Multiparametric Magnetic Resonance Imaging Research Articles

Technical aspects and clinical applications of synthetic MRI: a scoping review

Abstract Introduction Synthetic magnetic resonance imaging (SyMRI) is a non-invasive, robust MRI technique that generates multiple contrast-weighted images by acquiring a single MRI sequence within a few minutes, along with quantitative maps, automatic brain segmentation, and volumetry. Since its inception, it has undergone technical advancements and has also been tested for feasibility in various organs and pathological conditions. This scoping review comprehensively pinpoints the critical technical aspects and maps the wide range of clinical applications/benefits of SyMRI. Content A comprehensive search was conducted across five databases, PubMed, Scopus, Web of Science, Embase, and CINAHL Ultimate, using appropriate keywords related to SyMRI. A total of 99 studies were included after a 2-step screening process. Data related to the technical factors and clinical application was charted. Summary SyMRI provides quantitative maps and segmentation techniques comparable to conventional MRI and has demonstrated feasibility and applications across neuroimaging, musculoskeletal, abdominal and breast pathologies spanning the entire human lifespan, from prenatal development to advanced age. Certain drawbacks related to image quality have been encountered that can be overcome with technical advances, especially AI-based algorithms. Outlook SyMRI has immense potential for being incorporated into routine imaging for various pathologies due to its added advantage of providing quantitative measurements for more robust diagnostic and prognostic work-up with faster acquisitions and greater post-processing options.

Enhancing deep learning methods for brain metastasis detection through cross-technique annotations on SPACE MRI.

Gadolinium-enhanced "sampling perfection with application-optimized contrasts using different flip angle evolution" (SPACE) sequence allows better visualization of brain metastases (BMs) compared to "magnetization-prepared rapid acquisition gradient echo" (MPRAGE). We hypothesize that this better conspicuity leads to high-quality annotation (HAQ), enhancing deep learning (DL) algorithm detection of BMs on MPRAGE images. Retrospective contrast-enhanced (gadobutrol 0.1 mmol/kg) SPACE and MPRAGE data of 157 patients with BM were used, either annotated on MPRAGE resulting in normal annotation quality (NAQ) or on coregistered SPACE resulting in HAQ. Multiple DL methods were developed with NAQ or HAQ using either SPACE or MRPAGE images and evaluated on their detection performance using positive predictive value (PPV), sensitivity, and F1 score and on their delineation performance using volumetric Dice similarity coefficient, PPV, and sensitivity on one internal and four additional test datasets (660 patients). The SPACE-HAQ model reached 0.978 PPV, 0.882 sensitivity, and 0.916 F1-score. The MPRAGE-HAQ reached 0.867, 0.839, and 0.840, the MPRAGE NAQ 0.964, 0.667, and 0.798, respectively (p ≥ 0.157). Relative to MPRAGE-NAQ, the MPRAGE-HAQ F1-score detection increased on all additional test datasets by 2.5-9.6 points (p < 0.016) and sensitivity improved on three datasets by 4.6-8.5 points (p < 0.001). Moreover, volumetric instance sensitivity improved by 3.6-7.6 points (p < 0.001). HAQ improves DL methods without specialized imaging during application time. HAQ alone achieves about 40% of the performance improvements seen with SPACE images as input, allowing for fast and accurate, fully automated detection of small (< 1 cm) BMs. Training with higher-quality annotations, created using the SPACE sequence, improves the detection and delineation sensitivity of DL methods for the detection of brain metastases (BMs)on MPRAGE images. This MRI cross-technique transfer learning is a promising way to increase diagnostic performance. Delineating small BMs on SPACE MRI sequence results in higher quality annotations than on MPRAGE sequence due to enhanced conspicuity. Leveraging cross-technique ground truth annotations during training improved the accuracy of DL models in detecting and segmenting BMs. Cross-technique annotation may enhance DL models by integrating benefits from specialized, time-intensive MRI sequences while not relying on them. Further validation in prospective studies is needed.

