Contrast-enhanced T1-weighted MRI Research Articles

Defining tumor growth in vestibular schwannomas: a volumetric inter-observer variability study in contrast-enhanced T1-weighted MRI.

For patients with vestibular schwannomas (VS), a conservative observational approach is increasingly used. Therefore, the need for accurate and reliable volumetric tumor monitoring is important. Currently, a volumetric cutoff of 20% increase in tumor volume is widely used to define tumor growth in VS. The study investigates the tumor volume dependency on the limits of agreement (LoA) for volumetric measurements of VS by means of an inter-observer study. This retrospective study included 100 VS patients who underwent contrast-enhanced T1-weighted MRI. Five observers volumetrically annotated the images. Observer agreement and reliability was measured using the LoA, estimated using the limits of agreement with the mean (LOAM) method, and the intraclass correlation coefficient (ICC). The 100 patients had a median average tumor volume of 903 mm3 (IQR: 193-3101). Patients were divided into four volumetric size categories based on tumor volume quartile. The smallest tumor volume quartile showed a LOAM relative to the mean of 26.8% (95% CI: 23.7-33.6), whereas for the largest tumor volume quartile this figure was found to be 7.3% (95% CI: 6.5-9.7) and when excluding peritumoral cysts: 4.8% (95% CI: 4.2-6.2). Agreement limits within volumetric annotation of VS are affected by tumor volume, since the LoA improves with increasing tumor volume. As a result, for tumors larger than 200 mm3, growth can reliably be detected at an earlier stage, compared to the currently widely used cutoff of 20%. However, for very small tumors, growth should be assessed with higher agreement limits than previously thought.

Integrating lipid metabolite analysis with MRI-based transformer and radiomics for early and late stage prediction of oral squamous cell carcinoma

BackgroundOral Squamous Cell Carcinoma (OSCC) presents significant diagnostic challenges in its early and late stages. This study aims to utilize preoperative MRI and biochemical indicators of OSCC patients to predict the stage of tumors.MethodsThis study involved 198 patients from two medical centers. A detailed analysis of contrast-enhanced T1-weighted (ceT1W) and T2-weighted (T2W) MRI were conducted, integrating these with biochemical indicators for a comprehensive evaluation. Initially, 42 clinical biochemical indicators were selected for consideration. Through univariate analysis and multivariate analysis, only those indicators with p-values less than 0.05 were retained for model development. To extract imaging features, machine learning algorithms in conjunction with Vision Transformer (ViT) techniques were utilized. These features were integrated with biochemical indicators for predictive modeling. The performance of model was evaluated using the Receiver Operating Characteristic (ROC) curve.ResultsAfter rigorously screening biochemical indicators, four key markers were selected for the model: cholesterol, triglyceride, very low-density lipoprotein cholesterol and chloride. The model, developed using radiomics and deep learning for feature extraction from ceT1W and T2W images, showed a lower Area Under the Curve (AUC) of 0.85 in the validation cohort when using these imaging modalities alone. However, integrating these biochemical indicators improved the model’s performance, increasing the validation cohort AUC to 0.87.ConclusionIn this study, the performance of the model significantly improved following multimodal fusion, outperforming the single-modality approach.Clinical relevance statementThis integration of radiomics, ViT models, and lipid metabolite analysis, presents a promising non-invasive technique for predicting the staging of OSCC.

An automated metaheuristic-optimized approach for diagnosing and classifying brain tumors based on a convolutional neural network

Brain tumors must be classified to determine their severity and appropriate therapy. Applying Artificial Intelligence to medical imaging has enabled remarkable developments. The presented framework classifies patients with brain tumors with high accuracy and efficiency using CNN, pre-trained models, and the Manta Ray Foraging Optimization (MRFO) algorithm on X-ray and MRI images. Additionally, the CNN and Transfer Learning (TL) hyperparameters will be optimized through MRFO, resulting in improved performance of the pre-trained model. Two public datasets from Kaggle were used to build the models. The first dataset consists of two X-ray classes, while the 2nd dataset includes three contrast-enhanced T1-weighted MRI classes. First, a patient should be diagnosed as “Healthy” (or “Tumor”). When the scan returns the result “Healthy,” the patient has no abnormalities in their brain. If a scan reveals that the patient has a tumor, an MRI will be performed on them. After that, the type of tumor (i.e., meningioma, pituitary, and glioma) will be identified using the second proposed classifier. A comparative analysis of the models used in the two-class dataset showed that VGG16's pre-trained model outperformed other models. Besides, the Xception pre-trained model achieved the best results in the three-class dataset. A manual review of misclassifications was conducted to determine the reasons for the misclassifications and correct them. The evaluation of the suggested architecture yielded an accuracy of 99.96% for X-rays and 98.64% for T1-weighted contrast-enhanced MRIs. The proposed deep learning framework outperformed most current deep learning models.

