Contrast-enhanced MRI Scans Research Articles

Differences in Molecular Subtype Reference Standards Impact AI-based Breast Cancer Classification with Dynamic Contrast-enhanced MRI.

Background Breast cancer tumors can be identified as different luminal molecular subtypes depending on either immunohistochemical (IHC) staining or St Gallen criteria that includes Ki-67. Purpose To characterize molecular subtypes and understand the impact of disagreement among IHC and St Gallen molecular subtype reference standards on artificial intelligence classification of luminal A and luminal B tumors with use of radiomic features extracted from dynamic contrast-enhanced (DCE) MRI scans. Materials and Methods In this retrospective study, 28 radiomic features previously extracted from DCE-MRI scans of breast tumors imaged between February 2015 and October 2017 were examined in the following groups: (a) tumors classified as luminal A by both reference standards ("agreement"), (b) tumors classified as luminal A by IHC and luminal B by St Gallen ("disagreement"), and (c) tumors classified as luminal B by both ("agreement"). Luminal A or luminal B tumor classification with use of radiomic features was conducted with use of three sets: (a) IHC molecular subtyping, (b) St Gallen molecular subtyping, and (c) agreement tumors. The Kruskal-Wallis test was followed by the Mann-Whitney U test to determine pair-wise differences of radiomic features among agreement and disagreement tumors. Fivefold cross-validation with use of stepwise feature selection and linear discriminant analysis classified tumors in each set, with performance measured with use of area under the receiver operating characteristic curve (AUC). Results A total of 877 breast cancer tumors from 872 women (mean age, 48 years [range, 19-75 years]) were analyzed. Six features (sphericity, irregularity, surface area to volume ratio, variance of radial gradient histogram, sum average, volume of most enhancing voxels) were different (P ≤ .001) among agreement and disagreement tumors. AUC (median, 0.74 [95% CI: 0.68, 0.80]) was higher than when using tumors subtyped by either reference standard (IHC, 0.66 [0.60, 0.71], P = .003; St Gallen, 0.62 [0.58, 0.67], P = .001). Conclusion Differences in reference standards can hinder artificial intelligence classification performance of luminal molecular subtypes with dynamic contrast-enhanced MRI. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Bae in this issue.

Brain Tumor Classification Using Convolutional Neural Network and Feature Extraction

Convolutional Neural Networks (CNNs) are the most popular neural network model for the image classification problem, which has seen a surge in interest in recent years thanks to its potential to improve medical picture classification. CNN employs a number of building pieces, including convolution layers, pooling layers, and fully connected layers, to determine features in an adaptive manner via backpropagation. In this study, we aimed to create a CNN model that could identify and categorize brain cancers in T1-weighted contrast-enhanced MRI scans. There are two main phases to the proposed system. To identify images using CNN, first they must be preprocessed using a variety of image processing techniques. A total of 3064 photos of glioma, meningioma, and pituitary tumors are used in the investigation. Testing accuracy for our CNN model was 94.39%, precision was 93.33%, and recall was 93% on average. The suggested system outperformed numerous well-known current algorithms and demonstrated satisfactory accuracy on the dataset. We have performed several procedures on the data set to get it ready for usage, including standardizing the pixel sizes of the photos and dividing the dataset into 80% for train, 10% for test, and 10% for validation. The proposed classifier achieves a high level of accuracy of 95.3%.

Deep Learning to Simulate Contrast-enhanced Breast MRI of Invasive Breast Cancer.