The Value of MRI and Radiomics for the Diagnostic Evaluation of Thyroid-Associated Ophthalmopathy

Thyroid-associated ophthalmopathy (TAO) is a vision-threatening autoimmune disease that involves the extraocular muscles (EOMs) and periorbital fat. Typical signs of TAO include eyelid recession, proptosis, diplopia, and decreased visual acuity. As a self-limited disease, there is major bipolarity in clinical outcomes in TAO population. The early diagnosis and prediction of these refractory and relapsed patients is essential. Unfortunately, commonly used tools such as CAS/NOSPECTS, are based on clinical symptoms and signs alone, have significant limitations. Some imaging techniques or examinations, such as magnetic resonance imaging (MRI), can be very effective in assisting TAO assessment, from exhaustive whiteboard notes to optimized patient outcomes. Being one of the most commonly used and accurate objective examinations for TAO assessment, MRI boosts no ionizing radiation, high soft tissue contrast, better reflection of tissue water content, and the ability to quantify multiple parameters. In addition, novel MR sequences are becoming increasingly more familiar in TAO and other areas of clinical and scientific research. Moreover, radiomics, a method involving the extraction of a large number of features from medical images through algorithms, is a more recent approach used in the analysis and characterization of TAO data. Thus, this review aims to summarize and compare the value of routine and novel functional MRI sequences and radiomics prediction models in the diagnosis and evaluation of TAO.

ISSLS Prize in Clinical Science 2025: Cartilage End Plate Defects Precede and Initiate Bony End Plate Defects and Disc Degeneration- An 'Integrated Total End Plate Score' Identify Preclinical Discs at Risk for Degeneration.

We utilized the Fast Low Angle Shot (FLASH) sequence to document the sequential changes in cartilaginous (CEP) and bony end plate (BEP) to study the influence on disc degeneration (DD). Routine MRI and FLASH sequences were used in 500 lumbar discs in 100 each of healthy volunteers (HV), low back pain patients treated conservatively (CG) and surgically (SG) to document CEP and BEP status, Pfirrmann Grade (PG) and various MRI parameters. The three groups were identical demographically but had a significantly different number of healthy discs (p < 0.01) and changes in CEP and BEP (p < 0.01), with patients having a higher severity of end plate changes and DD, even in asymptomatic discs. CEP abnormalities always appeared first, followed by a sequence of BEP defects of different severity, allowing the development of an 'Integrated Total End Plate Score' (I-TEPS). There was a good correlation between I-TEPS and PG, with a steep escalation of DD after a score of 7. A score of ≥ 7 was also associated with higher surgical incidence in patients with both degenerated and herniated discs. The most significant influencing factors for surgery was a combination of I-TEPS ≥ 7 with herniation (OR7.7;p-0.00), smoking (OR4.63;p-0.02), and an I-TEPS ≥ 7 (OR3.37;p-0.04). CEP changes identified by FLASH preceded BEP defects and DD. I-TEPS was superior to TEPS in identifying a subgroup of discs that had CEP abnormalities without BEP. An I-TEPS ≥ 7 had a significant correlation to the severity of DD, influenced variations in herniation and also surgical incidence.

Single-Port Robotic Transvesical Partial Prostatectomy: A Novel Approach for Focal Treatment in Prostate Cancer.