DIPG-73. FOCUSED ULTRASOUND FOR TREATMENT OF CHILDREN DIAGNOSED WITH DIFFUSE MIDLINE GLIOMAS

Abstract BACKGROUND Diffuse midline glioma (DMG) is a highly aggressive and fatal brain tumor that predominantly impacts children and young adults. The infiltrative pattern and rapid growth characteristics of DMG causes a significant challenge in finding an effective treatment. Additionally, the blood-brain barrier (BBB) poses a significant challenge by restricting the passage of therapeutic drugs to tumor bed. Currently the established standard of care is radiotherapy (RT), which is transitory. Thus there is a need for improved treatment options. OBJECTIVE Focused ultrasound (FUS) is an advancing non-invasive technology that has been used to open BBB and enhance drug delivery. We hypothesize that a combination therapy of FUS with pharmacological agents will be an effective and non-invasive treatment option for children diagnosed with DMG. METHOD Syngeneic DMG murine models were used to study the combination therapy of FUS with ONC201, FUS with anti-PD1 and FUS with RT. FUS mediated BBB opening was monitored with contrast-enhanced T1-weighted MRI. FUS-mediated drug delivery was validated with western blot, assessment of ROS, and optical label imaging. Kaplan Meier analysis was used to assess survival benefits. Single cell RNA-sequencing (scRNA-Seq) analysis is being done to assess the treatment effects of FUS and RT combination therapy. RESULTS/CONCLUSION We found that FUS and ONC201 combination therapy resulted in a decrease in NDUFA12 protein (a biomarker of ONC201 response) and an increase in ROS production compared to ONC201 monotherapy. For FUS and anti-PD1 combination therapy, an increased anti-PD1 delivery to the brainstem was detected. The overall survival of mice treated with FUS and anti-PD1 combination therapy increased when compared to anti-PD1 alone. Our preliminary scRNA-Seq analysis showed that combination therapy of FUS and RT induces innate immune response compared to the untreated group. Combination therapy with focused ultrasound shows a promising outcome by increasing the efficacy of therapeutic drugs and immunotherapy.

Radiomics signatures for predicting the Ki-67 level and HER-2 status based on bone metastasis from primary breast cancer.

This study explores the potential of radiomics to predict the proliferation marker protein Ki-67 levels and human epidermal growth factor receptor 2 (HER-2) status based on MRI images of patients with spinal metastasis from primary breast cancer. A total of 110 patients with pathologically confirmed spinal metastases from primary breast cancer were enrolled between Dec. 2017 and Dec. 2021. All patients underwent T1-weighted contrast-enhanced MRI scans. The PyRadiomics package was used to extract features from the MRI images based on the intraclass correlation coefficient and least absolute shrinkage and selection operator. The most predictive features were used to develop the radiomics signature. The Chi-Square test, Fisher's exact test, Student's t-test, and Mann-Whitney U test were used to evaluate the clinical and pathological characteristics between the high- and low-level Ki-67 groups and the HER-2 positive/negative groups. The radiomics models were compared using receiver operating characteristic curve analysis. The area under the receiver operating characteristic curve (AUC), sensitivity (SEN), and specificity (SPE) were generated as comparison metrics. From the spinal MRI scans, five and two features were identified as the most predictive for the Ki-67 level and HER-2 status, respectively. The developed radiomics signatures generated good prediction performance for the Ki-67 level in the training (AUC = 0.812, 95% CI: 0.710-0.914, SEN = 0.667, SPE = 0.846) and validation (AUC = 0.799, 95% CI: 0.652-0.947, SEN = 0.722, SPE = 0.833) cohorts. Good prediction performance for the HER-2 status was also achieved in the training (AUC = 0.796, 95% CI: 0.686-0.906, SEN = 0.720, SPE = 0.776) and validation (AUC = 0.705, 95% CI: 0.506-0.904, SEN = 0.733, SPE = 0.762) cohorts. The results of this study provide a better understanding of the potential clinical implications of spinal MRI-based radiomics on the prediction of Ki-67 levels and HER-2 status in breast cancer.

Diagnostic Accuracy of MR Spectroscopic Imaging and 18F-FET PET for Identifying Glioma: A Biopsy-Controlled Hybrid PET/MRI Study.

