (4/8, 50.0%). EH was still present in three participants in the V group and one in the CV group (4/8, 50.0%). One participant in the V group and another in the CV group presented with residual inner ear EH and developed typical symptomatic Meniere's disease (2/8, 25.0%). Approximately 7% of healthy participants showed asymptomatic EH. Therefore, EH is not the definitive marker for making a diagnosis of Meniere's disease or the suitable predictor for the development of Meniere's disease. Among these participants, 25% maintained EH and subsequently developed typical Meniere's disease within the next 5 years. Schellong-positive participants maintained persistent EH in the inner ear, and participants with higher scores on the self-rating depression scale developed Meniere's symptoms after 5 years. 2a.

Contrast-enhanced pseudo-CT of finger joints: Improved diagnostic confidence for detection of inflammatory erosions

PurposeTo evaluate the interest of additional pseudo-CT images to standard clinical contrast-enhanced MR images (CE-MRI) in the detection of inflammatory erosions and to differentiate them from intraosseous ganglia at the finger joints. Method47 prospectively included patients with suspected or diagnosed rheumatoid arthritis received a CE-MRI of the fingers. Additionally, pseudo-CT images were derived from non-contrast MRI (pCT) and from contrast-enhanced MRI data (CE-pCT) using a high-resolution gradient-echo 3D fast low-angle shot sequence (FLASH), respectively. CE-MRI, pCT, and CE-pCT images were evaluated for erosions and intraosseous ganglia at the metacarpophalangeal and proximal interphalangeal joints by two musculoskeletal radiologists. Findings were defined on CE-MRI by an experienced independent 3rd reader which served as reference standard. Diagnostic confidence (1 = worst,4 = best) was rated for 3 different image sets (1 = CE-MRI,2 = CE-MRI + pCT,3 = CE-MRI + CE-pCT) and compared. ResultsReference standard revealed 44 erosions and 37 intraosseous ganglia on CE-MRI. Diagnostic confidence for CE-MRI was moderate for erosions (both readers) and intraosseous ganglia (reader 2). For the combination of CE-MRI and pCT, high confidence levels were observed for erosions for both readers (median 3 (Q1–Q3: 2–3) and 3 (3–3);p<.001 and p<.001) and for intraosseous ganglia for reader 2 (3 (3–3); p<.001). CE-MRI combined with CE-pCT showed very high confidence levels for both readers for erosions (4 (3–4) and 4 (4–4);p<.001 and p<.001) and for intraosseous ganglia (4 (3.5–4) and 4 (4–4);p<.001 and p<.001). ConclusionsPseudo-CT combined with CE-MRI increases the confidence in detection of inflammatory erosions at the finger joints, with CE-pCT being superior to pCT.

New strategy of LI-RADS v2018 to improve the sensitivity for small hepatocellular carcinoma ≤ 3.0 cm on extracellular-contrast enhanced MRI

IntroductionWe aimed to modify LI-RADS version 2018 to improve sensitivity and determine the value of the combination of high alpha-fetoprotein (AFP) levels for small HCC (sHCC, ≤ 30 mm) diagnosis. MethodsA total of 984 patients at high risk for HCC, with 1204 observations (including 997 small observations ≤ 30 mm), who underwent extracellular contrast-enhanced MRI were enrolled from five independent centers. Blinded readers evaluated the LI-RADS features and categorized each observation according to the LI-RADS v2018, modified LI-RADS and EASL. Odds ratios of LI-RADS major features (MFs) and several high AFP levels for sHCC diagnosis were analyzed using multivariable logistic regression. The modified LR-5 criteria was developed by including no APHE at any size with two MFs, and non-rim APHE with one MF (≥ 10 mm) or with two MFs (< 10 mm). The diagnostic performance of each version of the LR-5 was compared using generalized estimating equations. ResultsAPHE, washout, enhancing capsule and five high AFP levels were independently associated with sHCC. In three datasets, the modified LI-RADS had higher sensitivities for sHCC (76.8 ∼ 85.5 % vs. 73.7 ∼ 75.9 %, P < 0.05) to the LR-5 v2018. The modified LI-RADS with AFP ≥ 200 ng/mL as an additional feature or as an alternative to threshold growth provided higher sensitivities for sHCC than LI-RADS v2018 (82.1 ∼ 90.1 % vs. 73.7 ∼ 75.9 %, all P < 0.05), modified LI-RADS (82.1 ∼ 90.1 % vs. 76.8 ∼ 85.5 %, all P < 0.05) and EASL version 2018 (82.1 ∼ 90.1 % vs. 73.3 ∼ 74.7 %, all P < 0.05), with comparable specificities (all P > 0.05). ConclusionThe new strategy of LI-RADS v2018 provides significantly higher sensitivity and comparable specificity than those of LI-RADS v2018 for sHCC diagnosis on ECA-MRI.