Introduction and Objective: Given the favorable cancer-specific survival rates in localized prostate cancer and the negative impact of whole-gland treatments on functional outcomes, the field is moving toward precision strategies such as focal therapy and organ-sparing surgery. We aim to report medium-term functional and oncologic outcomes for the initial Single-Port Robotic Transvesical Partial Prostatectomy (SP-TVRAPP) patient cohort. Materials and Methods: We analyzed a prospectively maintained database of 20 patients who underwent SP-TVRAPP between February 2021 and March 2024. Inclusion criteria were prostate-specific antigen (PSA) ≤10 ng/mL, clinical stage ≤ cT2b, ISUP Grade Group ≤3, unilateral lesions on multiparametric magnetic resonance imaging (mpMRI) with positive biopsy cores on the same side, and preoperative IIEF-5 ≥ 17. We also considered bilateral prostate cancer in the anterior zone and invisible mpMRI tumors confirmed by unilateral positive biopsies. Results: At baseline, patients had an average age of 61 years, a median PSA of 4.8 ng/mL (interquartile range [IQR]: 3.7-7.7), and a median Sexual Health Inventory for Men (SHIM) score of 24 (IQR: 18-25). All procedures were completed without complications, need for additional ports, or conversion. After a median hospital stay of 4.2 hours, 94% of cases were discharged without opioid prescriptions, and Foley catheters were removed after approximately 4 days. At 6 weeks, 3, 6, and 12 months postprocedure, potency rates, defined as a SHIM score ≥17, were 45.0%, 77.7%, 83.3%, and 87.5%, respectively. When potency was defined as having erections sufficient for penetration, the rates were 80.0%, 88.8%, 88.8%, and 93.7% for the same time intervals. Regarding urinary function, 60.0% were continent at 1 week, increasing to 85.0% by 6 weeks, 88.8% at both 3 and 6 months, and reaching 93.7% at 12 months postsurgery. Oncologically, 30.0% experienced upgrading and 40.0% upstaging within this cohort. Negative surgical margins were attained in 85.0% of the cases and the median PSA was 0.4 ng/mL 12 months after SP-TVRAPP. Two men were found to have residual GG1 cancer in the protocol biopsies and are currently on active surveillance. At a mean follow-up of 15.5 months (0.2-34.8) months, none of the patients has required secondary interventions, and all remain free of both clinically significant residual prostate cancer and metastatic disease. Conclusions: SP-TVRAPP represents a promising treatment for certain patients with localized prostate cancer. This targeted surgical method has been associated with faster postoperative recovery and has demonstrated high rates of early recovery in erectile function and urinary continence while ensuring oncologic safety.

Correlation of clinically significant prostate cancer sites across multiparametric MRI, prostate biopsy, and whole-mount pathology for optimal prostate biopsy strategy

Data from patients with suspicious lesions on multiparametric magnetic resonance (mpMRI) and a diagnosis of clinically significant prostate cancer (csPCa) who underwent radical prostatectomy (RP) were collected. The aim was to compare csPCa sites identified through whole-mount pathological analysis (WMA) after RP with PI-RADS ≥ 3 lesions identified on mpMRI, and with csPCa foci detected through targeted-biopsy (TB) or combined targeted + systematic biopsy (TSB). A paired Student’s t-test and Pearson correlation analysis were performed to evaluate the agreement between these diagnostic methods. A total of 106 patients were included in the TSB group and 95 in the TB group. The correlation between mpMRI, PB, and WMA was moderate and comparable in both groups. No correlation between PB and WMA was found in the TB group for PI-RADS3 lesions, while a moderate-strong correlation was observed when comparing mpMRI, PB, and WMA for PI-RADS > 3 lesions. About 50% of csPCa sites remained undetected by mpMRI. TSB was able to identify 16.5% more csPCa sites than TB. mpMRI is an accurate method in the diagnosis of PCa, especially for PI-RADS > 3 lesions, although some csPCa sites remained undetected. The use of TSB improved the location agreement between PB and WMA, increasing detection rates up to 79%.

Can multiparametric MRI clear cell likelihood scores differentiate fat-Poor AML from CcRCC in subcentimeter lesions?