Background There is increasing interest in noncontrast breast MRI alternatives for tumor visualization to increase the accessibility of breast MRI. Purpose To evaluate the feasibility and accuracy of generating simulated contrast-enhanced T1-weighted breast MRI scans from precontrast MRI sequences in biopsy-proven invasive breast cancer with use of deep learning. Materials and Methods Women with invasive breast cancer and a contrast-enhanced breast MRI examination that was performed for initial evaluation of the extent of disease between January 2015 and December 2019 at a single academic institution were retrospectively identified. A three-dimensional, fully convolutional deep neural network simulated contrast-enhanced T1-weighted breast MRI scans from five precontrast sequences (T1-weighted non-fat-suppressed [FS], T1-weighted FS, T2-weighted FS, apparent diffusion coefficient, and diffusion-weighted imaging). For qualitative assessment, four breast radiologists (with 3-15 years of experience) blinded to whether the method of contrast was real or simulated assessed image quality (excellent, acceptable, good, poor, or unacceptable), presence of tumor enhancement, and maximum index mass size by using 22 pairs of real and simulated contrast-enhanced MRI scans. Quantitative comparison was performed using whole-breast similarity and error metrics and Dice coefficient analysis of enhancing tumor overlap. Results Ninety-six MRI examinations in 96 women (mean age, 52 years ± 12 [SD]) were evaluated. The readers assessed all simulated MRI scans as having the appearance of a real MRI scan with tumor enhancement. Index mass sizes on real and simulated MRI scans demonstrated good to excellent agreement (intraclass correlation coefficient, 0.73-0.86; P < .001) without significant differences (mean differences, -0.8 to 0.8 mm; P = .36-.80). Almost all simulated MRI scans (84 of 88 [95%]) were considered of diagnostic quality (ratings of excellent, acceptable, or good). Quantitative analysis demonstrated strong similarity (structural similarity index, 0.88 ± 0.05), low voxel-wise error (symmetric mean absolute percent error, 3.26%), and Dice coefficient of enhancing tumor overlap of 0.75 ± 0.25. Conclusion It is feasible to generate simulated contrast-enhanced breast MRI scans with use of deep learning. Simulated and real contrast-enhanced MRI scans demonstrated comparable tumor sizes, areas of tumor enhancement, and image quality without significant qualitative or quantitative differences. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Slanetz in this issue. An earlier incorrect version appeared online. This article was corrected on January 17, 2023.

Autosegmentation of brain metastases using 3D FCNN models and methods to manage GPU memory limitations

Aims. To explore the efficacy of two different approaches to train a Fully Convolutional Neural Network (FCNN) with Graphical Processing Unit (GPU) memory limitations and investigate if pre-trained two-dimensional weights can be transferred into a three-dimensional model for the purpose of brain tumour segmentation. Materials and methods. Models were developed in Python using TensorFlow and Keras. T1 contrast-enhanced MRI scans and associated contouring data from 104 patients were used to train and validate the model. The data was resized to one-quarter of its original resolution, and the original data was also split into four quarters for comparison to fit within GPU limitations. Transferred weights from a two-dimensional VGG16 model trained on ImageNet were transformed into three dimensions for comparison with randomly generated initial weights. Results. Resizing the data produced superior dice similarity coefficients with fewer false positives than quartering the data. Quartering the data yielded a superior sensitivity. Transforming and transferring two-dimensional weights was not able to consistently produce improvement in training or final metrics. Conclusion. For segmentation of brain tumours, resizing the data results in better performance than quartering the data. For the model and approaches used in this report, transferring weights were not able to demonstrate any benefit.

Radiomics for prediction of response to EGFR-TKI based on metastasis/brain parenchyma (M/BP)-interface.

To evaluate the potential of subregional radiomics as a novel tumor marker in predicting epidermal growth factor receptor (EGFR) mutation status and response to EGFR-tyrosine kinase inhibitor (TKI) therapy in NSCLC patients with brain metastasis (BM). We included 230 patients from center 1, and 80 patients were included from center 2 to form a primary and external validation cohort, respectively. Patients underwent contrast-enhanced T1-weighted and T2-weighted MRI scans before treatment. The individual- and population-level clustering was used to partition the peritumoral edema area (POA) into phenotypically consistent subregions. Radiomics features were calculated and selected from the tumor active area (TAA), POA and subregions, and used to develop models. Prediction values of each region were investigated and compared with receiver operating characteristic curves and Delong test. For predicting EGFR mutations, a multi-region combined model (EGFR-Fusion) was developed based on joint of the partitioned metastasis/brain parenchyma (M/BP)-interface and TAA, and generated the highest prediction performance in the training (AUC = 0.945, SEN = 0.878, SPE = 0.937), internal validation (AUC = 0.880, SEN = 0.733, SPE = 0.969), and external validation (AUC = 0.895, SEN = 0.875, SPE = 0.800) cohorts. For predicting response to EGFR-TKI, the developed multi-region combined model (TKI-Fusion) yielded predictive AUCs of 0.869 (SEN = 0.717, SPE = 0.884), 0.786 (SEN = 0.708, SPE = 0.818), and 0.802 (SEN = 0.750, SPE = 0.800) in the training, internal validation and external validation cohort, respectively. Our study revealed that complementary information regarding the EGFR status and response to EGFR-TKI can be provided by subregional radiomics. The proposed radiomics models may be new markers to guide treatment plans for NSCLC patients with BM.