Contrast-enhanced MRI is the method of choice for brain tumor diagnostics, despite its low specificity for tumor tissue. This study compared the contribution of MR spectroscopic imaging (MRSI) and amino acid PET to improve the detection of tumor tissue. Methods: In 30 untreated patients with suspected glioma, O-(2-[18F]fluoroethyl)-l-tyrosine (18F-FET) PET; 3-T MRSI with a short echo time; and fluid-attenuated inversion recovery, T2-weighted, and contrast-enhanced T1-weighted MRI were performed for stereotactic biopsy planning. Serial samples were taken along the needle trajectory, and their masks were projected to the preoperative imaging data. Each sample was individually evaluated neuropathologically. 18F-FET uptake and the MRSI signals choline (Cho), N-acetyl-aspartate (NAA), creatine, myoinositol, and derived ratios were evaluated for each sample and classified using logistic regression. The diagnostic accuracy was evaluated by receiver operating characteristic analysis. Results: On the basis of the neuropathologic evaluation of tissue from 88 stereotactic biopsies, supplemented with 18F-FET PET and MRSI metrics from 20 areas on the healthy-appearing contralateral hemisphere to balance the glioma/nonglioma groups, 18F-FET PET identified glioma with the highest accuracy (area under the receiver operating characteristic curve, 0.89; 95% CI, 0.81-0.93; threshold, 1.4 × background uptake). Among the MR spectroscopic metabolites, Cho/NAA normalized to normal brain tissue showed the highest diagnostic accuracy (area under the receiver operating characteristic curve, 0.81; 95% CI, 0.71-0.88; threshold, 2.2). The combination of 18F-FET PET and normalized Cho/NAA did not improve the diagnostic performance. Conclusion: MRI-based delineation of gliomas should preferably be supplemented by 18F-FET PET.

Radiomic features define risk and are linked to DNA methylation attributes in primary CNS lymphoma.

The prognostic roles of clinical and laboratory markers have been exploited to model risk in patients with primary CNS lymphoma, but these approaches do not fully explain the observed variation in outcome. To date, neuroimaging or molecular information is not used. The aim of this study was to determine the utility of radiomic features to capture clinically relevant phenotypes, and to link those to molecular profiles for enhanced risk stratification. In this retrospective study, we investigated 133 patients across 9 sites in Austria (2005-2018) and an external validation site in South Korea (44 patients, 2013-2016). We used T1-weighted contrast-enhanced MRI and an L1-norm regularized Cox proportional hazard model to derive a radiomic risk score. We integrated radiomic features with DNA methylation profiles using machine learning-based prediction, and validated the most relevant biological associations in tissues and cell lines. The radiomic risk score, consisting of 20 mostly textural features, was a strong and independent predictor of survival (multivariate hazard ratio = 6.56 [3.64-11.81]) that remained valid in the external validation cohort. Radiomic features captured gene regulatory differences such as in BCL6 binding activity, which was put forth as testable treatment target for a subset of patients. The radiomic risk score was a robust and complementary predictor of survival and reflected characteristics in underlying DNA methylation patterns. Leveraging imaging phenotypes to assess risk and inform epigenetic treatment targets provides a concept on which to advance prognostic modeling and precision therapy for this aggressive cancer.

Time-to-event overall survival prediction in glioblastoma multiforme patients using magnetic resonance imaging radiomics

PurposeGlioblastoma Multiforme (GBM) represents the predominant aggressive primary tumor of the brain with short overall survival (OS) time. We aim to assess the potential of radiomic features in predicting the time-to-event OS of patients with GBM using machine learning (ML) algorithms.Materials and methodsOne hundred nineteen patients with GBM, who had T1-weighted contrast-enhanced and T2-FLAIR MRI sequences, along with clinical data and survival time, were enrolled. Image preprocessing methods included 64 bin discretization, Laplacian of Gaussian (LOG) filters with three Sigma values and eight variations of Wavelet Transform. Images were then segmented, followed by the extraction of 1212 radiomic features. Seven feature selection (FS) methods and six time-to-event ML algorithms were utilized. The combination of preprocessing, FS, and ML algorithms (12 × 7 × 6 = 504 models) was evaluated by multivariate analysis.ResultsOur multivariate analysis showed that the best prognostic FS/ML combinations are the Mutual Information (MI)/Cox Boost, MI/Generalized Linear Model Boosting (GLMB) and MI/Generalized Linear Model Network (GLMN), all of which were done via the LOG (Sigma = 1 mm) preprocessing method (C-index = 0.77). The LOG filter with Sigma = 1 mm preprocessing method, MI, GLMB and GLMN achieved significantly higher C-indices than other preprocessing, FS, and ML methods (all p values < 0.05, mean C-indices of 0.65, 0.70, and 0.64, respectively).ConclusionML algorithms are capable of predicting the time-to-event OS of patients using MRI-based radiomic and clinical features. MRI-based radiomics analysis in combination with clinical variables might appear promising in assisting clinicians in the survival prediction of patients with GBM. Further research is needed to establish the applicability of radiomics in the management of GBM in the clinic.