To investigate the potential of multiparametric MRI clear cell likelihood scores (ccLS) for differentiating between fat-poor angiomyolipoma (AML) and clear cell renal cell carcinoma (ccRCC) in subcentimeter Lesions (1cm or smaller). This retrospective study included consecutive patients with subcentimeter renal masses who underwent multiparametric MRI between September 2009 and September 2022 across three hospitals. Clinical and MRI findings were analyzed to differentiate between fat-poor AML and ccRCC. Lesions were categorized using the ccLS and receiver operating characteristic curve analysis was performed to assess ccLS performance. Thirty-eight patients (mean age: 52 years ± 12; 19 women) with 39 lesions were included. Of the 39 lesions [mean size: 9.1mm ± 1.0 (range, 6.0-10.0mm)], 20 (51%) were ccRCC and 19 (49%) were fat-poor AML. Compared to the ccRCC, subcentimeter fat-poor AMLs were more likely to show hypointensity on T2WI (P < 0.001), homogeneous enhancement (P = 0.010), the presence of microscopic fat (P = 0.036), and the absence of a pseudocapsule (P = 0.020). The diagnostic percentage of fat-poor AML was 47% for a ccLS of 1 or 2, and ccRCC accounted for 75% in the ccLS 4 or 5 category. The AUC for discrimination was 0.846 (95% CI: 0.695-0.941, P < 0.001), with a sensitivity of 75.00% (95% CI: 50.9-91.3) and a specificity of 89.47% (95% CI: 66.9-98.7). Multiparametric MRI clear cell likelihood scores can potentially be used to differentiate between fat-poor AML and ccRCC in lesions 1cm or smaller.

Dual-Time-Point Radiomics for Prognosis Prediction in Colorectal Liver Metastasis Treated with Neoadjuvant Therapy Before Radical Resection: A Two-Center Study.

Optimal prognostic stratification for colorectal liver metastases (CRLM) patients undergoing surgery with neoadjuvant therapy (NAT) remains elusive. This study aimed to develop and validate dual-time-point radiomic models for CRLM prognosis prediction using pre- and post-NAT imaging features. Radiomic features were extracted from four MRI sequences in 100 cases of CRLM patients who underwent NAT and radical resection. RAD scores were generated, and clinical/pathologic variables were incorporated into uni- and multivariate Cox regression analyses to construct prognosis models. Time-ROC, time-C index, decision curve analysis (DCA), and calibration curves assessed the predictive performance of Fong score and pre- and post-NAT models for overall survival (OS) and disease-free survival (DFS) in a testing set. The final models included four variables for OS and three variables for DFS. The post-NAT models outperformed the pre-NAT models in time-ROC, time-C index, calibration, and DCA analysis, except for the 1-year DFS area under the curve (AUC). The Fong score models underperformed. The post-NAT OS RAD score effectively stratified patients into prognostic subgroups. The radiomic models incorporating pre- and post-NAT MRI features and clinical/pathologic variables effectively stratified CRLM patients prognositically. The post-NAT models demonstrated superior performance.

Diagnostic value and efficacy of multimodal magnetic resonance imaging in differentiating radiation necrosis from tumor recurrence in glioblastomas.

Distinguishing radiation necrosis (RN) from recurrent tumor (RT) in patients with gliomas treated with radiation therapy presents an important clinical dilemma. To evaluate the diagnostic performance of multiparametric magnetic resonance imaging (MRI) techniques in distinguishing RN from RT in patients with histologically confirmed glioma treated previously with radiotherapy and chemotherapy or without chemotherapy using a combination of dynamic susceptibility-weighted contrast-enhanced (DSC) perfusion MRI, diffusion tensor imaging (DTI), and MR spectroscopy (MRS). Patients with glioma who developed a new enhancing mass after standard treatment were retrospectively evaluated. Conventional MRI, DTI, DSC, and MRS were performed. The region of interest (ROI) was manually drawn in the enhancing lesions, peri-lesional white matter edema, and the contralateral normal-appearing white matter. Apparent diffusion coefficient (ADC), fractional anisotropy (FA), relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), N-acetylaspartate (NAA), choline (Cho), creatine (Cr), NAA/Cr, Cho/NAA, and Cho/Cr were calculated. Two-tailed t-test and receiver operating characteristic (ROC) curve analysis were performed. In total, 34 patients with RT and 25 with RN met our inclusion criteria. FA, rCBF, rCBV, Cho/NAA, Cho/Cr were statistically significant between the two groups (P < 0.05). The sensitivity and specificity of FA, rCBF, rCBV, Cho/NAA, and Cho/Cr in the diagnosis of RT were 70.6%, 97.1%, 91.2%, 91.2%, and 82.4% and 64%, 100%, 100%, 96%, and 72% respectively. DTI, DSC, and MRS are of great value in the differential diagnosis of RN and RT of glioma. The diagnostic performance of DSC is better than DTI and MRS.