Tayside Screening For Cardiac Events (TASCFORCE) study: a prospective cardiovascular risk screening study

PurposeRisk factor-based models struggle to accurately predict the development of cardiovascular disease (CVD) at the level of the individual. Ways of identifying people with low predicted risk who will develop...

P01.06.B Interim results from CAR-Study B: An ongoing randomized trial on the effect of SRS or WBRT on cognitive performance in patients with 11-20 brain metastases

Abstract Background Both stereotactic radiosurgery (SRS) and whole brain radiation therapy (WBRT) have proven to be effective treatments for multiple brain metastases (BM) with similar overall survival. Cognition and Radiation (CAR) Study B is a randomized trial on the effect of Gamma Knife radiosurgery (GKRS) or WBRT on cognitive performance in patients with 11-20 BM. The primary and secondary aim of this interim analysis were to check whether Bayesian stopping rules for cognitive failure were met, and to compare cognitive changes after treatment respectively, for the first 45 patients enrolled. Material and Methods Patients with 11-20 newly diagnosed BM on a triple-dose contrast-enhanced MRI-scan, expected survival &amp;gt;3 months and Karnofsky Performance Status (KPS) ≥70, were stratified by age, histology, total BM volume, systemic treatment, KPS, and baseline Hopkins Verbal Learning total recall (HVLT-R TR) score, and randomized 1:1 (minimization) to GKRS or WBRT. Neuropsychological tests were administered before (T0) treatment (n=21 vs n=20), and at 3 (T3; n=16 vs n=14) and 6 (T6; n=9 vs n=9) months thereafter. A decline of ≥5 points in HVLT-R TR score was considered a cognitive failure. The trial would be halted if the posterior probability for a higher cognitive failure rate in one group versus the other was &amp;gt;0·975 at T3 or T6 according to the employed beta (2.09,2.91) prior (prior mean of 42%), based on the average failure rates at 4 months reported by Chang et al. (2009). Between-group differences in changes of test performances over 6 months were analyzed using mixed ANOVAs. Proportions of cognitive changes (T0-T6) at the individual level based on reliable change indices correcting for practice effects, were determined. Results HVLT-R TR failure rates in the GKRS versus WBRT group were 31% versus 29% at T3, and 0% versus 33% at T6. The observed failure rates after WBRT at T3 and T6 were lower than the average failure rates of Chang et al. (2009). Posterior probabilities were 0·451 at T3 and 0·918 at T6. Over 6 months, changes in performance on tests of immediate (p=·003) and delayed recall (p=·024), and information processing speed (p=·003) were significantly different between groups (large effect sizes), with significant declines after WBRT, but not after GKRS. Over 6 months, at the individual patient level, there were no declines in performances across all tests in the GKRS group (n=8) while performances declined in 4 out of 8 patients in the WBRT group. Conclusion The stopping rules were not met since the posterior probabilities did not cross the threshold. Other preliminary findings in this small sample suggest that cognitive decline, both at group and individual level, is more pronounced after WBRT compared to GKRS. Accrual is continued (NCT02953717; ZonMw 842003006).

A deep learning approach with subregion partition in MRI image analysis for metastatic brain tumor.

