Journal of Magnetic Resonance Imaging最新文献

{"title":"Editorial for \"Personalized Cine Cardiac MRI Protocol Optimization Using a Retrieval-Augmented Large Language Model: A Prospective Comparison With Technologist-Optimized Protocol\".","authors":"Mohammad H Jalili, Hamid Chalian, Arash Bedayat","doi":"10.1002/jmri.70576","DOIUrl":"https://doi.org/10.1002/jmri.70576","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149127692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Slice-Wise Quality Assessment of High b Value Breast DWI via Deep Learning-Based Artifact Detection.","authors":"Ameya Markale, Luise Brock, Ihor Horishnyi, Dominika Skwierawska, Tri-Thien Nguyen, Dominique Hadler, Hannes Schreiter, Shirin Heidarikahkesh, Lorenz A Kapsner, Michael Uder, Sabine Ohlmeyer, Frederik B Laun, Andrzej Liebert, Sebastian Bickelhaupt","doi":"10.1002/jmri.70572","DOIUrl":"https://doi.org/10.1002/jmri.70572","url":null,"abstract":"<p><strong>Background: </strong>Diffusion-weighted imaging (DWI) can support lesion detection and characterization in breast MRI. Therefore, DWI is increasingly being incorporated into breast MRI protocols to address some of the shortcomings of routine clinical breast MRI. However, especially high b value DWI can be prone to intensity artifacts that can affect diagnostic image assessment.</p><p><strong>Purpose: </strong>To detect hyper- and hypointense artifacts on high b value DWI (b = 1500 s/mm²) using deep learning, employing either a binary classification (artifact presence) or a multiclass classification (artifact intensity) approach on a slice-wise dataset.</p><p><strong>Study type: </strong>Retrospective.</p><p><strong>Study population: </strong>11,806 DWI slices were acquired from 156 examinations performed on female patients between 2022 and mid-2023. For both hyper- and hypointense artifact classification, the slices were split into test, validation, and holdout test set with 8164/1806/1836 and 8164/1820/1822 slices in these sets respectively.</p><p><strong>Field strength/sequence: </strong>3 T spin-echo planar diffusion-weighted sequence with b values: 50, 750, and 1500 s/mm².</p><p><strong>Assessment: </strong>Three convolutional neural network (CNN) architectures (DenseNet121, ResNet18, and SEResNet50) were trained for binary classification of hyper- and hypointense artifacts on slice-wise labeled data. The best performing model (DenseNet121) was applied to an independent holdout test set and further trained separately for multiclass classification.</p><p><strong>Statistical tests: </strong>Performance of the networks was evaluated using accuracy, precision, recall, and areas under the receiver operating characteristic curve (AUROC) and under the precision recall curve (AUPRC). Radiologist evaluated bounding box positions on a 5-point Likert-like scale across 150 slices for validation and 200 slices for the holdout test set, derived from the network's Grad-CAM heatmaps.</p><p><strong>Results: </strong>DenseNet121 achieved AUROCs of 0.91 and 0.95 for hyper- and hypointense artifact detection, respectively, and weighted AUROCs of 0.84 and 0.90 for multiclass classification on single-slice high b value diffusion-weighted images, achieving mean scores of 3.52 ± 0.81 for hyperintense artifacts and 3.62 ± 0.65 for hypointense artifacts.</p><p><strong>Data conclusion: </strong>Hyper- and hypointense artifact detection in slice-wise breast DWI MRI dataset (b = 1500 s/mm²) using CNNs, particularly DenseNet121, suggests potential reliability and requires further validation.</p><p><strong>Evidence level: </strong>3.</p><p><strong>Technical efficacy: </strong>Stage 3.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial for \"Slice-Wise Quality Assessment of High b-Value Breast DWI via Deep Learning-Based Artifact Detection\".","authors":"Ziyu Fu","doi":"10.1002/jmri.70574","DOIUrl":"https://doi.org/10.1002/jmri.70574","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of Pre-Operative Cartilage and Meniscus Ultrashort Echo Time T2* as Biomarkers of Long-Term Clinical Outcomes Following Anterior Cruciate Ligament Reconstruction.","authors":"Ashley A Williams, Nicholas P Drain, Jayme C B Koltsov, Yongxian Qian, Kathryn J Stevens, Constance R Chu","doi":"10.1002/jmri.70571","DOIUrl":"https://doi.org/10.1002/jmri.70571","url":null,"abstract":"<p><strong>Background: </strong>Ultrashort echo time T2* (UTE-T2*) mapping is sensitive to the composition of deep cartilage and menisci. UTE-T2* elevations are common after anterior cruciate ligament (ACL) injuries, but their long-term prognostic importance is unknown.