Magnetic Resonance Imaging Scans Research Articles

Morphometric analysis of ventricular indexes and intracranial cerebrospinal fluid spaces in the brain using magnetic resonance imaging healthy in Van cats

BackgroundVentricular indexes are defined as a numerical marker of ventricular dimensions in domestic mammals. The anatomical size of the brain ventricles has been the subject of many studies so far and has been accepted as a potential indicator of many brain disorders in the clinical field. Currently, the number of studies describing the morphometry of the brain ventricular system in cats is limited. Furthermore, no studies have been conducted specifically on indexes characterizing the numerical compatibility of the brain and brain ventricles in cats. The aim of this study was to reveal the morphometric status of intracranial cerebrospinal fluid spaces and ventricular indexes in healthy Van cats by magnetic resonance imaging method.ResultsFor this retrospective study, T2-weighted magnetic resonance imaging (MRI) scans were performed at 1.5 T on 20 (10 male and 10 female) Van cats, under general anaesthesia. The animals were at the age of mean 4 (3–5 age). All Van cats were selected from individuals who had not undergone any surgical procedures, were not neutered, and had no visible anomalies. The statistical analysis of first, descriptive statistics such as mean and standard deviation were calculated. In line with the mean results obtained, the difference between sex was examined statistically. ‘A Mann-Whitney U test’ was applied to detect sex differences in measurement parameters in the study. The results are as follows (mean ± standard deviation): İntracranial cerebrospinal fluid spaces measurements: R-fss: 1.02 ± 0.19 mm, L-fss: 1.03 ± 0.18 mm, A-if: 1.06 ± 0.26 mm, R-sf: 1.38 ± 0.32 mm, L-sf: 1.37 ± 0.27, V1: 4.26 ± 0.53 mm. Indexes: Fourth ventricle: 15.95 ± 1.73%, Bifrontal: 17.45 ± 1.78%, Bioccipital: 47.53 ± 9.36%, Evans: 13.76 ± 2.93%, Lateral ventricle: 35.41 ± 2.50%, Callosal angle: 85.06°±4.42°.ConclusionsThe present study provides baseline values of intracranial cerebrospinal fluid spaces and linear indexes of the ventricles in the Van cats. The acquisition of these data contributes to filling the knowledge void on important anatomical and morphological features of the Van cats brain.

Inflammatory myofibroblastic tumor in the liver after bone marrow transplantation: case report and literature review

IntroductionInflammatory myofibroblastic tumor (IMT) is a rare low-grade malignant neoplasm in the liver. Timely diagnosis and treatment of IMT are challenging due to its atypical symptoms and imaging results.Case reportWe report a 46-year-old woman who presented to our hospital with persistent hyperpyrexia and discomfort in the right upper abdomen for 2 months post bone marrow transplantation. Radiological findings revealed a space-occupying lesion of uncertain nature in the liver. Since the histological examination of the biopsy specimen indicated IMT, she underwent surgical resection. Subsequently, the postoperative pathology confirmed the diagnosis of IMT. The patient’s febrile condition subsided after the surgery. A magnetic resonance imaging (MRI) scan performed 8 months later showed no signs of recurrence.ConclusionIMTs are caused by genetic rearrangements. Diagnosing IMT can be challenging especially in this case as we had to differentiate the tumor from inflammatory diseases associated with bone marrow transplantation. Hence, a thorough pathological immunohistochemical examination is required to confirm its diagnosis. Local IMTs should be treated with radical surgical resection. In cases of distant metastasis or incomplete resection cases, chemotherapy, targeted therapy, or immunotherapy can be utilized. Regular follow-up is crucial for improving the patient’s survival rate.

Federated Learning Framework for Brain Tumor Detection Using MRI Images in Non-IID Data Distributions.