P17.10.B VOLUMETRIC DIFFERENCES IN TARGET DEFINITON OF RECURRENT HIGH-GRADE GLIOMAS USING CONTRAST ENHANCED MRI VS.[18F] FDOPA PET IN RADIOSURGICAL PLANNING

Abstract BACKGROUND Radiosurgical treatment of high-grade gliomas usually bases target definition on contrast-enhanced T1-weighted MRI images. Despite advances in several MRI techniques, a number of limitations exist, primarily due to the disruption of the blood-brain barrier, making the target delineation less accurate, particularly when treating recurrences. Several studies support the use of amino acid PET/MRI in distinguishing progression from post-operative/post-radiation inflammation. However, no established guidelines define the biological target volume when using amino acid PET/MRI in radiosurgical treatment planning. We aimed to explore the volumetric differences in target volumes obtained with standard contrast-enhanced MRI versus [18F]FDOPA PET in patients with recurrent high-grade gliomas (rHGG). MATERIAL AND METHODS Ten patients obtained [18F]FDOPA PET/MRI within 1 week before radiosurgical treatment. We co-registered these images with the 3D MPRAGE-MRI typically used for stereotactic target definition and performed a follow-up PET/MRI scan 4-6 weeks after treatment. We reviewed and compared the method of tumor delineation, target volumes, and stereotactic accuracy in both modalities. RESULTS Radiosurgical treatment delivers high radiation doses to defined targets. The target volume and the proximity to eloquent structures limit the dose, thus depending on high-accuracy imaging. The invasive nature of malignant brain tumors makes anatomical delineation insufficient, particularly when treating recurrences. In such cases, amino acid PET supplies additional information. However, the spatial resolution afforded by [18F]FDOPA uptake alone is insufficient for stereotactic definition, which to a large degree is based on contrast-enhanced T1-weighted MRI images in order to achieve the sub-millimeter resolution needed for radiosurgical planning. Hybrid PET/MRI allows the combination of metabolic information of [18F]FDOPA-PET with the anatomical precision of stereotactic MRI. In the recurrent setting, [18F]FDOPA-PET supplies the necessary information for targeting the biologically active tumour, differentiating it from radiation necrosis, thus limiting the target volumes. This enables us to treat the recurrent tumours more efficiently. CONCLUSION [18F]FDOPA-PET is a useful supplement in radiosurgical planning and treatment of rHGG. It is helpful in the delineation of the biologically active tumor, thus limiting the target volume significantly. This enables higher radiation doses and potentially more efficient treatment. Further results will be presented.

Automated, fast, robust brain extraction on contrast-enhanced T1-weighted MRI in presence of brain tumors: an optimized model based on multi-center datasets

ObjectivesExisting brain extraction models should be further optimized to provide more information for oncological analysis. We aimed to develop an nnU-Net–based deep learning model for automated brain extraction on contrast-enhanced T1-weighted (T1CE) images in presence of brain tumors.MethodsThis is a multi-center, retrospective study involving 920 patients. A total of 720 cases with four types of intracranial tumors from private institutions were collected and set as the training group and the internal test group. Mann–Whitney U test (U test) was used to investigate if the model performance was associated with pathological types and tumor characteristics. Then, the generalization of model was independently tested on public datasets consisting of 100 glioma and 100 vestibular schwannoma cases.ResultsIn the internal test, the model achieved promising performance with median Dice similarity coefficient (DSC) of 0.989 (interquartile range (IQR), 0.988–0.991), and Hausdorff distance (HD) of 6.403 mm (IQR, 5.099–8.426 mm). U test suggested a slightly descending performance in meningioma and vestibular schwannoma group. The results of U test also suggested that there was a significant difference in peritumoral edema group, with median DSC of 0.990 (IQR, 0.989–0.991, p = 0.002), and median HD of 5.916 mm (IQR, 5.000–8.000 mm, p = 0.049). In the external test, our model also showed to be robust performance, with median DSC of 0.991 (IQR, 0.983–0.998) and HD of 8.972 mm (IQR, 6.164–13.710 mm).ConclusionsFor automated processing of MRI neuroimaging data presence of brain tumors, the proposed model can perform brain extraction including important superficial structures for oncological analysis.Clinical relevance statementThe proposed model serves as a radiological tool for image preprocessing in tumor cases, focusing on superficial brain structures, which could streamline the workflow and enhance the efficiency of subsequent radiological assessments.Key Points• The nnU-Net–based model is capable of segmenting significant superficial structures in brain extraction.• The proposed model showed feasible performance, regardless of pathological types or tumor characteristics.• The model showed generalization in the public datasets.