Comparative analysis of audiovestibular testing results in vestibular schwannomas

Objective:This study aimed to analyze the results of auditory and vestibular function tests in patients with unilateral vestibular schwannoma and explore their association with tumor size. Methods:Clinical data from 81 patients diagnosed with unilateral vestibular schwannoma who underwent pure-tone audiometry（PTA）, cervical and ocular vestibular evoked myogenic potentials（c/oVEMP）, as well as video head impulse test（vHIT）, and subsequently underwent surgical treatment, were retrospectively analyzed. Patients were categorized into groups based on tumor size: small （≤ 15 mm）, medium（16-30 mm）, and large （>30 mm）, determined by the maximum tumor diameter on contrast-enhanced MRI scans. Results:PTA results indicated hearing loss in 73 cases（90.1%）; vestibular function tests revealed abnormal rates for the anterior semicircular canal, horizontal semicircular canal, posterior semicircular canal, utricle, and saccule at 29.6%, 77.8%, 54.3%, 90.1%, and 92.6%, respectively. Statistically, no significant differences were found in preoperative hearing test results among patients in different groups（F=0.393, P=0.676）. However, significant differences were observed in horizontal semicircular canal gain（r=-0.248, P=0.025）, abnormal rates of horizontal semicircular canal catch-up saccades（r=0.507, P<0.001）, as well as cVEMP（χ²=15.111, P=0.004） and oVEMP thresholds（χ²=18.948, P<0.001） across varying tumor size groups. Conclusion:The extent of hearing loss in patients with vestibular schwannoma is not correlated with tumor size, whereas the degree of vestibular dysfunction demonstrates a correlation with tumor size.

Diagnostic Efficacy of Multimodal Ultrasound for Mass and Non-Mass Enhancements in Dynamic Contrast-Enhanced MRI of Idiopathic Granulomatous Mastitis and Breast Cancer

PurposeIdiopathic granulomatous mastitis (IGM) poses diagnostic challenges due to its diverse clinical and radiological presentations, often mimicking malignancies. This study aimed to assess the diagnostic efficacy of multimodal ultrasound for mass and non-mass enhancements in Dynamic Contrast-Enhanced MRI (DCE-MRI) of IGM and breast cancer. MethodsA retrospective analysis involved patients confirmed histopathologically with IGM and BC. All patients underwent conventional ultrasound (C-US), ultrasound elastography (UE), contrast-enhanced ultrasound (CEUS), and DCE-MR examinations. Blind experienced radiologists assessed imaging findings. Diagnostic accuracy, sensitivity, and specificity were calculated for mass and non-mass enhancements. ResultsFor mass enhancements (ME), multimodal ultrasound demonstrated strong efficacy (AUC = 0.8651, 95% CI: 0.7431 to 0.9871), exhibiting high sensitivity (83.3%) and specificity (92.4%) in differentiating IGM from breast cancer. However, for non-mass enhancements (NME), multimodal ultrasound showed limited accuracy (AUC = 0.6306) with lower sensitivity (65.6%) and specificity (81.2%) in distinguishing between IGM and breast cancer. ConclusionMultimodal ultrasound displayed good diagnostic efficacy for mass enhancements in DCE-MRI for idiopathic granulomatous mastitis and breast cancer, while for non-mass enhancement patterns, DCE-MRI remains the most valuable radiological modality for comprehensively assessing this condition's complexities.