Brain tumor detection from medical images, especially magnetic resonance imaging (MRI) scans, is a critical task in early diagnosis and treatment planning. Traditional machine learning approaches often rely on centralized data, raising concerns about data privacy, security, and the difficulty of obtaining large annotated datasets. Federated learning (FL) has emerged as a promising solution for training models across decentralized devices while maintaining data privacy. However, challenges remain in dealing with non-IID (independent and identically distributed) data, which is common in real-world scenarios. In this research, we used a client-server-based federated learning framework for brain tumor detection using MRI images, leveraging VGG19 as the backbone model. To improve clinical relevance and model interpretability, we have included explainability techniques, particularly Grad-CAM. We trained our model across four clients with non-IID data distribution to simulate real-world conditions. For performance evaluation, we used a centralized test dataset, consisting of 20% of the original data, with the test set used collectively for evaluating model performance after completing federated learning rounds. Using a separate test dataset ensures that all models are evaluated on the same data, making comparisons fair. Since the test dataset is not part of the FL training process, it does not violate the privacy-preserving nature of FL. The experimental results demonstrate that the VGG19 model achieves a high test accuracy of 97.18% (FedAVG), 98.24% (FedProx), and 98.45% (Scaffold) than other state-of-the-art models, showcasing the effectiveness of federated learning in handling distributed and non-IID data. Our findings highlight the potential of federated learning to address privacy concerns in medical image analysis while maintaining high performance even in non-IID settings. This approach provides a promising direction for future research in privacy-preserving AI for healthcare applications.

Determinants and Prognostic Value of Early Gadolinium Enhancement-Derived Myocardial Salvage Index in STEMI.

T2-weighted imaging is commonly used to measure myocardial salvage in reperfused myocardial infarction but is hindered by poor reproducibility and indistinct boundaries. Early gadolinium enhancement (EGE) emerges as an alternative for measuring the area at risk. This study aims to evaluate the determinants of the myocardial salvage index (MSI) derived from EGE and its prognostic implications. We analyzed acute cardiac magnetic resonance scans of 453 reperfused patients with ST-segment-elevation myocardial infarction (mean age, 60±12 years; 389 men) from April 2017 to July 2023 at a single center retrospectively. EGE was collected at 3 minutes after contrast agent injection, where hyperintense areas (signal intensities > mean+2SD of remote myocardium) were considered as the area at risk, plus the hypointense core within. MSI was calculated as the ratio of salvageable myocardium to the area at risk. Major adverse cardiovascular events included cardiovascular death, hospitalization for heart failure, reinfarction, and unplanned revascularization for the target vessel. During a median follow-up of 3.2 years (interquartile range, 1.6-4.7 years), at least one major adverse cardiac event occurred in 91 participants (20.1%). The median MSI was 35.0% (interquartile range, 22.9-59.5%), with smaller MSI observed in patients with larger infarcts (P<0.001). Linear regression identified prepercutaneous coronary intervention Thrombolysis in Myocardial Infarction flow (β=3.35, P<0.001) and microvascular obstruction (β=-11.92, P<0.001) as independent determinants of MSI. Multivariable Cox regression showed that every 10% increase in MSI was associated with a 32% reduction in major adverse cardiac event risk (hazard ratio, 0.68 [95% CI, 0.53-0.86]; P=0.001). A graded response was observed between MSI and cardiovascular death and reinfarction. MSI greater than the median was associated with nontarget vessel-related reinfarctions but not target vessel-related ones (nontarget, P=0.027; target vessel, P=0.36). Good reproducibility was reported with EGE-measured area at risk (intraobserver, intraclass correlation coefficient [ICC]=0.95; interobserver, ICC=0.89). EGE-derived MSI was associated with prepercutaneous coronary intervention Thrombolysis in Myocardial Infarction flow and microvascular injuries. It was an independent predictor of major adverse cardiac events. Our results highlight the prognostic potential of EGE imaging in acute myocardial infarction.

Radiographic Measurement of Anteriorization After Tibial Tubercle Osteotomy.