Inter-reader agreement for diagnosing thymic cysts on chest MRI in two tertiary referral centers.

Despite the substantial role of chest MRI for the diagnosis and follow-up of thymic cysts, information about inter-reader agreement and optimal MR sequences is still limited. We aimed to investigate the inter-reader agreement for diagnosing thymic cysts using various combinations of MR sequences and to assess the effect of the addition of CT on inter-reader agreement. A total of 76 anterior mediastinal lesions (≤ 30mm) from two tertiary referral hospitals (55 from Institution A and 21 from Institution B) who underwent chest CT and contrast-enhanced chest MR were included. Internal and external reading sets consisted of different combinations of MR sequences (pre- and post-contrast T1-weighted, T2-weighted, subtracted images, and diffusion-weighted imaging [DWI]/apparent diffusion coefficient [ADC] map) and CT. Four and three radiologists independently reviewed internal and external reading sets. The overall inter-reader agreement was moderate (κ = 0.50-0.57) for diagnosing cysts without significant differences between MR sequence combinations (all p-values > 0.05). The mean pairwise inter-reader agreement was the highest (κ = 0.65) when both the subtracted image and DWI/ADC map were provided. The addition of CT had no positive effect on the inter-reader agreement in the internal reading set (κ, from 0.57 to 0.50) but increased inter-reader agreement in the external reading set from moderate (κ = 0.48) to substantial (κ = 0.74). In conclusion, the overall inter-reader agreement for diagnosing thymic cysts on MRI was moderate. MR sequences including both the subtracted image and DWI/ADC map may be optimal in terms of inter-reader agreement.

Multiparametric MRI and transfer learning for predicting positive margins in breast-conserving surgery: a multi-center study.

This study aimed to predict positive surgical margins in breast-conserving surgery (BCS) using multiparametric MRI (mpMRI) and radiomics. A retrospective analysis was conducted on data from 444 BCS patients from three Chinese hospitals between 2019 and 2024, divided into four cohorts and five datasets. Radiomics features from preoperative mpMRI, along with clinicopathological data, were extracted and selected using statistical methods and LASSO logistic regression. Eight machine learning classifiers, integrated with a transfer learning (TL) method, were applied to enhance model generalization. The model achieved an AUC of 0.889 in the internal test set and 0.771 in the validation set. Notably, TL significantly improved performance in two external validation sets, increasing the AUC from 0.533 to 0.902 in XAH and from 0.359 to 0.855 in YNCH. These findings highlight the potential of combining mpMRI and TL to provide accurate predictions for positive surgical margins in BCS, with promising implications for broader clinical application across multiple hospitals.

Imaging and Image Processing Techniques for High-Resolution Visualization of Connective Tissue with MRI: Application to Fascia, Aponeurosis, and Tendon

Recent interest in musculoskeletal connective tissues like tendons, aponeurosis, and deep fascia has led to a greater focus on in vivo medical imaging, particularly MRI. Given the rapid T2* decay of collagenous tissues, advanced ultra-short echo time (UTE) MRI sequences have proven useful in generating high-signal images of these tissues. To further these advances, we discuss the integration of UTE with Diffusion Tensor Imaging (DTI) and explore image processing techniques to enhance the localization, labeling, and modeling of connective tissues. These techniques are especially valuable for extracting features from thin tissues that may be difficult to distinguish. We present data from lower leg scans of 30 healthy subjects using a non-Cartesian MRI sequence to acquire axial 2D images to segment skeletal muscle and connective tissue. DTI helped differentiate aponeurosis from deep fascia by analyzing muscle fiber orientations. The dual echo imaging methods yielded high-resolution images of deep fascia, where in-plane spatial resolutions were between 0.3 × 0.3 mm to 0.5 × 0.5 mm with a slice thickness of 3–5 mm. Techniques such as K-Means clustering, FFT edge detection, and region-specific scaling were most effective in enhancing images of deep fascia, aponeurosis, and tendon to enable high-fidelity modeling of these tissues.