Case report: Cerebral syphilitic gumma: a case retrospective report of eight cases

BackgroundCerebral syphilitic gumma (CSG), a rare manifestation of neurosyphilis, presents characteristics akin to intracranial tumors, often leading to clinical misdiagnosis.ObjectiveThis study aimed to summarize the clinical experience in diagnosing and treating CSG.Materials and methodsThe present study conducted a retrospective analysis of clinical data, encompassing the baseline characteristics, clinical presentation, diagnosis, treatment, and prognosis of eight patients with CSG who were treated and diagnosed by our institution.ResultsThe median age at the initial diagnosis was 57.5 years, comprising six males and two females. The predominant clinical manifestations included headaches in eight cases, motor and sensory disorders in four cases, epileptic seizures in one case, and dysarthria in one patient. The serum Treponema pallidum haemagglutination assay (TPHA) and toluidine red unheated serum test (TRUST) yielded positive in all eight patients. The contrast-enhanced magnetic resonance imaging (MRI) revealed irregular ring-shaped enhancement of lesions in three patients, while nodular enhancement of lesions was observed in five patients. In terms of treatment, seven patients underwent surgery. The postoperative pathological tissue showed granulomatous inflammation. Six patients exhibited an elevated protein concentration in their cerebrospinal fluid (CSF), while two patients demonstrated a reduced CSF glucose level. Additionally, positive results were obtained for both CSF TPHA and TRUST in all eight patients. The clinical diagnosis of CSG was confirmed in eight patients who exhibited notable clinical improvement following penicillin treatment. Subsequent reevaluation of the imaging findings demonstrated complete resolution of the enhanced lesions.ConclusionIn clinical practice, CSG should be considered for patients with intracranial lesions and positive serum syphilis antibodies. Timely and accurate diagnosis enables patients with CSG to achieve a more favorable prognosis through active anti-syphilis treatment.

EARLY DETECTION OF BRAIN TUMOR USING MRI AND TRANSFER LEARNING

Brain tumors pose significant risks to cognitive functions, making early detection crucial for improving patient survival rates. Accurate tumor detection aids neuro-oncologists in diagnosing tumor types and recommending appropriate treatments. However, manual detection is challenging, time-consuming, and prone to human error. Recently, Deep Learning (DL) models have demonstrated substantial potential in efficiently classifying large image datasets. This paper presents three novel and efficient multi-class DL architectures leveraging transfer learning to classify different brain tumors. Our primary contribution is to enhance the predictive accuracy while minimizing the usage of computational resource compared to other state-of-art models in the literature by forgoing preprocessing, segmentation, hybrid models, and augmentation techniques. Additionally, we reduce the number of FC layers to streamline computation. We conduct extensive experiments to evaluate the performance of our models using the brain tumor Figshare T1-weighted contrast-enhanced MRI dataset, comprising 3064 images from three distinct tumor types. The InceptionV3 model records a 98.36% accuracy level with five-fold cross-validation. By incorporating batch normalization and optimizing the learning rate for the Adam optimizer, the Xception model reached 99.19% accuracy. Finally, we utilize Particle Swarm Optimization (PSO) to fine-tune the learning rate of the Stochastic Gradient Descent (SGD) optimizer. The Xception model attained 98.85% accuracy. These results highlight the novelty of our approach, offering a practical solution for neuro-oncologists, particularly through the fine-tuned Xception model, which delivers early and accurate tumor detection with minimal computational resources.

Prediction of therapeutic response to transarterial chemoembolization plus systemic therapy regimen in hepatocellular carcinoma using pretreatment contrast-enhanced MRI based habitat analysis and Crossformer model.

To develop habitat and deep learning (DL) models from multi-phase contrast-enhanced magnetic resonance imaging (CE-MRI) habitat images categorized using the K-means clustering algorithm. Additionally, we aim to assess the predictive value of identified regions for early evaluation of the responsiveness of hepatocellular carcinoma (HCC) patients to treatment with transarterial chemoembolization (TACE) plus molecular targeted therapies (MTT) and anti-PD-(L)1. A total of 102 patients with HCC from two institutions (A, n = 63 and B, n = 39) who received TACE plus systemic therapy were enrolled from September 2020 to January 2024. Multiple CE-MRI sequences were used to outline 3D volumes of interest (VOI) of the lesion. Subsequently, K-means clustering was applied to categorize intratumoral voxels into three distinct subgroups, based on signal intensity values of images. Using data from institution A, the habitat model was built with the ExtraTrees classifier after extracting radiomics features from intratumoral habitats. Similarly, the Crossformer model and ResNet50 model were trained on multi-channel data in institution A, and a DL model with Transformer-based aggregation was constructed to predict the response. Finally, all models underwent validation at institution B. The Crossformer model and the habitat model both showed high area under the receiver operating characteristic curves (AUCs) of 0.869 and 0.877 (training cohort). In validation, AUC was 0.762 for the Crossformer model and 0.721 for the habitat model. The habitat model and DL model based on CE-MRI possesses the capability to non-invasively predict the efficacy of TACE plus systemic therapy in HCC patients, which is critical for precision treatment and patient outcomes.