PurposeTo propose a deep learning network with subregion partition for predicting metastatic origins and EGFR/HER2 status in patients with brain metastasis.MethodsWe retrospectively enrolled 140 patients with clinico-pathologically confirmed brain metastasis originated from primary NSCLC (n = 60), breast cancer (BC, n = 60) and other tumor types (n = 20). All patients underwent contrast-enhanced brain MRI scans. The brain metastasis was subdivided into phenotypically consistent subregions using patient-level and population-level clustering. A residual network with a global average pooling layer (RN-GAP) was proposed to calculate deep learning-based features. Features from each subregion were selected with least absolute shrinkage and selection operator (LASSO) to build logistic regression models (LRs) for predicting primary tumor types (LR-NSCLC for the NSCLC origin and LR-BC for the BC origin), EGFR mutation status (LR-EGFR) and HER2 status (LR-HER2).ResultsThe brain metastasis can be partitioned into a marginal subregion (S1) and an inner subregion (S2) in the MRI image. The developed models showed good predictive performance in the training (AUCs, LR-NSCLC vs. LR-BC vs. LR-EGFR vs. LR-HER2, 0.860 vs. 0.909 vs. 0.850 vs. 0.900) and validation (AUCs, LR-NSCLC vs. LR-BC vs. LR-EGFR vs. LR-HER2, 0.819 vs. 0.872 vs. 0.750 vs. 0.830) set.ConclusionOur proposed deep learning network with subregion partitions can accurately predict metastatic origins and EGFR/HER2 status of brain metastasis, and hence may have the potential to be non-invasive and preoperative new markers for guiding personalized treatment plans in patients with brain metastasis.

Contrast-enhanced MRI for T Restaging of Locally Advanced Rectal Cancer Following Neoadjuvant Chemotherapy and Radiation Therapy.

Background Accurate restaging of rectal cancer is crucial in the selection of candidates for local excision after neoadjuvant chemotherapy and radiation therapy (NCRT). The conventional approach of combined T2-weighted imaging and diffusion-weighted imaging (DWI) at MRI has been found to have limitations in restaging. Purpose To determine the diagnostic performance of contrast-enhanced MRI in distinguishing between pathologic stage ypT0-1 and ypT2-4 rectal cancer after NCRT compared with T2-weighted imaging and DWI by using surgical pathologic specimens as the reference standard. Materials and Methods This retrospective study evaluated MRI scans in all consecutive patients with locally advanced rectal cancer who underwent total mesorectal excision after NCRT in Peking University Cancer Hospital (Beijing, China) from January 2014 to October 2018. All MRI features obtained before and after NCRT were evaluated by two experienced radiologists, independently and blinded to personal, clinical, and histopathologic information. The post-NCRT yT stage was assigned based on high b value (b = 1000 sec/mm2) DWI with T2-weighted imaging (protocol 1) in the first round and on contrast-enhanced MRI scans (protocol 2) in a second round. The diagnostic accuracies for the differentiation of pathologic stage ypT0-1 from ypT2-4 tumors with the two protocols were compared. Multivariable regression analysis was used to explore the independent predictors of pathologic stage ypT0-1 tumors. Results A total of 328 patients (mean age, 57 years ± 10 [SD]; 227 men; 69%) were enrolled. The area under the receiver operating characteristic curve of the contrast-enhanced MRI protocol in predicting pathologic stage ypT0-1 tumors was 0.81 (95% CI: 0.77, 0.85), which was better than that of the T2-weighted DWI protocol (0.66; 95% CI: 0.60, 0.71; P < .001). Multivariable logistic regression analysis showed that yT stage after NCRT on contrast-enhanced MRI scans was the only independent predictor of pathologic stage ypT0-1 tumors (P < .001). Conclusion Contrast-enhanced MRI provides accurate differentiation of ypT0-1 from ypT2-4 tumors after neoadjuvant chemotherapy and radiation therapy. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Zins and Santiago in this issue.

Fully Automated MRI Segmentation and Volumetric Measurement of Intracranial Meningioma Using Deep Learning.

Accurate and rapid measurement of the MRI volume of meningiomas is essential in clinical practice to determine the growth rate of the tumor. Imperfect automation and disappointing performance for small meningiomas of previous automated volumetric tools limit their use in routine clinical practice. To develop and validate a computational model for fully automated meningioma segmentation and volume measurement on contrast-enhanced MRI scans using deep learning. Retrospective. A total of 659 intracranial meningioma patients (median age, 59.0 years; interquartile range: 53.0-66.0 years) including 554 women and 105 men. The 1.0 T, 1.5T, and 3.0 T; three-dimensional, T1 -weighted gradient-echo imaging with contrast enhancement. The tumors were manually segmented by two neurosurgeons, H.K. and C.-K.P., with 10 and 26 years of clinical experience, respectively, for use as the ground truth. Deep learning models based on U-Net and nnU-Net were trained using 459 subjects and tested for 100 patients from a single institution (internal validation set [IVS]) and 100 patients from other 24 institutions (external validation set [EVS]), respectively. The performance of each model was evaluated with the Sørensen-Dice similarity coefficient (DSC) compared with the ground truth. According to the normality of the data distribution verified by the Shapiro-Wilk test, variables with three or more categories were compared by the Kruskal-Wallis test with Dunn's post hoc analysis. A two-dimensional (2D) nnU-Net showed the highest median DSCs of 0.922 and 0.893 for the IVS and EVS, respectively. The nnU-Nets achieved superior performance in meningioma segmentation than the U-Nets. The DSCs of the 2D nnU-Net for small meningiomas less than 1cm3 were 0.769 and 0.780 with the IVS and EVS, respectively. A fully automated and accurate volumetric measurement tool for meningioma with clinically applicable performance for small meningioma using nnU-Net was developed. 3 TECHNICAL EFFICACY: Stage 2.