Advanced Multiparametric MRI Strategies for Tumor Restaging After Neoadjuvant Therapy in Locally Advanced Gastric Cancer.

This study tested the diagnostic accuracy of multiparametric magnetic resonance imaging (mpMRI) in restaging locally advanced gastric cancer after neoadjuvant therapy (NAT) using pathologic T stage (ypT) and pathologic N stage (ypN) as the reference standard. Between August 2022 and September 2023, the study enrolled a prospective cohort of 70 gastric cancer patients who underwent NAT and subsequent surgical resection. MRI procedures, including DLSB T2-weighted imaging (T2WI), ZOOMit diffusion-weighted imaging (DWI), and XD-VIBE dynamic contrast-enhanced imaging (DCE), were performed after NAT and before surgery. Four abdominal radiologists independently assigned radiologic T stage (yrT) and radiologic N stage (yrN) based on individual and combined sequences. Inter-reader agreement was quantified using Kendall's coefficient. Diagnostic accuracy was determined by comparing MRI assessments and pathologic outcomes, with pairwise comparisons analyzed via the McNemar test. Subgroup analysis evaluated the performance in identifying good responders to NAT. Inter-reader agreement was almost perfect for T restaging and substantial for N restaging. Diagnostic accuracy for T restaging was 0.432 using DLSB-T2WI, 0.586 using ZOOMit DWI, 0.557 using XD-VIBE DCE, and 0.586 using mpMRI. The accuracy demonstrated by DWI, DCE and mpMRI was superior to that of T2WI (all P < 0.05). For N restaging, the accuracy of the mpMRI protocol was 0.443. Notably, mpMRI achieved an AUC of 0.879 (95% confidence interval 0.835-0.915) for differentiating ypT0-1 tumors. Advanced mpMRI strategies can serve as a valuable tool for restaging gastric cancer after NAT. Accurately differentiating good responders to neoadjuvant therapy through mpMRI holds significant clinical implications for personalized treatment strategies and improved patient outcomes.

Computer-aided diagnosis based on 3D deep convolutional neural network system using novel 3D magnetic resonance imaging sequences for high-grade prostate cancer

Abstract Background With the rising incidence of prostate cancer (PCa), there is a global demand for assistive tools that aid in the diagnosis of high-grade PCa. This study aimed to develop a diagnostic support system for high-grade PCa using innovative magnetic resonance imaging (MRI) sequences in conjunction with artificial intelligence (AI). Materials and methods We examined image sequences of 254 patients with PCa obtained from diffusion-weighted and T2-weighted imaging, using novel MRI sequences before prostatectomy, to elucidate the characteristics of the 3-dimensional (3D) image sequences. The presence of PCa was determined based on the final diagnosis derived from pathological results after prostatectomy. A 3D deep convolutional neural network (3DCNN) was used as the AI for image recognition. Data augmentation was conducted to enhance the image dataset. High-grade PCa was defined as Gleason grade group 4 or higher. Results We developed a learning system using a 3DCNN as a diagnostic support system for high-grade PCa. The sensitivity and area under the curve values were 85% and 0.82, respectively. Conclusions The 3DCNN-based AI diagnostic support system, developed in this study using innovative 3D multiparametric MRI sequences, has the potential to assist in identifying patients at a higher risk of pretreatment of high-grade PCa.

Radiomics Analysis of Different Machine Learning Models based on Multiparametric MRI to Identify Benign and Malignant Testicular Lesions.