Automatic segmentation of demyelination lesions in multiple sclerosis

Aim – To evaluate the effectiveness of the YOLOv8 algorithm for automatic segmentation of demyelination lesions in various locations in patients with multiple sclerosis. Material and methods. The study included 120 patients with a clinically confirmed diagnosis of multiple sclerosis who underwent contrast-enhanced MRI. MRI data from patients with different types of disease progression were analyzed. T1-weighted, T2-weighted, and FLAIR sequences were used for analysis. The YOLOv8 algorithm was adapted for medical imaging and trained on manually annotated MRI scans. Model performance was evaluated using precision, recall, and F1-Score metrics. Results. The YOLOv8 model demonstrated high segmentation performance with a precision of 0,79, recall of 00,73, and F1-Score of 0.65. The model effectively identified demyelination lesions in various locations typical for multiple sclerosis. However, there remains a need to improve recall to minimize missed lesions. Testing on independent data confirmed the stability of the model’s results. Conclusion. The YOLOv8 algorithm shows significant potential for automatic segmentation of demyelination lesions in multiple sclerosis patients. This method could be successfully implemented in clinical practice, enabling faster diagnosis and improved monitoring of disease progression. Further optimization of the model, through data augmentation techniques and hybrid architectures, may enhance both segmentation accuracy and recall.

Evaluation of temporomandibular joint involvement in juvenile idiopathic arthritis patients

ObjectiveJuvenile idiopathic arthritis (JIA) is a common, chronic and inflammatory rheumatological disease of childhood. The disease can affect all synovial joints in the body. Temporomandibular joints (TMJs) are important areas of involvement in JIA, which are frequently involved but often not noticed because the involvement is usually asymptomatic. The aim of this study is to determine the frequency and risk factors of TMJ joint involvement in juvenile idiopathic arthritis patients admitted to our clinic, and to guide for early diagnosis and treatment.MethodsPatients who applied to this study with the diagnosis of JIA between January 2014 and May 2017 at Pediatric Rheumatology Clinic, were followed up regularly in our clinic, had a accessible medical history, and a rheumatology polyclinic record. Patients with contrast-enhanced TMJ Magnetic Resonance Imaging (MRI) taken and reported by the radiologist were included.ResultsTMJ involvement was detected in 51.2% of the 41 patients included in the study. It was found that 71.5% of the patients with TMJ involvement were asymptomatic and 71.5% of the patients had chronic involvement. When the patients with and without TMJ involvement were compared according to the contrast-enhanced TMJ MRI results; In the patient group with involvement, the polyarticular onset subtype was seen at a higher rate (p = 0.005), the age of onset was earlier (p = 0.003), the disease duration was longer (p = 0.037), more joints were involved (p = 0.005), the ESR values ​​were higher (p = 0.0001), and the treatment compliance and treatment responses of the patients in this group were worse (p = 0.001, p = 0.0001).ConclusionTMJ involvement is common in JIA patients and can occur at any stage of the disease. It is often asymptomatic and progresses insidiously, leading to chronic and degenerative changes in the mandible at an early stage. Due to its asymptomatic nature, the insidious progression, and the risk of causing chronic, irreversible sequelae, it is crucial to screen high-risk JIA patients regularly with contrast-enhanced TMJ MRI, which remains the gold standard method. While specific risk factors are difficult to pinpoint, some factors may increase the likelihood of TMJ involvement. To better identify these high-risk patients and determine which individuals require regular screening, larger-scale and multicenter studies are essential.