Deep Learning to Detect Triangular Fibrocartilage Complex Injury in Wrist MRI: Retrospective Study with Internal and External Validation.

Objective: To use deep learning to predict the probability of triangular fibrocartilage complex (TFCC) injury in patients’ MRI scans. Methods: We retrospectively studied medical records over 11 years and 2 months (1 January 2009–29 February 2019), collecting 332 contrast-enhanced hand MRI scans showing TFCC injury (143 scans) or not (189 scans) from a general hospital. We employed two convolutional neural networks with the MRNet (Algorithm 1) and ResNet50 (Algorithm 2) framework for deep learning. Explainable artificial intelligence was used for heatmap analysis. We tested deep learning using an external dataset containing the MRI scans of 12 patients with TFCC injuries and 38 healthy subjects. Results: In the internal dataset, Algorithm 1 had an AUC of 0.809 (95% confidence interval—CI: 0.670–0.947) for TFCC injury detection as well as an accuracy, sensitivity, and specificity of 75.6% (95% CI: 0.613–0.858), 66.7% (95% CI: 0.438–0.837), and 81.5% (95% CI: 0.633–0.918), respectively, and an F1 score of 0.686. Algorithm 2 had an AUC of 0.871 (95% CI: 0.747–0.995) for TFCC injury detection and an accuracy, sensitivity, and specificity of 90.7% (95% CI: 0.787–0.962), 88.2% (95% CI: 0.664–0.966), and 92.3% (95% CI: 0.763–0.978), respectively, and an F1 score of 0.882. The accuracy, sensitivity, and specificity for radiologist 1 were 88.9, 94.4 and 85.2%, respectively, and for radiologist 2, they were 71.1, 100 and 51.9%, respectively. Conclusions: A modified MRNet framework enables the detection of TFCC injury and guides accurate diagnosis.

Role of autoantibodies and blood–brain barrier leakage in cognitive impairment in systemic lupus erythematosus

ObjectiveCognitive impairment is common in patients with SLE but the cause is unknown. The current cross-sectional study examined the association between select SLE-related autoantibodies, other serological biomarkers and extensive blood–brain...

MRI Radiogenomics of Pediatric Medulloblastoma: A Multicenter Study.

Background Radiogenomics of pediatric medulloblastoma (MB) offers an opportunity for MB risk stratification, which may aid therapeutic decision making, family counseling, and selection of patient groups suitable for targeted genetic analysis. Purpose To develop machine learning strategies that identify the four clinically significant MB molecular subgroups. Materials and Methods In this retrospective study, consecutive pediatric patients with newly diagnosed MB at MRI at 12 international pediatric sites between July 1997 and May 2020 were identified. There were 1800 features extracted from T2- and contrast-enhanced T1-weighted preoperative MRI scans. A two-stage sequential classifier was designed-one that first identifies non-wingless (WNT) and non-sonic hedgehog (SHH) MB and then differentiates therapeutically relevant WNT from SHH. Further, a classifier that distinguishes high-risk group 3 from group 4 MB was developed. An independent, binary subgroup analysis was conducted to uncover radiomics features unique to infantile versus childhood SHH subgroups. The best-performing models from six candidate classifiers were selected, and performance was measured on holdout test sets. CIs were obtained by bootstrapping the test sets for 2000 random samples. Model accuracy score was compared with the no-information rate using the Wald test. Results The study cohort comprised 263 patients (mean age ± SD at diagnosis, 87 months ± 60; 166 boys). A two-stage classifier outperformed a single-stage multiclass classifier. The combined, sequential classifier achieved a microaveraged F1 score of 88% and a binary F1 score of 95% specifically for WNT. A group 3 versus group 4 classifier achieved an area under the receiver operating characteristic curve of 98%. Of the Image Biomarker Standardization Initiative features, texture and first-order intensity features were most contributory across the molecular subgroups. Conclusion An MRI-based machine learning decision path allowed identification of the four clinically relevant molecular pediatric medulloblastoma subgroups. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Chaudhary and Bapuraj in this issue.