An Explainable MRI-Radiomic Quantum Neural Network to Differentiate Between Large Brain Metastases and High-Grade Glioma Using Quantum Annealing for Feature Selection

Solitary large brain metastases (LBM) and high-grade gliomas (HGG) are sometimes hard to differentiate on MRI. The management differs significantly between these two entities, and non-invasive methods that help differentiate between them are eagerly needed to avoid potentially morbid biopsies and surgical procedures. We explore herein the performance and interpretability of an MRI-radiomics variational quantum neural network (QNN) using a quantum-annealing mutual-information (MI) feature selection approach. We retrospectively included 423 patients with HGG and LBM (> 2 cm) who had a contrast-enhanced T1-weighted (CE-T1) MRI between 2012 and 2019. After exclusion, 72 HGG and 129 LBM were kept. Tumors were manually segmented, and a 5-mm peri-tumoral ring was created. MRI images were pre-processed, and 1813 radiomic features were extracted. A set of best features based on MI was selected. MI and conditional-MI were embedded into a quadratic unconstrained binary optimization (QUBO) formulation that was mapped to an Ising-model and submitted to D’Wave’s quantum annealer to solve for the best combination of 10 features. The 10 selected features were embedded into a 2-qubits QNN using PennyLane library. The model was evaluated for balanced-accuracy (bACC) and area under the receiver operating characteristic curve (ROC-AUC) on the test set. The model performance was benchmarked against two classical models: dense neural networks (DNN) and extreme gradient boosting (XGB). Shapley values were calculated to interpret sample-wise predictions on the test set. The best 10-feature combination included 6 tumor and 4 ring features. For QNN, DNN, and XGB, respectively, training ROC-AUC was 0.86, 0.95, and 0.94; test ROC-AUC was 0.76, 0.75, and 0.79; and test bACC was 0.74, 0.73, and 0.72. The two most influential features were tumor Laplacian-of-Gaussian-GLRLM-Entropy and sphericity. We developed an accurate interpretable QNN model with quantum-informed feature selection to differentiate between LBM and HGG on CE-T1 brain MRI. The model performance is comparable to state-of-the-art classical models.

Evaluation of glymphatic-meningeal lymphatic system with intravenous gadolinium-based contrast-enhancement in cerebral small-vessel disease.

To investigate the feasibility of using noninvasive neuroimaging methods in visualizing and evaluating the clearance of the glymphatic-meningeal lymphatic system (GMLS) in patients with arteriosclerotic cerebral small-vessel disease (CSVD) and controls. This observational study recruited patients with high-burden CSVD and controls (age 50-80years). At multiple time points before and after intravenous administration of a contrast agent, three-dimensional (3D) brain volume T1-weighted imaging and 3D Cube T2-fluid attenuated inversion recovery imaging were performed to visualize and assess the clearance of the glymphatics and meningeal lymphatic vessels (mLVs). We measured the signal intensity ratio (SIR) of four regions of interest representing the glymphatics and mLVs at each time point. The clearance rate at 24h (CR24h) and changes in the SIR from baseline to 24h (∆SIR) were defined as the clearance function. The analysis of variance was used to evaluate the group differences after adjusting for hypertension. A total of 20 CSVD patients and 15 controls were included. Cortical periarterial enhancement and the enhancement of enlarged perivascular spaces in the basal ganglia were respectively observed in 11 (55.00%) and 16 (80.00%) CSVD patients, but in none of controls. All CSVD patients and most of controls (80.00%) showed cortical perivenous enhancement. Para-sinus enhancement was observed in all participants. CSVD patients showed lower CR24h and higher ∆SIR of the glymphatics and mLVs (all p < 0.05). The impaired drainage of the GMLS in patients with high-burden CSVD could be visually evaluated using noninvasive neuroimaging methods with intravenous gadolinium-based contrast-enhancement. Dynamic intravenous contrast-enhanced MRI could visually evaluate the impaired drainage of the glymphatic-meningeal lymphatic system in patients with high-burden cerebral small-vessel disease and could help to explore a new therapeutic target. • Signal intensity changes in regions representing the glymphatic-meningeal lymphatic system (GMLS) can reflect the drainage function based on contrast-enhanced 3D-FLAIR and 3D T1-weighted MRI. • Impaired drainage of the GMLS in patients with high-burden CSVD can be visually evaluated using dynamic intravenous contrast-enhanced MRI. • This direct, noninvasive technique could serve as a basis for further GMLS studies and could help to explore a new therapeutic target in CSVD patients.