</p><p><strong>Hypothesis: </strong>Higher pre-operative UTE-T2* values will associate with greater symptoms and radiographic knee osteoarthritis (KOA) at long-term follow-up.</p><p><strong>Study type: </strong>Retrospective longitudinal cohort study using prospectively acquired baseline MRI and follow-up clinical outcomes.</p><p><strong>Population: </strong>Twenty-two participants (13 female, mean [SD] age: 29 [9] years) with ACL reconstruction (ACLR).</p><p><strong>Field strength/sequence: </strong>3T, 3D acquisition-weighted stack of spirals.</p><p><strong>Assessment: </strong>Pre-operative UTE-T2* maps from participants who underwent ACLR (2009-2012) were compared to 11-year follow-up Knee Injury and Osteoarthritis Outcome Scores (KOOS), the Marx Activity Rating Scale scores, and radiographs.</p><p><strong>Statistical tests: </strong>Linear regression models assessed relationships between pre-operative cartilage and meniscus UTE-T2* and follow-up KOOS Symptoms or Marx scores. Mann-Whitney U-tests assessed differences in pre-operative UTE-T2* by radiographic KOA status (Kellgren-Lawrence grade ≥ 2).</p><p><strong>Significance: </strong>two-sided α < 0.05.</p><p><strong>Results: </strong>KOOS and Marx scores were available for 22 participants at a median of 11.3 years (range: 10.3-12.4) after ACLR; radiographs were available for 11 participants. After adjustment for meniscus tear status, higher pre-operative UTE-T2* values in the deep medial femoral cartilage and medial meniscus were associated with worse KOOS Symptoms at 11 years (adjusted R² = 0.146-0.332). Higher femoral cartilage UTE-T2* trended with lower Marx scores (adjusted R² = 0.208, p = 0.056), while higher posterior meniscus UTE-T2* was associated with higher Marx scores (adjusted R² = 0.308). In the small radiographic subset, median pre-operative posterior meniscus UTE-T2* was 69% higher in patients with radiographic KOA.</p><p><strong>Data conclusion: </strong>Pre-operative UTE-T2* in medial compartment cartilage and menisci were associated with knee symptoms and activity levels 11 years after ACLR. These findings support the potential utility of elevated cartilage and meniscus UTE-T2* as markers of pre-osteoarthritis reflecting symptomatic and structural KOA risk after ACLR.</p><p><strong>Level of evidence: </strong>Level 3.</p><p><strong>Technical efficacy: </strong>Stage 3 Diagnostic Thinking.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13641164/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial for \"Evaluation of Pre-Operative Cartilage and Meniscus Ultrashort Echo Time T2* as Biomarkers of Long-Term Clinical Outcomes Following Anterior Cruciate Ligament Reconstruction\".","authors":"Dimitri A Kessler","doi":"10.1002/jmri.70563","DOIUrl":"https://doi.org/10.1002/jmri.70563","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial for \"Diagnostic Performance of a Metasurface Coil Compared With a Commercial 12-Channel Coil for Knee MRI at 1.5 T\".","authors":"Scott N Hwang, Cristy N French","doi":"10.1002/jmri.70579","DOIUrl":"https://doi.org/10.1002/jmri.70579","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149109924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalized Cine Cardiac MRI Protocol Optimization Using a Retrieval-Augmented Large Language Model: A Prospective Comparison With Technologist-Optimized Protocol.","authors":"Sadegh Dehghani, Ali Farashahi, Ibtihal Raheem Hammood","doi":"10.1002/jmri.70566","DOIUrl":"https://doi.org/10.1002/jmri.70566","url":null,"abstract":"<p><strong>Background: </strong>Cine cardiac magnetic resonance imaging (MRI) protocols are typically based on static templates, which may not optimally address individual patient limitations such as arrhythmia or limited breath-hold capacity.