341 SOncomagnetics: A Potential New Noninvasive Magnetic Stimulation Technology to Treat Glioblastoma

INTRODUCTION: A new noninvasive oscillating magnetic field (OMF)-generating device called Oncomagnetic (OM) device, developed in our laboratory, disrupts mitochondrial electron transport to elevate intracellular reactive oxygen species (ROS) levels in patient derived glioblastoma (GBM) cells. METHODS: GBM cell death, ROS elevation and caspase 3 activation were studied by bright-field and fluorescence microscopy involving specific fluorescent probes. Device treatment of FOXN1-Nude mice with GBM xenografts was for 4 hours twice a day. Treatment of a 53-year-old male patient with recurrent GBM was done under a compassionate use protocol for 2–6 hours each weekday. Contrast-enhanced MRI scans assessed tumor size. RESULTS: OMF caused selective death of GBM cells, but not normal human cortical neurons and astrocytes. It altered mitochondrial oxygen consumption, increased mitochondrial and cellular ROS, and activated caspase 3. The overall survival of xenografted mice was prolonged by 6 days with treatment. A 26-day device treatment was well tolerated by the patient and decreased contrast-enhanced tumor volume by &gt;30%. Patient’s autopsy after his death from closed head injury showed treatment effect, no tumor mass, and some evidence of active tumor. CONCLUSION: Noninvasive stimulation with the OM device is a potential powerful new treatment for GBM involving a novel mechanism of action.

O-RADS MRI Classification of Indeterminate Adnexal Lesions: Time-Intensity Curve Analysis Is Better Than Visual Assessment.

Background The MRI Ovarian-Adnexal Reporting and Data System (O-RADS) enables risk stratification of sonographically indeterminate adnexal lesions, partly based on time-intensity curve (TIC) analysis, which may not be universally available. Purpose To compare the diagnostic accuracy of visual assessment with that of TIC assessment of dynamic contrast-enhanced MRI scans to categorize adnexal lesions as benign or malignant and to evaluate the influence on the O-RADS MRI score. Materials and Methods The European Adnex MR Study Group, or EURAD, database, a prospective multicenter study of women undergoing MRI for indeterminate adnexal lesions between March 2013 and March 2018, was queried retrospectively. Women undergoing surgery for an adnexal lesion with solid tissue were included. Solid tissue enhancement relative to outer myometrium was assessed visually and with TIC. Contrast material washout was recorded. Lesions were categorized according to the O-RADS MRI score with visual and TIC assessment. Per-lesion diagnostic accuracy was calculated. Results A total of 320 lesions (207 malignant, 113 benign) in 244 women (mean age, 55.3 years ± 15.8 [standard deviation]) were analyzed. Sensitivity for malignancy was 96% (198 of 207) and 76% (157 of 207) for TIC and visual assessment, respectively. TIC was more accurate than visual assessment (86% [95% CI: 81, 90] vs 78% [95% CI: 73, 82]; P < .001) for benign lesions, predominantly because of higher specificity (95% [95% CI: 92, 98] vs 76% [95% CI: 68, 81]). A total of 21% (38 of 177) of invasive lesions were rated as low risk visually. Contrast material washout and high-risk enhancement (defined as earlier enhancement than in the myometrium) were highly specific for malignancy for both TIC (97% [95% CI: 91, 99] and 94% [95% CI: 90, 97], respectively) and visual assessment (97% [95% CI: 92, 99] and 93% [95% CI: 88, 97], respectively). O-RADS MRI score was more accurate with TIC than with visual assessment (area under the receiver operating characteristic curve, 0.87 [95% CI: 0.83, 0.90] vs 0.73 [95% CI: 0.68, 0.78]; P < .001). Conclusion Time-intensity curve analysis was more accurate than visual assessment for achieving optimal diagnostic accuracy with the Ovarian-Adnexal Reporting and Data System MRI score. Clinical trial registration no. NCT01738789 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Vargas and Woo in this issue. An earlier incorrect version appeared online. This article was corrected on March 7, 2022.

