Deep Learning Algorithms for Bladder Cancer Segmentation on Multi-Parametric MRI.

BACKGROUND: Bladder cancer (BC) segmentation on MRI images is the first step to determining the presence of muscular invasion. This study aimed to assess the tumor segmentation performance of three deep learning (DL) models on multi-parametric MRI (mp-MRI) images. METHODS: We studied 53 patients with bladder cancer. Bladder tumors were segmented on each slice of T2-weighted (T2WI), diffusion-weighted imaging/apparent diffusion coefficient (DWI/ADC), and T1-weighted contrast-enhanced (T1WI) images acquired at a 3Tesla MRI scanner. We trained Unet, MAnet, and PSPnet using three loss functions: cross-entropy (CE), dice similarity coefficient loss (DSC), and focal loss (FL). We evaluated the model performances using DSC, Hausdorff distance (HD), and expected calibration error (ECE). RESULTS: The MAnet algorithm with the CE+DSC loss function gave the highest DSC values on the ADC, T2WI, and T1WI images. PSPnet with CE+DSC obtained the smallest HDs on the ADC, T2WI, and T1WI images. The segmentation accuracy overall was better on the ADC and T1WI than on the T2WI. The ECEs were the smallest for PSPnet with FL on the ADC images, while they were the smallest for MAnet with CE+DSC on the T2WI and T1WI. CONCLUSIONS: Compared to Unet, MAnet and PSPnet with a hybrid CE+DSC loss function displayed better performances in BC segmentation depending on the choice of the evaluation metric.

MRI-based radiomic features for identifying recurrent prostate cancer after proton radiation therapy.

PURPOSE: Magnetic Resonance Imaging (MRI) evaluation of recurrent prostate cancer (PCa) following proton beam therapy is challenging due to radiation-induced tissue changes. This study aimed to evaluate MRI-based radiomic features so as to identify the recurrent PCa after proton therapy. METHODS: We retrospectively studied 12 patients with biochemical recurrence (BCR) following proton therapy. Two experienced radiologists identified prostate lesions from multi-parametric MRI (mpMRI) images post-proton therapy and marked control regions of interest (ROIs) on the contralateral side of the prostate gland. A total of 210 radiomic features were extracted from lesions and control regions on the T2-weighted (T2WI) and Apparent Diffusion Coefficient (ADC) image series. Recursive Feature Elimination with Cross-Validation method (RFE-CV) was used for feature selection. A Multilayer Perceptron (MLP) neural network was developed to classify three classes: cancerous, benign, and healthy tissue. The 12-core biopsy results were used as the gold standard for the segmentations. The classifier performance was measured using specificity, sensitivity, the area under receiver operating characteristic curve (AUC), and other statistical indicators. RESULTS: Based on biopsy results, 10 lesions were identified as PCa recurrence while eight lesions were confirmed to be benign. Ten radiomic features (10/210) were selected to build the multi-class classifier. The radiomics classifier gave an accuracy of 0.83 in identifying cancerous, benign, and healthy tissue with a sensitivity of 0.80 and specificity of 0.85. The model yielded an AUC of 0.87, 95% CI [0.72-1.00] in differentiating cancer from the benign and healthy tissues. CONCLUSIONS: Our proof-of-concept study demonstrates the potential of using radiomic features as part of the differential diagnosis of PCa on mpMRI following proton therapy. The results need to be validated in a larger cohort.

Quality assurance for magnetic resonance angiography of the chest in patients suspected of pulmonary embolism during iodinated contrast shortage in the emergency department setting.

Objectives: COVID-19 lockdowns resulted in a global shortage of iodinated contrast media. Therefore, alternative imaging protocols were devised to evaluate patients arriving to the emergency department (ED) with suspicion of pulmonary embolism (PE). This quality assurance (QA) aims to compare diagnostic potential between alternative magnetic resonance angiography (MRA) protocol over the gold standard computed tomography angiography (CTA) by evaluating MRA imaging quality, scanner type/imaging sequence, and any risk of misdiagnosis in patients with symptoms of PE. Material and Methods: This retrospective study compromised of 55 patients who arrived to ED and underwent MRA of the chest for suspicion of PE during the months of May to June 2022. Data regarding their chief complaints, imaging sequence, and MRA results were collected. Two fellowship-trained faculty radiologists reviewed the MRA scans of the patients and scored the quality using a Likert scale. Results: Two patients were positive for PE and 53 patients showed negative results. Regarding the scan quality issues, motion was noted in 80% of the 55 studies that we reviewed. Significant associations (P < 0.009) between Likert scale scores and initial complaint category were found. The characteristic symptoms associated with suspicion of PE, namely, shortness of breath, chest pain, and cough were distributed among the 1 and 2 categories, reflecting the most optimal vessel opacification scores. We found no risk of misdiagnosis after reviewing the electronic medical record for follow-up appointments within 6 months of ED visit. Conclusion: Patients were screened for PE with MRA as an alternative imaging tool during times of contrast shortage. Further, evaluation of MRA with CTA, side by side, in a larger patient population is required to increase the validity of our QA study.

MRI-based nomograms and radiomics in presurgical prediction of extraprostatic extension in prostate cancer: a systematic review.