</p><p><strong>Purpose: </strong>To test whether an artificial intelligence (AI) framework using a large language model with retrieval-augmented generation (LLM-RAG) can provide cine cardiac MRI protocols that are equivalent or noninferior to those of technologist-optimized protocols.</p><p><strong>Study type: </strong>Prospective, intra-individual comparative, nonrandomized.</p><p><strong>Population/subjects: </strong>Sixty-eight patients (mean age 46 ± 18 years; 39 women) referred for clinical cardiac MRI, including patients with normal function as well as those with arrhythmia, implants, or limited breath-hold capacity.</p><p><strong>Field strength/sequence: </strong>1.5 T, cine cardiac MRI.</p><p><strong>Assessment: </strong>Each participant underwent cine cardiac MRI twice in one session. First with a technologist-optimized protocol, then with an AI-optimized protocol (fixed order). Image quality and artifact suppression were assessed using four-point Likert scales by two blinded readers. Left ventricular (LV) functional parameters and repeat acquisitions were recorded.</p><p><strong>Statistical tests: </strong>Wilcoxon signed-rank test, McNemar's test, paired t-tests, and Bland-Altman analysis. Significance set at p < 0.05.</p><p><strong>Results: </strong>Compared to technologist-, AI-optimized cine protocols demonstrated significantly higher image quality (3.69 ± 0.47 vs. 3.00 ± 0.49, p < 0.001) and artifact suppression (3.19 ± 0.35 vs. 2.17 ± 0.45, p < 0.001). The AI-optimized protocol reduced repeat acquisitions by 68% (6 vs. 19 repeats, p = 0.028). No significant differences were observed in LV functional parameters between both protocols (all p > 0.05). Bland-Altman analysis showed minimal bias for all LV functional parameters.</p><p><strong>Data conclusion: </strong>In this nonrandomized study, a LLM-RAG framework generated cine cardiac MRI protocols that were noninferior to technologist-optimized protocols for image quality and demonstrated high agreement for LV functional measurements with fewer repeat acquisitions. These findings demonstrated the feasibility of AI-guided patient-adaptive protocoling. Randomized trials are warranted.</p><p><strong>Evidence level: </strong>2.</p><p><strong>Technical efficacy: </strong>Stage 2.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110770","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diffusion Basis Spectrum Imaging (DBSI): A Multicompartment Diffusion Model Linking Regional Microstructural Heterogeneity With Proliferation and Survival in High-Grade Gliomas.","authors":"Nengjin Zhu, Liwei Mazu, Zuqi Xia, Siyuan He, Yingqian Huang, Rulin Xu, Longfei Zhao, Shanmei Zeng, Kejun He, Jianping Chu","doi":"10.1002/jmri.70573","DOIUrl":"https://doi.org/10.1002/jmri.70573","url":null,"abstract":"<p><strong>Background: </strong>Diffusion basis spectrum imaging (DBSI) has emerged as an advanced model capable of characterizing central nervous system tissue injury and glioma pathology. However, the associations of DBSI with key characteristics of high-grade gliomas (HGGs), including proliferation and survival outcomes, remain underexplored.</p><p><strong>Purpose: </strong>To evaluate the associations between DBSI-derived histogram features with Ki-67 level and overall survival (OS) in HGGs.</p><p><strong>Study type: </strong>Retrospective.</p><p><strong>Subjects: </strong>Ninety-one patients (36 women, 53.41 ± 12.62 years) with HGGs.</p><p><strong>Field/sequence: </strong>3-T, spin-echo echo-planar imaging (SE-EPI) sequence with q-space Cartesian grid sampling.</p><p><strong>Assessment: </strong>DBSI was used to estimate restricted fraction, hindered fraction, water fraction, anisotropic fraction, and mean isotropic apparent diffusion coefficient (iso ADC). Histogram features were extracted from both the contrast-enhancing regions (CER) and non-enhancing T2-weighted imaging (T2WI)/fluid attenuated inversion recovery (FLAIR) hyperintense regions (NEHR). Patients were stratified by Ki-67 labeling index (LI), and OS was defined as the time from surgery to death or last follow-up.</p><p><strong>Statistical tests: </strong>Receiver operating characteristic (ROC) curve analysis and Cox regression were applied to evaluate features in relation to Ki-67 LI and OS, respectively. p < 0.05 was considered significant.