Hand and foot MRI in contemporary undifferentiated arthritis: in which patients is MRI valuable to detect rheumatoid arthritis early? A large prospective study.

ObjectivesIdentifying patients that will develop RA among those presenting with undifferentiated arthritis (UA) remains a clinical dilemma. Although MRI is helpful according to EULAR recommendations, this has only been determined in UA patients not fulfilling 1987 RA criteria, while some of these patients are currently considered as RA because they fulfil the 2010 criteria. Therefore, we studied the predictive value of MRI for progression to RA in the current UA population, i.e. not fulfilling RA classification criteria (either 1987 or 2010 criteria) and not having an alternate diagnosis. Additionally, the value of MRI was studied in patients with a clinical diagnosis of UA, regardless of the classification criteria.MethodsTwo UA populations were studied: criteria-based UA as described above (n = 405) and expert-opinion-based UA (n = 564), i.e. UA indicated by treating rheumatologists. These patients were retrieved from a large cohort of consecutively included early arthritis patients that underwent contrast-enhanced MRI scans of hand and foot at baseline. MRIs were scored for osteitis, synovitis and tenosynovitis. Patients were followed for RA development during the course of 1 year. Test characteristics of MRI were determined separately for subgroups based on joint involvement and autoantibody status.ResultsAmong criteria-based UA patients (n = 405), 21% developed RA. MRI-detected synovitis and MRI-detected tenosynovitis were predictive for progression to RA. MRI-detected tenosynovitis was independently associated with RA progression (odds ratio (OR) 2.79; 95% CI 1.40, 5.58), especially within ACPA-negative UA patients (OR 2.91; 95% CI 1.42, 5.96). Prior risks of RA development for UA patients with mono-, oligo- and polyarthritis were 3%, 19% and 46%, respectively. MRI results changed this risk most within the oligoarthritis subgroup: positive predictive value was 27% and negative predictive value 93%. Similar results were found in expert-opinion-based UA (n = 564).ConclusionThis large cohort study showed that MRI is most valuable in ACPA-negative UA patients with oligoarthritis; a negative MRI could aid in preventing overtreatment.

RETROSPECTIVE ANALYSIS OF FUNDUS CHANGES IN POST COVID RHINO ORBITAL CEREBRAL MUCORMYCOSIS - A CASE SERIES

Aim:To study the fundus changes in post COVID- Rhino-orbital-cerebral mucormycosis. Methods: The study was done by collecting data from 30 cases of ROCM admitted in GGH, Kurnool. Detailed history was taken along with systemic ,ENT, ophthalmic and neurological examination and all necessary investigations were done including contrast enhanced MRI. Treatment was started with systemic and retrobulbar amphotericin-B injections. Fundus pictures were taken. Results: All of them had history of infection with covid-19 dated about 3-5 weeks back. Among them 18 had corticosteroid administration, 12 had oxygen with nasal prongs/mask, 2 had high flow/non-invasive ventilation. All of them were diabetics and 21 were hypertensives.Most of them had orbital/facial pain &amp; edema, headache, 24 patients had proptosis, 16 had ptosis, 20 had ocular movement restriction,18 had loss of vision. In Contrast Enhanced MRI scan, 28 cases showed diffuse PNS involvement,4 had medial orbital involvement, 8 had diffuse orbital involvement,18 had involvement of orbital apex, 6 had CNS involvement. Fundus examination revealed optic atrophy in 15 cases, 5 had CRAO and 3 had CRVO,8 had diabetic retinopathy,4 had hypertensive retinopathy, others had no significant abnormality. Conclusion: Mucormycosis is a rapidly progressive angioinvasive fungal infection which has been on rise in India with the 2nd wave of COVID-19. Early diagnosis and management are essential to halt the spread of infection and prevent diminution of vision and therefore, further improve the visual outcome and overall prognosis of the patient.

The sternum: An important review area when reporting pre-surgical staging breast MRI studies.