There is growing interest in sagittal plane malalignment as a risk factor for patellofemoral chondral wear and, correspondingly, as an important measure to correct when performing certain tibial tubercle osteotomy (TTO) procedures. However, a radiographic method to measure anteriorization after TTO has not been described. To develop and validate a radiographic method of measuring anteriorization after TTO. Cross-sectional study (diagnosis); Level of evidence, 3. Patients treated by 2 high-volume sports medicine surgeons at a single institution who underwent a TTO from 2015 to 2023 with available pre- and postoperative radiographic and magnetic resonance imaging (MRI) scans were identified. Approximately 10 mm and 0 mm of operative anteriorization were targeted for the anteromedializing and straight distalizing TTOs, respectively. Two methods to assess anteriorization after TTO on lateral knee radiographs were developed, using the preoperative to postoperative difference in distance between the anterior-most aspect of the tibial tubercle and either the center of the tibial shaft or the anterior tibial plateau. To validate the radiographic techniques, intraclass correlation coefficients (ICCs) were calculated between each method of radiographic measurement and the gold standard MRI measurement (preoperative to postoperative difference in sagittal tibial tubercle-trochlear groove distance). There were 70 patients (52 [74%] women) with a mean age of 31.5 ± 9.2 years. The mean anteriorization amount among the 57 anteriorizing TTOs was 4.9 ± 2.5 mm on the x-ray (XR) shaft technique, 4.6 ± 2.6 mm on the XR plateau technique, and 5.3 ± 2.7 mm on MRI (P = .35). The mean anteriorization amount among the 13 straight distalizing TTOs was 0.1 ± 2.5 mm on the XR shaft technique, -0.3 ± 2.2 mm on the XR plateau technique, and 0.6 ± 2.6 mm on MRI (P = .66). There was excellent agreement with MRI for both the XR shaft (ICC, 0.89) and XR plateau (ICC, 0.82) techniques. Interrater reliability was excellent for both techniques (ICC, 0.94-0.95). Anteriorization after TTO can be measured using routine pre- and postoperative radiographs. Additionally, the amount of anteriorization achieved with modern anteromedializing TTO techniques was less than that traditionally targeted. Moving forward, surgeons can assess the amount of anteriorization achieved during TTO on standard radiographs, while researchers may investigate the potential role of anteriorization on postoperative outcomes.

Protocol of the follow-up of patients with transthyretin amyloid cardiomyopathy by multimodality imaging (FAITH) study: a prospective observational study in patients with ATTR-CM undergoing treatment with tafamidis

IntroductionThis prospective observational study of patients with transthyretin amyloid cardiomyopathy (ATTR-CM) undergoing treatment with tafamidis aims at identifying quantitative image markers and comparing imaging modalities regarding the follow-up and prognostication...

Influence of infrapatellar fat pad size on the development and severity of chondromalacia patella.

The inflammatory role of the infrapatellar fat pad (IPFP) in cartilage damage has been well-documented, yet its potential protective function as a shock absorber remains underexplored. This retrospective cohort study aimed to evaluate the relationship between the IPFP size and chondromalacia patella (CP), while also examining the effects of age, sex, and body mass index (BMI). Magnetic resonance imaging scans from 311 patients aged 40 to 65 years were retrospectively analyzed. Axial sequences classified CP severity, and sagittal sequences measured IPFP areas. CP was graded according to the International Cartilage Repair Society system, and patients were grouped into control (no CP), mild CP (grades 1-2), and severe CP (grades 3-4) categories. Demographic data, including age, sex, and BMI, were collected, and statistical analysis explored the relationships between IPFP area, CP severity, and these factors. Of the patients, 145 (46.6%) had no CP, while 166 (53.4%) had varying CP severity. Patients with CP had significantly smaller IPFP areas (6.16 ± 0.67 cm2) compared to controls (6.96 ± 0.87 cm2, P < .001). The mean IPFP area decreased progressively with increasing CP severity. After adjusting for confounders, a smaller IPFP area was significantly associated with the presence and severity of CP (P < .001). These findings provide evidence that a larger IPFP area plays a protective role in maintaining patellar cartilage integrity and mitigating CP progression, as demonstrated by a significant inverse correlation between IPFP area and CP severity, independent of age, sex, and BMI. A comprehensive, multidisciplinary approach integrating biomechanical, metabolic, and inflammatory factors is warranted to fully elucidate the role of IPFP in CP.

Altered cerebellar subregion functional connectivity and structure in patients with pediatric bipolar depression.