</p><p><strong>Results: </strong>In HGGs, CER showed higher restricted and anisotropic fractions, whereas NEHR showed a higher hindered fraction, water fraction, and iso ADC. For Ki-67 assessment, the robust mean absolute deviation (RMAD) of restricted fraction in CER was the best single feature (AUC = 0.714), and the combination of DBSI features improved the AUC to 0.791. Furthermore, DBSI features from NEHR, together with multifocality, were significant factors associated with OS.</p><p><strong>Data conclusion: </strong>DBSI-derived histogram analysis provides a noninvasive quantitative characterization of HGG heterogeneity. Proliferative activity may be primarily reflected within CER, while survival outcomes may be related to DBSI features in NEHR.</p><p><strong>Evidence level: </strong>3.</p><p><strong>Technical efficacy: </strong>Stage 3.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multicenter, Multivendor Development and Validation of Automated Liver Prescription.","authors":"Garrett C Fullerton, Jitka Starekova, Collin J Buelo, David T Harris, Raphael do Vale Souza, Aaron Faacks, Alexandra A Anagnostopoulos, Veronica P Christofi, Austin Murphy, Diana Kadi, Takeshi Yokoo, Mustafa R Bashir, Scott B Reeder, Diego Hernando","doi":"10.1002/jmri.70575","DOIUrl":"https://doi.org/10.1002/jmri.70575","url":null,"abstract":"<p><strong>Background: </strong>Manual prescription of liver MRI volumes introduces inter-operator variability and prolongs exam duration. Artificial intelligence (AI)-based prescription has been demonstrated, but existing work has been limited to single-vendor settings.</p><p><strong>Purpose: </strong>To develop and validate an AI model for automated liver MRI prescription using a large, consecutive, multicenter dataset encompassing multiple vendors, field strengths, localizer sequences, and liver pathologies.</p><p><strong>Study type: </strong>Retrospective.</p><p><strong>Population: </strong>A total of 11,012 patients (5928 female) who underwent liver MRI exams across three vendors (GE HealthCare, Philips Healthcare, Siemens Healthineers) at three institutions, split into training/validation/test sets (70/10/20%). Liver disorders included cirrhosis, ascites, portal hypertension, and iron overload.</p><p><strong>Field strength/sequence: </strong>Three-plane localizers acquired using single-shot fast spin echo (68.3%) and gradient echo (31.7%) at 1.5 T (52.8%) and 3 T (47.2%).</p><p><strong>Assessment: </strong>A YOLOv8 object detection model was trained to detect the liver, torso, and arms on localizer images. Automated prescription accuracy was assessed using three-dimensional intersection-over-union (IoU_3D) for axial, coronal, and sagittal prescriptions. Subgroup analyses evaluated performance across technical and patient-related factors.</p><p><strong>Statistical tests: </strong>Mann-Whitney U tests with Bonferroni correction, Kruskal-Wallis tests, Wilcoxon signed-rank tests, Spearman correlation. Effect sizes were calculated using rank-biserial correlation (r_rb) and epsilon-squared (ε²). p < 0.05 was considered statistically significant.</p><p><strong>Results: </strong>Median axial prescription IoU_3D was 0.963 (interquartile range: 0.946-0.976), with median IoU_3D above 0.940 for coronal and sagittal prescriptions. For axial prescriptions, margins less than 4 mm in each direction were required to achieve complete coverage of manual prescriptions in 90% of cases in the held-out test set. Effects were negligible to small across vendors, field strengths, localizer sequences, and patient factors (r_rb ≤ 0.16, ε² ≤ 0.01).</p><p><strong>Data conclusion: </strong>A single multivendor model achieved accurate automated liver MRI prescription without clinically meaningful performance differences across vendors, field strengths, localizer sequences, or patient factors.</p><p><strong>Evidence level: </strong>2.</p><p><strong>Technical efficacy: </strong>Stage 1.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial for \"Multicenter, Multivendor Development and Validation of Automated Liver Prescription\".","authors":"Grace McIlvain, Benjamin Sloan","doi":"10.1002/jmri.70577","DOIUrl":"https://doi.org/10.1002/jmri.70577","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149110667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}