Identification and assessment of extramammary findings on contrast-enhanced breast MRI scans is particularly important in the setting of newly diagnosed invasive cancer as metastatic lesions may be encountered in the liver, lungs, pleural cavity or bones. Establishing that stage IV disease is present has a profound effect on patient management. The sternum is routinely included on breast MRI studies and can be an early site for breast cancer metastases. These appear as enhancing lesions with high signal on fat-suppressed T2-weighted images. However, incidental benign lesions, notably haemangiomas, may also be encountered, and careful analysis is required to avoid false-positive results. Clinical context is important with a much lower likelihood of malignancy in the setting of routine screening of young women with no personal history of breast cancer. This pictorial essay illustrates findings encountered with lesions in the sternum and offers insights into how to interpret and manage them.

Mucormycosis in COVID-19 pandemic and its neurovascular spread.

PurposeRhinocerebral mucormycosis is a rapidly progressive angioinvasive fungal infection commonly seen in diabetics. In the COVID-19 pandemic we have witnessed a sudden surge in these cases. We aimed to evaluate the disease presentation, patterns of spread, and any association with the COVID-19 virus.MethodsThis prospective study was conducted on mucormycosis patients operated between March and July 2021. The diagnosis was confirmed either on KOH staining, fungal culture or histopathological examination.ResultsThirty one cases (21 males, 10 females) with a mean age of 53.3 years were included, of which 9 (29.1%) were COVID positive on presentation, 17 (54.8%) were post-COVID, while 5 (16.1%) had radiological evidence of COVID sequelae. Most common symptoms were cheek numbness (87.1%), headache (83.9%), visual disturbances (77.4%), and palate involvement (58.1%). Blackening of turbinates was uncommon (22.6%). Ethmoid sinus was involved in all patients. Pterygopalatine fossa involvement was present in 77.4%, and was accurately diagnosed on contrast enhanced MRI scan. There were 8 (25.8%) deaths, while the remaining are discharged or under treatment.ConclusionAn increase in the incidence of mucormycosis in the COVID-19 pandemic is probably due to a compromise in host immunity along with a synergistic effect in thrombotic microangiopathy. Spread of infection to the soft tissues of the infratemporal fossa, orbit or palate occur via neurovascular structures rather than by bone erosion. The pterygopalatine fossa is involved in most individuals.

Deep learning radiomic nomogram to predict recurrence in soft tissue sarcoma: a multi-institutional study.

To evaluate the performance of a deep learning radiomic nomogram (DLRN) model at predicting tumor relapse in patients with soft tissue sarcomas (STS) who underwent surgical resection. In total, 282 patients who underwent MRI and resection for STS at three independent centers were retrospectively enrolled. In addition, 113 of the 282 patients received additional contrast-enhanced MRI scans. We separated the participants into a development cohort and an external test cohort. The development cohort consisted of patients from one center and the external test cohort consisted of patients from two other centers. Two MRI-based DLRNs for prediction of tumor relapse after resection of STS were established. We universally tested the DLRNs and compared them with other prediction models constructed by using widespread adopted predictors (i.e., staging systems and Ki67) instead of radiomics features. The DLRN1 model incorporated plain MRI-based radiomics signature into the clinical data, and the DLRN2 model integrated radiomics signature extracted from plain and contrast-enhanced MRI with the clinical predictors. Across both study sets, the two MRI-based DLRNs had relatively better prognostic capability (C index ≥ 0.721 and median AUC ≥ 0.746; p < 0.05 compared with most other models and predictors) and less opportunity for prediction error (integrated Brier score ≤ 0.159). The decision curve analysis indicates that the DLRNs have greater benefits than staging systems, Ki67, and other models. We selected appropriate cutoff values for the DLRNs to divide STS recurrence into three risk strata (low, medium, and high) and calculated those groups' cumulative risk rates. The DLRNs were shown to be a reliable and externally validated tool for predicting STS recurrence by comparing with other prediction models. • The prediction of a high recurrence rate of STS before emergence of local recurrence can help to determine whether more active treatment should be implemented. • Two MRI-based DLRNs for prediction of tumor relapse were shown to be a reliable and externally validated tool for predicting STS recurrence. • We used the DLRNs to divide STS recurrence into three risk strata (low, medium, and high) to facilitate more targeted postoperative management in the clinic.