To explore the structural and functional changes of Cerebellar Subregion in patients with pediatric bipolar disorder (PBD) patients and its clinical significance by using multimodal magnetic resonance imaging, so as to further explore the specific role of the cerebellum in PBD. This study included 48 pediatric patients with bipolar disorder (PBD) in the depressive phase from the outpatient clinic of the Department of Psychosomatic Medicine of the First Affiliated Hospital of Nanchang University. 22 healthy controls (HCs) matched for gender, age, handedness and education level were chosen from the community as the control group. All subjects underwent 3.0T resting-state functional magnetic resonance imaging (rs-fMRI) scans and completed clinical scales, including the Hamilton Depression Scale (HAMD) and Young Mania Rating Scale (YMRS). The cerebellum was categorized into 34 distinct subregions (R17, L17) based on SUIT and designated as seed points to perform whole-brain functional connectivity (FC). Group differences in categorical variables were assessed using the chi-square test, while continuous variables were compared employing the two-sample t-test. Correlations between FC and clinical parameters were analyzed for differential brain intervals. Compared with HCs, PBD patients in the depressive phase showed reduced FC between the left cerebellar lobules I-IV and the occipital inferior lobe (Occipital_Inf_L), cerebellar vermis VIIIa, and VIIIb; decreased FC between cerebellar cerebellar vermis VI and the frontal inferior orbital gyrus (Frontal_Inf_Orb_L), as well as the right cerebellar Crus 1; reduced FC between the left cerebellar Crus I and the dentate nucleus; decreased FC between cerebellar vermis VIIIa and the left superior frontal gyrus; reduced FC between the right cerebellar lobule IX and the right lingual gyrus; lowered FC between the left dentate and the dorsolateral prefrontal cortex, and the left lingual gyrus; FC between the left fastigial nucleus and the right cerebellar X decreased. Depressive phase of PBD patients exhibit altered functional connectivity within various subregions of the cerebellum, suggesting that the cerebellum is involved in central neural reorganization in PBD, which may be instructive for the understanding of central mechanisms and its future diagnostic and therapeutic target development.

Radiological evaluation of facial nerve and facial canal in patients with Bell's palsy.

To evaluate the diameter of the facial nerve (FN) and facial canal (FC) at different segments in Bell's palsy (BP) patients and compare the facial nerve/facial canal width ratio (FN/FC) of the the affected and healthy sides to determine the entrapped sites. This retrospective study enrolled 47 patients with BP that were referred to a tertiary referral hospital. From patients' records, initial House-Brackmann grades, cranial computurized tomography (CT) and contrast enhanced temporal magnetic resonance imaging (MRI) scans were revieved. FC and FN diameter measurements were performed at the midpoint of labyrinthine segment, geniculate ganglion and tympanic segment in the axial plane, and mastoid segment in the sagittal plane. Contrast-enhanced segments on MRI were evaluated. There was no statistically significant difference between the FC diameters of the affected and the healthy sides on CT at any measured segments. FN diameter was increased significantly for all segments except mastoid segment on the affected side compared to the healthy side. FN/FC ratio was significantly greater on the affected side than the healthy side in all measured segments. Evident contrast enhancement was observed in all affected FNs, with geniculate ganglion and labyrinthine segment being the most common sites with contrast enhancement on MRI. Anatomical differences in FC may not be the main predisposing factor in BP. As the geniculate ganglion and labyrinthine segments had the highest FN/FC ratio and showed clear contrast enhancement on MRI, these findings may indicate a higher risk of inflammation and entrapment in these segments.

Impact of a Sedation Reduction Protocol in Infants Undergoing MRI Scanning.

Brain magnetic resonance imaging (MRI) is an important diagnostic tool for infants with possible brain abnormalities. While sedation may be necessary for high quality images, it carries risks of complications. The sedation rate for MRI procedure varies widely, ranging from 0% to 100%, influenced by infant characteristics and institutional practices, with an increasing focus on non-sedated or minimally sedated approaches to reduce risks. We studied sedation utilization in infants undergoing MRI scanning before and after implementing an MRI bundle. This cohort study utilized a pre- post-intervention design. An MRI bundle, including a process map, a safety checklist and a questionnaire collecting detailed information on sedation, were developed for our off-unit MRI suite. Pre-intervention group included infants scanned March 2018 to February 2019, and Post-intervention group March 2019 to February 2022. We hypothesized that sedation rates would significantly decrease following the intervention. In the study, 229 infants in the Pre-Intervention group and 764 infants in the Post-Intervention group underwent MRI scanning. Sedation use decreased by 62%, from 29% pre-intervention to 18% post-intervention (P =0.0003). Post-intervention infants were 47.6% less likely to be sedated, adjusting for gestation-corrected age (OR 0.524 [0.369, 0.745]; P <0.01). Each 1-week increase in gestation-corrected age was associated with a 7.1% increase in the odds of sedation, controlling for the intervention time-period (OR 1.071 [1.022, 1.122]; P =0.004). The questionnaire was completed 72% of the time in the post-intervention group. A standardized approach and protocol development can significantly reduce sedation for neonatal MRI. This study offers guidance for future research and integrated care interventions across medical teams.