Radiomics for the non-invasive prediction of PD-L1 expression in patients with brain metastases secondary to non-small cell lung cancer

BackgroundThe expression level of the programmed cell death ligand 1 (PD-L1) appears to be a predictor for response to immunotherapy using checkpoint inhibitors in patients with non-small cell lung cancer (NSCLC). As differences in terms of PD-L1 expression levels in the extracranial primary tumor and the brain metastases may occur, a reliable method for the non-invasive assessment of the intracranial PD-L1 expression is, therefore of clinical value. Here, we evaluated the potential of radiomics for a non-invasive prediction of PD-L1 expression in patients with brain metastases secondary to NSCLC.Patients and methodsFifty-three NSCLC patients with brain metastases from two academic neuro-oncological centers (group 1, n = 36 patients; group 2, n = 17 patients) underwent tumor resection with a subsequent immunohistochemical evaluation of the PD-L1 expression. Brain metastases were manually segmented on preoperative T1-weighted contrast-enhanced MRI. Group 1 was used for model training and validation, group 2 for model testing. After image pre-processing and radiomics feature extraction, a test-retest analysis was performed to identify robust features prior to feature selection. The radiomics model was trained and validated using random stratified cross-validation. Finally, the best-performing radiomics model was applied to the test data. Diagnostic performance was evaluated using receiver operating characteristic (ROC) analyses.ResultsAn intracranial PD-L1 expression (i.e., staining of at least 1% or more of tumor cells) was present in 18 of 36 patients (50%) in group 1, and 7 of 17 patients (41%) in group 2. Univariate analysis identified the contrast-enhancing tumor volume as a significant predictor for PD-L1 expression (area under the ROC curve (AUC), 0.77). A random forest classifier using a four-parameter radiomics signature, including tumor volume, yielded an AUC of 0.83 ± 0.18 in the training data (group 1), and an AUC of 0.84 in the external test data (group 2).ConclusionThe developed radiomics classifiers allows for a non-invasive assessment of the intracranial PD-L1 expression in patients with brain metastases secondary to NSCLC with high accuracy.

Prediction of model generalizability for unseen data: Methodology and case study in brain metastases detection in T1-Weighted contrast-enhanced 3D MRI

Background and purposeA medical AI system's generalizability describes the continuity of its performance acquired from varying geographic, historical, and methodologic settings. Previous literature on this topic has mostly focused on “how” to achieve high generalizability (e.g., via larger datasets, transfer learning, data augmentation, model regularization schemes), with limited success. Instead, we aim to understand “when” the generalizability is achieved: Our study presents a medical AI system that could estimate its generalizability status for unseen data on-the-fly. Materials and methodsWe introduce a latent space mapping (LSM) approach utilizing Fréchet distance loss to force the underlying training data distribution into a multivariate normal distribution. During the deployment, a given test data's LSM distribution is processed to detect its deviation from the forced distribution; hence, the AI system could predict its generalizability status for any previously unseen data set. If low model generalizability is detected, then the user is informed by a warning message integrated into a sample deployment workflow. While the approach is applicable for most classification deep neural networks (DNNs), we demonstrate its application to a brain metastases (BM) detector for T1-weighted contrast-enhanced (T1c) 3D MRI. The BM detection model was trained using 175 T1c studies acquired internally (from the authors' institution) and tested using (1) 42 internally acquired exams and (2) 72 externally acquired exams from the publicly distributed Brain Mets dataset provided by the Stanford University School of Medicine. Generalizability scores, false positive (FP) rates, and sensitivities of the BM detector were computed for the test datasets. Results and conclusionThe model predicted its generalizability to be low for 31% of the testing data (i.e., two of the internally and 33 of the externally acquired exams), where it produced (1) ∼13.5 false positives (FPs) at 76.1% BM detection sensitivity for the low and (2) ∼10.5 FPs at 89.2% BM detection sensitivity for the high generalizability groups respectively. These results suggest that the proposed formulation enables a model to predict its generalizability for unseen data.