To develop and validate a machine learning-based prediction model for the use of multiparametric magnetic resonance imaging(MRI) to predict benign and malignant lesions in the testis. The study retrospectively enrolled 148 patients with pathologically confirmed benign and malignant testicular lesions, dividing them into: training set (n=103) and validation set (n=45). Radiomics characteristics were derived from T2-weighted(T2WI)、contrast-enhanced T1-weighted(CE-T1WI)、diffusion-weighted imaging(DWI) and Apparent diffusion coefficient(ADC) MRI images, followed by feature selection. A machine learning-based combined model was developed by incorporating radiomics scores (rad scores) from the optimal radiomics model along with clinical predictors. Draw the receiver operating characteristic (ROC) curve and use the area under the curve (AUC) to evaluate and compare the predictive performance of each model. The diagnostic efficacy of the various machine learning models was evaluated using the Delong test. Radiomics features were extracted from four sequence-based groups(CE-T1WI+DWI+ADC+T2WI), and the model that combined Logistic Regression(LR) machine learning showed the best performance in the radiomics model. The clinical model identified one independent predictors. The combined clinical-radiomics model showed the best performance, whose AUC value was 0.932(95% confidence intervals(CI)0.868-0.978), sensitivity was 0.875, specificity was 0.871 and accuracy was 0.884 in validation set. The combined clinical-radiomics model can be used as a reliable tool to predict benign and malignant testicular lesions and provide a reference for clinical treatment method decisions.

MR Imaging Techniques for Microenvironment Mapping of the Glioma Tumors: ASystematic Review.

The tumor microenvironment (TME) is a critical regulator of cancer progression, metastasis, and treatment response. Currently, various imaging approaches exist to assess the pathophysiological features of the TME. This systematic review provides an overview of magnetic resonance imaging (MRI) methods used in clinical practice to characterize the pathophysiological features of the gliomas TME. This review involved a systematic comprehensive search of original open-access articles reporting the clinical use of MR imaging in glioma patients of all ages in the PubMed, Scopus, and Web of Science databases between January 2010 and December 2023. We restricted our research to papers published in the English language. A total of 1137 studies were preliminarily identified through electronic database searches. After duplicate studies were removed, 44 studies met the eligibility criteria. The glioma TME was accompanied by alterations in metabolism, pH, vascularity, oxygenation, and extracellular matrix components, including tumor-associated macrophages, and sodium concentration. Multiparametric MRI is capable of noninvasively assessing the pathophysiological features and tumor-supportive niches of the TME, which is in line with its application in personalized medicine.

MMIT-DDPM – Multilateral medical image translation with class and structure supervised diffusion-based model

Unified translation of medical images from one-to-many distinct modalities is desirable in healthcare settings. A ubiquitous approach for bilateral medical scan translation is one-to-one mapping with GANs. However, its efficacy in encapsulating diversity in a pool of medical scans and performing one-to-many translation is questionable. In contrast, the Denoising Diffusion Probabilistic Model (DDPM) exhibits exceptional ability in image generation due to its scalability and ability to capture the distribution of whole training data. Therefore, we propose a novel conditioning mechanism for the deterministic translation of medical scans to any target modality from a source modality with a DDPM model. This model denoises the target modality under the guidance of a source-modality structure encoder and source-to-target class conditioner. Consequently, this mechanism serves as prior information for sampling the desired target modality during inference. The training and testing have been carried out on the T1-weighted, T2-weighted, and Fluid Attenuated Inversion Recovery (FLAIR) sequence of the BraTS 2021 dataset. The proposed model is capable of unified multi-lateral translation among six combinations of T1ce, T2, and FLAIR sequences of brain MRI, eliminating the need for multiple bilateral translation models. We have analyzed the performance of our architecture against State-of-the-art, Convolution, and Transformer-based GANs. The diffusion model efficiently covers the distribution of multiple modalities while producing better image quality of the translated sequences, as evidenced by the average improvement of 8.06 % in Multi-Scale Structural Similarity (MSSIM) and 2.52 in Fréchet Inception Distance (FID) metrics compared with the CNN and transformer-based GAN architecture.

Impact of software-assisted structured reporting on radiology residents approaching prostate MRI.