Incidence of temporal thumb sign in adult patients with idiopathic seizures with and without elevated opening cerebrospinal fluid pressure.

The temporal thumb sign (TTS) has emerged as a potential distinct magnetic resonance imaging (MRI) finding in pediatric patients with new-onset idiopathic seizures which is characterized by the protrusion of the inferior temporal gyrus into the skull base. This study investigated the prevalence and clinical implications of TTS in adult patients with idiopathic seizures, with or without elevated opening cerebrospinal fluid (CSF) pressure, and those diagnosed with idiopathic intracranial hypertension (IIH) without seizures. A retrospective study was conducted on 157 adults aged 18-40-years-old, divided into four distinct groups. Group 1 consisted of 52 patients diagnosed with idiopathic intracranial hypertension (IIH) without seizures. Group 2 included 27 patients with seizures and elevated opening CSF pressure. Group 3 comprised 39 patients with seizures but normal opening CSF pressure. Group 4 consisted of 39 controls. MRI or CT scans and lumbar puncture data were analyzed to evaluate the presence of TTS. Statistical analysis was performed to assess the prevalence, sensitivity, and specificity of TTS across groups. Significant differences were defined as p < 0.05. Significant differences in TTS prevalence were observed between all groups (p < 0.001). TTS was most frequently detected in patients with seizures and elevated opening pressure (92.6%), followed by IIH patients without seizures (63.5%), and patients with seizures and normal opening pressure (61.5%). The control group exhibited a lower frequency of TTS (12.8%). Logistic regression demonstrated the association of TTS with seizure and elevated intracranial pressure (p < 0.001). Inter-rater reliability analysis showed excellent agreement between observers. The results of this study suggest that the Kamali "Temporal Thumb Sign" is a novel neuroimaging feature associated with seizures and elevated opening CSF pressure in adult patients. High sensitivity (92.6%) of the TTS may yield it as a screening imaging marker in either brain CT or MRI scans in the emergency department, which could aid in suspecting possible elevated opening CSF pressure and proceeding with further workup. The association of TTSwith seizures suggests a potential link between elevated intracranial pressure and structural alterations in the skull basewithseizure presentation. The TTS may represent a distinct manifestation of intracranial hypertension, similar to temporal lobe encephaloceles. Question This study investigates the prevalence and clinical implications of TTS in adults with idiopathic seizures at a single institution. Findings TTS was detected in 92.6% of patients with seizures and elevated CSF pressure. Clinical relevance TTS may serve as a sensitive marker for identifying elevated intracranial pressure in patients presenting with seizures to the emergency department.

Diagnostic performance of contrast-enhanced CT combined with contrast-enhanced MRI for colorectal liver metastases: a case-control study