PURPOSE: Prediction of extraprostatic extension (EPE) is essential for accurate surgical planning in prostate cancer (PCa). Radiomics based on magnetic resonance imaging (MRI) has shown potential to predict EPE. We aimed to evaluate studies proposing MRI-based nomograms and radiomics for EPE prediction and assess the quality of current radiomics literature. METHODS: We used PubMed, EMBASE, and SCOPUS databases to find related articles using synonyms for MRI radiomics and nomograms to predict EPE. Two co-authors scored the quality of radiomics literature using the Radiomics Quality Score (RQS). Inter-rater agreement was measured using the intraclass correlation coefficient (ICC) from total RQS scores. We analyzed the characteristic s of the studies and used ANOVAs to associate the area under the curve (AUC) to sample size, clinical and imaging variables, and RQS scores. RESULTS: We identified 33 studies-22 nomograms and 11 radiomics analyses. The mean AUC for nomogram articles was 0.783, and no significant associations were found between AUC and sample size, clinical variables, or number of imaging variables. For radiomics articles, there were significant associations between number of lesions and AUC (p < 0.013). The average RQS total score was 15.91/36 (44%). Through the radiomics operation, segmentation of region-of-interest, selection of features, and model building resulted in a broader range of results. The qualities the studies lacked most were phantom tests for scanner variabilities, temporal variability, external validation datasets, prospective designs, cost-effectiveness analysis, and open science. CONCLUSION: Utilizing MRI-based radiomics to predict EPE in PCa patients demonstrates promising outcomes. However, quality improvement and standardization of radiomics workflow are needed.

Not all imagery is created equal: A functional Magnetic resonance imaging study of internally driven and symbol driven musical performance imagery.

Playing a musical instrument requires fast multimodal sensory-motor processing which can be activated by voluntary access to performance imagery. Musicians use different methods to activate imagery for the purpose of "mental practice". The aim of the present study was to investigate cortical activation patterns in different methods of mental practice of musical performance. While undergoing functional magnetic resonance imaging (fMRI), 7 male oud (fretless lute) players engaged in performance imagery of a pre-memorized short excerpt from mainstream oud repertoire using three common imagery methods (task conditions): From memory (internally driven) 1)eyes closed, 2)eyes open, and while following the musical score (symbol driven). The study design consisted of a four-task 16-epoch block design where the 4th task was an eyes-open rest tasks (EOR) included as a control condition. Each task was repeated four times in a pseudorandomized sequence. The superior temporal gyrus and transvers temporal gyrus (Heschl) were active in the left and right hemispheres in all imagery conditions. The occipital cortex, specifically the fusiform gyrus was active in all three conditions. Symbol driven imagery resulted in less prominent activations in frontal and parietal lobes. The findings suggest that not all imagery modalities activate sensory and motor areas similarly.

Diffusion tensor and volumetric magnetic resonance imaging findings in the brains of professional musicians.

Professional musicians represent an ideal model to study the training-induced brain plasticity. The current study aimed to investigate the brain volume and diffusion characteristics of musicians using structural magnetic resonance and diffusion tensor imaging (DTI). The combined use of volumetric and diffusion methods in studying musician brain has not been done in literature. Our study group consisted of seven male musicians playing an instrument and seven age- and gender-matched non-musicians. We evaluated the volumes of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) and calculated total intracranial volume (TIV) and measured the fractional anisotropy (FA) of pre-selected WM bundles: corpus callosum (CC), corticospinal tract (CST), superior longitudinal fasciculus (SLF), forceps major (ForMaj), forceps minor (ForMin), and arcuate fasciculus (AF). The mean WM/TIV volume in musicians was higher compared to non-musicians. The mean FA was lower in CC, SLF, ForMaj, ForMin, and right AF but higher in right CST in the musicians. The mean value of the total number of fibers was larger in the CST, SLF, left AF, and ForMaj in the musicians. The observed differences were not statistically significant between the groups (p>0.05). However, increased GM volume was found in the musicians compared to the non-musicians in the right and left cerebellum and supramarginal and angular gyrus, left superior and inferior parietal lobule and as well as left middle temporal gyrus. Our findings suggest differing brain structure in musicians and the confirmation of the results on a larger population.

Apparent diffusion coefficient in differentiation of pediatric posterior fossa tumors.

PURPOSE: To investigate the contribution of preoperative apparent diffusion coefficient (ADC) values in the differential diagnosis of pediatric posterior fossa tumors. METHODS: Forty-two pediatric patients (mean age 7.76 ± 4.58 years) with intra-axial tumors in the infra-tentorial region underwent magnetic resonance imaging. ADC measurement was performed using regions of interest, obtained from the solid component of the mass lesions. ADC ratios were calculated by dividing the ADC values from the mass lesions by the ADC values from normal cerebellar parenchyma. Lesions were categorized as juvenile pilocytic astrocytoma (JPA), ependymoma and medulloblastoma based on histopathological diagnosis. ADC values of the lesions and histopathological diagnoses were statistically correlated. RESULTS: Histopathological diagnosis showed that 14 lesions were JPA, 10 were ependymoma; 18 were medulloblastoma. Both ADC values and ADC ratios were significantly correlated with tumor types (p <0.05). Astrocytoma was distinguished from ependymoma with sensitivity 85.7% and specificity 90% using an ADC ratio ≥1.7 and medulloblastoma was distinguished from ependymoma with sensitivity 100% and specificity 88.89% using an ADC ratio ≤1.18. CONCLUSION: Preoperative ADC values could differentiate the main histological subtypes of pediatric posterior fossa tumors with high sensitivity and specificity.