A deep learning based four-fold approach to classify brain MRI: BTSCNet

Incorrect diagnosis of brain tumor types prevent appropriate response to medical assistance and reduces patients' chances of survival. Examining MRI images of the patient's brain manually is one traditional method for distinguishing brain tumors, but it is time consuming and prone to human errors. Thus, an automated and new deep learning based four-fold method, Brain Tumor Segmentation and Classification Network (BTSCNet), is proposed in this article which will be helpful for physicians to classify three types of brain tumors (meningioma, glioma, and pituitary tumor) from T1-weighted contrast-enhanced MRI (CE-MRI) images properly. The proposed four folds includes: segmentation of brain tumor region using Brain Tumor Segmentation Network (BTSNet), ROI selection using morphological operation, feature extraction using multi-region gray level co-occurrence matrix (MR-GLCM), andclassification using Sliding Window Euclidean distance (SWED) measure. The lack of annotated training samples is the main challenge in deep-learning-based brain MRIimage classification. Thus, while training the proposed system, the scale, orientation and flip of the input image is randomly changed in the first fold of BTSCNetso that the network can be trained with varied or augmentedimages with respect toscale, orientation and flip.Four performance measures are used to measure the efficiency of the proposed method. The output of the proposed method is tested on a public database where the correct classification rates obtained are96.6% (meningioma), 98.1% (glioma), and 95.3% (pituitary tumor), when considering MR-Contrast feature, MR-Correlation and MR-Homogeneity feature respectively, which indicates that the efficiency of the proposed method.

Transcranial cavitation mapping of blood–brain barrier opening regions in Alzheimer’s disease patients using a neuronavigation-guided focused ultrasound system

We present real-time cavitation monitoring and mapping in a clinical trial for blood-brain barrier (BBB) opening in Alzheimer’s disease (AD) patients using a neuronavigation-guided focused ultrasound (FUS) system. Six AD patients (N = 6, age = 68.5 ± 9.5) were sonicated at the right prefrontal lobe with a single-element FUS transducer (PNP = 0.2 MPa, fc = 0.25 MHz, pulse length = 10 ms, PRF = 2 Hz, duration = 2 min) with the microbubble administration (Definity). We performed passive cavitation detection (PCD) using a single-element hydrophone (N = 4), and passive acoustic mapping (PAM) with a 64-element phased array transducer (N = 2). Five patients showed successful BBB opening (volume = 654 ± 394 mm3) on day0, and closure on day3, confirmed by contrast-enhanced T1-weighted MRI, while one patient served as a negative BBB opening case due to the insufficient microbubble administration. Ultra-harmonic cavitation dose (CD) correlated with the opening volume (R2 = 0.66, N = 4), but harmonic and broadband CDs showed a relatively poor correlation (R2 = 0.06 and 0.33). Cavitation maps (N = 2) showed a spatial distribution of acoustic energy that roughly coincided with the respective BBB opening location. We thus demonstrated the successful BBB opening and cavitation monitoring with the neuronavigation-guided system that allowed for a cost-effective procedure compared to the MR-guided approaches. The PCD and PAM showed promising results for potentially predicting the BBB opening volume and location found in MRI.

Refined Automatic Brain Tumor Classification Using Hybrid Convolutional Neural Networks for MRI Scans.

Refined hybrid convolutional neural networks are proposed in this work for classifying brain tumor classes based on MRI scans. A dataset of 2880 T1-weighted contrast-enhanced MRI brain scans are used. The dataset contains three main classes of brain tumors: gliomas, meningiomas, and pituitary tumors, as well as a class of no tumors. Firstly, two pre-trained, fine-tuned convolutional neural networks, GoogleNet and AlexNet, were used for classification process, with validation and classification accuracy being 91.5% and 90.21%, respectively. Then, to improving the performance of the fine-tuning AlexNet, two hybrid networks (AlexNet-SVM and AlexNet-KNN) were applied. These hybrid networks achieved 96.9% and 98.6% validation and accuracy, respectively. Thus, the hybrid network AlexNet-KNN was shown to be able to apply the classification process of the present data with high accuracy. After exporting these networks, a selected dataset was employed for testing process, yielding accuracies of 88%, 85%, 95%, and 97% for the fine-tuned GoogleNet, the fine-tuned AlexNet, AlexNet-SVM, and AlexNet-KNN, respectively. The proposed system would help for automatic detection and classification of the brain tumor from the MRI scans and safe the time for the clinical diagnosis.