BackgroundColorectal liver metastases (CRLM) are a major determinant of prognosis in colorectal cancer (CRC) patients. Their early and accurate detection is essential for appropriate therapeutic planning and improving survival outcomes.PurposeTo evaluate the diagnostic capabilities of contrast-enhanced computed tomography (CT) and contrast-enhanced magnetic resonance imaging (MRI) in detecting colorectal liver metastases.Materials and methodsWe employed a case-control design to compare patients with histologically confirmed liver metastases against a control group without the condition. A total of 85 patients in each group were selected and retrospectively matched based on relevant factors. All subjects underwent both contrast-enhanced CT and MRI. The diagnostic performance of these imaging modalities was assessed by analysing sensitivity, specificity, positive and negative predictive values, and radiologists’ diagnostic confidence. Kappa statistics were used to evaluate inter-observer agreement. All MRI scans were performed using a 3-Tesla (3-T) MRI scanner to ensure high-quality imaging and detailed lesion characterization. And all the scans were reviewed by two radiologists.ResultsThe combination of contrast-enhanced CT and MRI demonstrated a statistically significant improvement in sensitivity (90.6% for MRI alone vs. 96.5% for combined modalities) and specificity (95.3% for MRI alone vs. 98.3% for combined modalities). Positive and negative predictive values were similarly enhanced. Radiologists’ diagnostic confidence was higher with combined imaging, achieving a ‘very high’ confidence level in 78.8% of cases compared with 64.7% for MRI alone. The inter-observer agreement reached ‘almost perfect’ status with the combined approach.ConclusionThe integration of contrast-enhanced CT with MRI significantly enhanced the diagnostic accuracy for colorectal liver metastases, representing a valuable tool for the preoperative evaluation of patients with CRC.Clinical trial numberNot applicable.

Morphometric Characterization of Levator Ani Subdivisions in Healthy Controls and Patients: An MRI Study Using 3D PICS.

To date, levator ani muscle (LAM) morphometry has been classified descriptively and semi-quantitatively. New MRI techniques enabling detailed visualization with the 3D pelvic inclination correction system (3D PICS) could offer a one-stop-shop diagnostic modality for quantitative assessment of LAM subdivisions. The aim of this controlled MRI study was to assess morphometric LAM subdivision characteristics in two distinct groups of premenopausal women, namely nulliparous asymptomatic controls and symptomatic patients (Pelvic Organ Prolapse Quantification [POP-Q] ≥ II). Magnetic resonance imaging scans of the 22 women in each group were analyzed applying the 3D PICS coordinate system. A second reading of MRI was used to calculate interrater reliability (IRR). Origins and insertions were expressed in the 3D-Cartesian coordinate system in relation to point 0/0/0 (inferior pubic point). Distances and angles between muscles and planes were described using mean and standard deviation or median with first and third quartiles for all LAM subdivisions. Moderate to good IRR was reported except for points close to point 0/0/0. Origins showed no difference between groups. Insertions differed notably in the vertically oriented pubovaginal, puboperineal, and puboanal muscles, with patients exhibiting lower positions along the superior-inferior axis by 6.1-7.7, 8.8, and 8.0-8.2mm respectively. In contrast, the insertions of the horizontally oriented puborectal muscle showed a smaller difference of 1.8mm. Muscle lengths were also 4% to 24% longer in cases. This in vivo MRI study reveals first geometric 3D data on LAM morphology in 3D PICS for both cases and controls. Exact 3D coordinates of origin/insertion points, lengths, and angles could serve as a basis for future imaging-based POP diagnostics.

Identification of key brain networks and functional connectivities of successful aging: A surface-based resting-state functional magnetic resonance study.

Successful aging (SA) refers to the ability to maintain high levels of physical, cognitive, psychological, and social engagement in old age, with high cognitive function being the key to achieving SA. To explore the potential characteristics of the brain network and functional connectivity (FC) of SA. Twenty-six SA individuals and 47 usual aging individuals were recruited from community-dwelling elderly, which were taken the magnetic resonance imaging scan and the global cognitive function assessment by Mini Mental State Examination (MMSE). The resting state-functional magnetic resonance imaging data were preprocessed by DPABISurf, and the brain functional network was conducted by DPABINet. The support vector machine model was constructed with altered functional connectivities to evaluate the identification value of SA. The results found that the 6 inter-network FCs of 5 brain networks were significantly altered and related to MMSE performance. The FC of the right orbital part of the middle frontal gyrus and right angular gyrus was mostly increased and positively related to MMSE score, and the FC of the right supramarginal gyrus and right temporal pole: Middle temporal gyrus was the only one decreased and negatively related to MMSE score. All 17 significantly altered FCs of SA were taken into the support vector machine model, and the area under the curve was 0.895. The identification of key brain networks and FC of SA could help us better understand the brain mechanism and further explore neuroimaging biomarkers of SA.