To evaluate the potential advantages of software-assisted structured reporting for radiology residents approaching multiparametric prostate MRI (mpMRI). MpMRI scans from 100 patients, performed for prostate cancer (PCa) detection or staging, were anonymized, and reviewed by six second-year radiology residents without previous experience in prostate MRI, following 6h of intensive training. The dataset was split into two subsets of 50 cases each. All residents were asked to report scans from the first subset using a basic text processor (narrative reports -NR-). For the second subset, one group used a dedicated software to produce structured reports (SR) while the other continued with NR. Report completeness was assessed using a PI-RADS-based checklist, and statistical analyses, including Wilcoxon rank sum and Pearson's Chi-squared tests, were performed to compare word count, reporting time, and concordance with an expert radiologist's findings. All readers adopting SR in the second batch demonstrated a significant increase in word count and a decrease in reporting time compared to the first batch. Image quality and final impressions were missing from all NR, while gland size, lesion description, and PI-RADS score were consistently included in nearly all reports (96-100%). One of the three residents using SR showed a statistically significant improvement in concordance with the expert radiologist on index lesion location and clinically significant PCa presence (p=0.001), while the other two exhibited positive trends (p=0.061-0.078). The adoption of SR allowed radiology residents to decrease their reporting time and improve the comprehensiveness of their reports, while increasing concordance with an expert radiologist.

Neuroimaging in Epilepsy.

The goal of this article is to review the use of neuroimaging in the treatment of new-onset seizures and epilepsy. The article will focus predominantly on MRI because this is the most common modality, but it will also review others that are often used in individuals being considered for epilepsy surgery. The article also reviews common causes of epilepsy and their appearance on imaging and various imaging tools that can be combined to assist in the care of people with epilepsy. MRI has revolutionized the ability to diagnose the cause of many forms of epilepsy. However, to maximize the diagnostic power of MRI, it is essential to order the correct imaging sequences. In this article, the harmonized neuroimaging of epilepsy structural sequences (HARNESS) MRI protocol proposed by the International League Against Epilepsy is discussed. MRI is the preferred imaging modality to identify lesions associated with epilepsy. Protocols should include thin-cut, no-gap sequences to permit the identification of small epileptogenic lesions, and studies should be reviewed with an understanding of all the clinical information to help guide the identification of potential lesions.

Deep learning-based overall survival prediction in patients with glioblastoma: An automatic end-to-end workflow using pre-resection basic structural multiparametric MRIs

PurposeAccurate and automated early survival prediction is critical for patients with glioblastoma (GBM) as their poor prognosis requires timely treatment decision-making. To address this need, we developed a deep learning (DL)-based end-to-end workflow for GBM overall survival (OS) prediction using pre-resection basic structural multiparametric magnetic resonance images (Bas-mpMRI) with a multi-institutional public dataset and evaluated it with an independent dataset of patients on a prospective institutional clinical trial. Materials and methodsThe proposed end-to-end workflow includes a skull-stripping model, a GBM sub-region segmentation model and an ensemble learning-based OS prediction model. The segmentation model utilizes skull-stripped Bas-mpMRIs to segment three GBM sub-regions. The segmented GBM is fed into the contrastive learning-based OS prediction model to classify the patients into different survival groups. Our datasets include both a multi-institutional public dataset from Medical Image Computing and Computer Assisted Intervention (MICCAI) Brain Tumor Segmentation (BraTS) challenge 2020 with 235 patients, and an institutional dataset from a 5-fraction SRS clinical trial with 19 GBM patients. Each data entry consists of pre-operative Bas-mpMRIs, survival days and patient ages. Basic clinical characteristics are also available for SRS clinical trial data. The multi-institutional public dataset was used for workflow establishing (90% of data) and initial validation (10% of data). The validated workflow was then evaluated on the institutional clinical trial data. ResultsOur proposed OS prediction workflow achieved an area under the curve (AUC) of 0.86 on the public dataset and 0.72 on the institutional clinical trial dataset to classify patients into 2 OS classes as long-survivors (>12 months) and short-survivors (<12 months), despite the large variation in Bas-mpMRI protocols. In addition, as part of the intermediate results, the proposed workflow can also provide detailed GBM sub-regions auto-segmentation with a whole tumor Dice score of 0.91. ConclusionOur study demonstrates the feasibility of employing this DL-based end-to-end workflow to predict the OS of patients with GBM using only the pre-resection Bas-mpMRIs. This DL-based workflow can be potentially applied to assist timely clinical decision-making.