Radiomics analysis of contrast-enhanced T1W MRI: predicting the recurrence of acute pancreatitis

To investigate the predictive value of radiomics based on T1-weighted contrast-enhanced MRI (CE-MRI) in forecasting the recurrence of acute pancreatitis (AP). A total of 201 patients with first-episode of acute pancreatitis were enrolled retrospectively (140 in the training cohort and 61 in the testing cohort), with 69 and 30 patients who experienced recurrence in each cohort, respectively. Quantitative image feature extraction was obtained from MR contrast-enhanced late arterial-phase images. The optimal radiomics features retained after dimensionality reduction were used to construct the radiomics model through logistic regression analysis, and the clinical characteristics were collected to construct the clinical model. The nomogram model was established by linearly integrating the clinically independent risk factor with the optimal radiomics signature. The five best radiomics features were determined by dimensionality reduction. The radiomics model had a higher area under the receiver operating characteristic curve (AUC) than the clinical model for estimating the recurrence of acute pancreatitis for both the training cohort (0.915 vs. 0.811, p = 0.020) and testing cohort (0.917 vs. 0.681, p = 0.002). The nomogram model showed good performance, with an AUC of 0.943 in the training cohort and 0.906 in the testing cohort. The radiomics model based on CE-MRI showed good performance for optimizing the individualized prediction of recurrent acute pancreatitis, which provides a reference for the prevention and treatment of recurrent pancreatitis.

Pretreatment findings on magnetic resonance imaging in primary central nervous system lymphoma may predict overall survival duration.

and purpose: Primary central nervous system lymphoma (PCNSL) lesions often show avid contrast enhancement on T1-weighted contrast-enhanced MRI sequences. However, several case reports and a clinical study have described PCNSL in patients with no contrast enhancement on MRI. We assessed whether overall survival (OS) time was related to any tumor characteristics (lesion location, volume, and number; contrast enhancement; necrosis; proximity to the subarachnoid space; and edema) on MRI in patients with PCNSL. We retrospectively reviewed records (MRI features, pathology, and survival data) of all patients at our institution with PCNSL who had been seen from, 2007 through 2017, and had undergone pretreatment MRI. We identified 79 patients (42 men, 37 women) with a mean age at diagnosis of 61.7 ± 10.4years. The mean OS duration was 44.6 ± 41.7months. The most common pathological diagnosis (74 patients) was diffuse large B-cell lymphoma. No associations were found between OS time and lesion location, volume, and number; contrast enhancement; necrosis; proximity to the subarachnoid space; or edema. However, a sole patient with non-enhancing PCNSL on MRI was found to have low-grade disease, with prolonged survival (>83months). Several other patients with leptomeningeal disease had a mean OS time of 80 months. Patients with hemorrhagic lesions had a mean OS of 25.5months. The survival time for patients with PCNSL may be longer than previously thought, especially for patients with leptomeningeal seeding and lesions with hemorrhagic components Also, non-enhancing tumors may be less aggressive than enhancing tumors.

Multi-Classification of Brain Tumors on Magnetic Resonance Images Using an Ensemble of Pre-Trained Convolutional Neural Networks.

Automatic classification of brain tumors is an important issue in computeraided diagnosis (CAD) for medical applications since it can efficiently improve the clinician's diagnostic performance and the current study focused on the CAD system of the brain tumors. Existing studies mainly focused on a single classifier either based on traditional machinelearning algorithms or deep learning algorithms with unsatisfied results. In this study, we proposed an ensemble of pre-trained convolutional neural networks to classify brain tumors into three types from their T1-weighted contrast-enhanced MRI (CE-MRI) images, which are meningioma, glioma, and pituitary tumor. Three pre-trained convolutional neural networks (Inception-v3, Resnet101, Densenet201) with the best classification performance (i.e. accuracy of 96.21%, 97.00%, 96.54%, respectively) on the CE-MRI benchmark dataset were selected as backbones of the ensemble model. The features extracted by backbone networks in the ensemble model were further classified by a support vector machine. The ensemble system achieved an average classification accuracy of 98.14% under a five-fold cross-validation process, outperforming any single deep learning model in the ensemble system and other methods in the previous studies. Performance metrics for each brain tumor type, including area under the curve, sensitivity, specificity, precision, and F-score, were calculated to show the ensemble system's performance. Our work addressed a practical issue by evaluating the model with fewer training samples. The classification accuracy was reduced to 97.23%, 96.87%, and 93.96% when 75%, 50%, and 25% training data was used to train the ensemble model, respectively. Our ensemble model has a great capacity and achieved the best performance in any single convolutional neural networks for brain tumors classification and is potentially applicable in real clinical practice.