Magnetic resonance imaging study of acromial bony erosion after arthroscopic repair of partial-thickness and small full-thickness rotator cuff tears

Magnetic resonance imaging study of acromial bony erosion after arthroscopic repair of partial-thickness and small full-thickness rotator cuff tears

Using Deep Learning to Perform Automatic Quantitative Measurement of Masseter and Tongue Muscles in Persons With Dementia: Cross-Sectional Study.

Sarcopenia (loss of muscle mass and strength) increases adverse outcomes risk and contributes to cognitive decline in older adults. Accurate methods to quantify muscle mass and predict adverse outcomes, particularly in older persons with dementia, are still lacking. This study's main objective was to assess the feasibility of using deep learning techniques for segmentation and quantification of musculoskeletal tissues in magnetic resonance imaging (MRI) scans of the head in patients with neurocognitive disorders. This study aimed to pave the way for using automated techniques for opportunistic detection of sarcopenia in patients with neurocognitive disorder. In a cross-sectional analysis of 53 participants, we used 7 U-Net-like deep learning models to segment 5 different tissues in head MRI images and used the Dice similarity coefficient and average symmetric surface distance as main assessment techniques to compare results. We also analyzed the relationship between BMI and muscle and fat volumes. Our framework accurately quantified masseter and subcutaneous fat on the left and right sides of the head and tongue muscle (mean Dice similarity coefficient 92.4%). A significant correlation exists between the area and volume of tongue muscle, left masseter muscle, and BMI. Our study demonstrates the successful application of a deep learning model to quantify muscle volumes in head MRI in patients with neurocognitive disorders. This is a promising first step toward clinically applicable artificial intelligence and deep learning methods for estimating masseter and tongue muscle and predicting adverse outcomes in this population.

Diagnosis of Metastatic Breast Tumor with an Iron-Based Hydrogen-Bonded Organic Framework via T2-Weighted Magnetic Resonance Imaging.

Magnetic resonance imaging (MRI) often employs contrast agents (CAs) to improve the visualization of lesions. Although iron-based oxides have been clinically approved as T2 CAs, various obstacles have hindered their widespread commercial use. Consequently, there is a pressing demand for innovative T2-type CAs. Herein, we synthesized an iron-based hydrogen-bonded organic framework (Fe-HOF) from Fe-TCPP and explored its potential as a T2-weighted MRI CA. The Fe-HOF demonstrated a superior relaxivity (r2) of 32.067 mM-1 s-1 and a higher r2/r1 ratio of 45.25 compared to Fe-TCPP. This enhancement may be attributed to the combination of the single-atom form of Fe3+ with its increased radius. Our findings indicate that a 6 μmol [Fe]/kg dose of Fe-HOF significantly improves lesion contrast in T2-weighted MRI scans of subcutaneous tumor model mice and liver metastasis model mice of breast tumor. The simplicity of Fe-HOF' s structure ensures the absence of complex metal ions or ligands during synthesis, and the iron component can be metabolized into the endogenous iron pool, resulting in remarkable biocompatibility and biosafety. These findings pave the way for the design of novel T2-weighted MRI probes tailored for cancer characterization at various stages.

Correlation between AI algorithm generated radiographic hip measurements for hip dysplasia patients and internal derangement findings on advanced MRI.

BackgroundHip dysplasia (HD) involves abnormal acetabular development, resulting in reduced femoral head coverage, labral tears, and cartilage injury. Machine learning (AI) advancements have enabled reproducible radiographic measurements for HD, such as lateral center edge angle (LCEA), Tonnis, and extrusion index. Moreover, incorporating 3D magnetic resonance imaging (MRI) alongside 2D MRI enhances diagnostic capabilities.PurposeTo correlate advanced MRI-assessed labral and cartilage injuries with validated AI-generated radiographic measurements.Material and MethodsThis study enrolled 139 patients (age range = 16-68 years) with HD, comprising a total of 156 hips. All patients had 2D and 3D MRI scans, four-view X-rays, and AI-generated radiographic measurements using a commercial AI program that utilized bony landmarks to generate measurements. Labral reconstructions were obtained for each hip, and a multi-reader study was conducted. Inter-reader (ICC) analysis and Spearman correlations were calculated.ResultsThe predominant location for the largest labral tear was anterosuperior (133/156, 90%), and paralabral cysts were observed in 53/156 (34%) cases. No statistically significant correlations were found between the length of labral tears and radiographic measurements. However, statistically significant correlations were observed between paralabral cysts and femoral head coverage, extrusion index, LCEA, and Tonnis measurements.ConclusionAI-generated radiographic measurements in HD exhibited weak correlations with advanced MRI findings, likely due to the condition's complex pathophysiology.

Prediction of preoperative tumor-related epilepsy using XGBoost radiomics models with 4 MRI sequences

Introduction. Tumor-related epilepsy is a prevalent condition in patients with gliomas. Accurate prediction of epilepsy is crucial for early treatment. This study aimed to evaluate the novel application of the eXtreme Gradient Boost (XGBoost) machine learning (ML) algorithm into a radiomics model predicting preoperative tumor-related epilepsy (PTRE). Its performance was compared with 4 conventional ML algorithms, including the least absolute shrinkage and selection operator (LASSO), elastic net, random forest, and support vector machine.Methods.This study used four magnetic resonance imaging (MRI) images consisting of four sequences (T1-weighted [T1W], T1-weighted contrast [T1WC], T2-weighted [T2W], and T2-weighted fluid-attenuated inversion recovery [T2W FLAIR]) acquired from 74 glioma patients, 30 with PTRE and 44 without PTRE. 394 radiomics features were extracted from the MRI scans usingPyradiomics, alongside 12 clinical features from the medical records. The ML algorithms were mixed and matched to create 20 radiomics models with two stages for: (1) feature selection and (2) prediction of PTRE. Nested cross-validation was used to tune the algorithms and select the stable features.Results.The XGBoost radiomics model demonstrated the second-highest balanced accuracy and F1-score of 0.81 ± 0.01 and 0.80 ± 0.01 respectively. It also achieved the highest recall of 0.81 ± 0.02. It used mostly textural radiomics features from the T1W, T2W and T2W FLAIR sequences to make the predictions.Conclusion.This study demonstrates that XGBoost is a viable alternative to conventional ML algorithms for developing a radiomics model to predict PTRE, as the model produced from XGBoost had among the highest metrics. XGBoost selected features with a higher predictive value than other models. The features selected by XGBoost were more stable, which is a useful property for radiomics analysis. Features selected from multiple MRI sequences were important in the model's decision.

Stages prediction of Alzheimer’s disease with shallow 2D and 3D CNNs from intelligently selected neuroimaging data

Detection of Alzheimer’s Disease (AD) is critical for successful diagnosis and treatment, involving the common practice of screening for Mild Cognitive Impairment (MCI). However, the progressive nature of AD makes it challenging to identify its causal factors. Modern diagnostic workflows for AD use cognitive tests, neurological examinations, and biomarker-based methods, e.g., cerebrospinal fluid (CSF) analysis and positron emission tomography (PET) imaging. While these methods are effective, non-invasive imaging techniques like Magnetic Resonance Imaging (MRI) are gaining importance. Deep Learning (DL) approaches for evaluating alterations in brain structure have focused on combining MRI and Convolutional Neural Networks (CNNs) within the spatial architecture of DL. This combination has garnered significant research interest due to its remarkable effectiveness in automating feature extraction across various multilayer perceptron models. Despite this, MRI’s noisy and multidimensional nature requires an intelligent preprocessing pipeline for effective disease prediction. Our study aims to detect different stages of AD from the multidimensional neuroimaging data obtained through MRI scans using 2D and 3D CNN architectures. The proposed preprocessing pipeline comprises skull stripping, spatial normalization, and smoothing. It is followed by a novel and efficient pixel count-based frame selection and cropping approach, which renders a notable dimension reduction. Furthermore, the learnable resizer method is applied to enhance the image quality while resizing the data. Finally, the proposed shallow 2D and 3D CNN architectures extract spatio-temporal attributes from the segmented MRI data. Furthermore, we merged both the CNNs for further comparative analysis. Notably, 2D CNN achieved a maximum accuracy of 93%, while 3D CNN reported the highest accuracy of 96.5%.