Journal of Magnetic Resonance Imaging最新文献

{"title":"Magnetic Resonance Imaging for Dental Pulp Assessment: A Comprehensive Review.","authors":"Bing Han, Na Chen, Jin Luo, Farzaneh Afkhami, Ove A Peters, Xiaoyan Wang","doi":"10.1002/jmri.29742","DOIUrl":"https://doi.org/10.1002/jmri.29742","url":null,"abstract":"<p><p>Magnetic resonance imaging (MRI) has recently emerged as a promising modality for dental applications, offering radiation-free imaging with superior soft tissue visualization capabilities compared to x-ray-based techniques such as spiral or cone beam computed tomography (CBCT). Conventional radiographic methods or CBCT cannot directly assess the condition of the dental pulp due to their primary focus on hard tissue visualization, whereas the dental pulp is primarily composed of connective tissue. Given the advantages of MRI in soft tissue imaging, this review aims to explore the current application of MRI for dental pulp tissue assessment. Relevant studies concerning the application of MRI for visualizing dental pulp were retrieved from databases including PubMed, Embase, and Scopus. The review explored and discussed the advancements in MRI hardware and software related to dental pulp visualization, as well as the advantages and limitations of MRI in dental pulp studies. Despite remaining limitations, such as scanning time and cost considerations, MRI offers notable benefits, including radiation-free imaging and potentially superior resolution and accuracy compared with other imaging techniques. Consequently, the continued advancement of MRI as a noninvasive diagnostic method in dentistry, particularly for assessing pulp condition, holds substantial promise for improving endodontic diagnosis and subsequent treatment decision-making.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425474","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 \"High-Fidelity MRI Assessment of Cerebral Perfusion in Healthy Neonates Less Than 1 Week of Age\".","authors":"Masaaki Hori, Kei Nakahara, Masahiro Kobayashi","doi":"10.1002/jmri.29745","DOIUrl":"https://doi.org/10.1002/jmri.29745","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425471","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":"Pre-Treatment Breast MRI: Clinical Indications, Outcomes, and Future Directions.","authors":"Brian N Dontchos, Matthew D Phelps, Habib Rahbar, Diana L Lam","doi":"10.1002/jmri.29741","DOIUrl":"https://doi.org/10.1002/jmri.29741","url":null,"abstract":"<p><p>Breast MRI is the most sensitive modality for assessing the extent of disease in patients with newly-diagnosed breast cancer because it identifies clinically- and mammographically-occult breast cancers. Though highly sensitive, breast MRI has lower specificity that may result in false positive findings and potential overestimation of disease if additional MRI findings are not biopsied prior to surgery. It had been anticipated that the superior cancer detection rate of pre-treatment MRI would translate to improved immediate (surgical re-excision) and long-term patient outcomes such as breast cancer recurrence and survival rates, but studies have not necessarily supported this assumption. In this review, current recommendations and utilization of breast MRI for pre-treatment local staging of breast cancer will be presented, with an emphasis on specific clinical scenarios for patient selection and its impact on short- and long-term patient clinical outcomes. We will also present new evidence that pre-treatment MRI may support de-escalation of treatment and discuss emerging advanced MRI techniques that may improve diagnostic performance.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425412","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":"High Impact Clinical Applications of Cardiac Magnetic Resonance Imaging in Women: A Review.","authors":"Alex Diaz, Chelsea Meloche, Mohamed Abdelmotleb, Hamid Chalian, Ana Paula Santos Lima, Luba Frank, Karen Ordovas","doi":"10.1002/jmri.29736","DOIUrl":"https://doi.org/10.1002/jmri.29736","url":null,"abstract":"<p><p>The diagnosis of cardiovascular disease in women poses an ongoing challenge due to lack of knowledge about sex differences in the manifestations of cardiovascular disease, since women have been underrepresented in cardiovascular research studies that guide current practice. The purpose of this article is to review a spectrum of cardiovascular disorders which occur exclusively or more frequently in women and to highlight the role that cardiovascular magnetic resonance (MR) plays in diagnosing and prognosticating these disorders. Specifically, this review focuses on cardio-oncologic, ischemic, inflammatory, autoimmune, peri-partum, and genetic manifestations of cardiomyopathy in women. We strive to draw attention to the added diagnostic value provided by cardiac MR, compared against alternative imaging modalities, and propose opportunities for further research on sex differences in imaging and diagnosing cardiovascular diseases. EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 3.</p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143408709","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":"High-Fidelity MRI Assessment of Cerebral Perfusion in Healthy Neonates Less Than 1 Week of Age.","authors":"Zhiyi Hu, Dengrong Jiang, Jennifer Shepard, Yuto Uchida, Kenichi Oishi, Wen Shi, Peiying Liu, Doris Lin, Vivek Yedavalli, Aylin Tekes, William Christopher Golden, Hanzhang Lu","doi":"10.1002/jmri.29740","DOIUrl":"10.1002/jmri.29740","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Perfusion imaging of the brain has important clinical applications in detecting neurological abnormalities in neonates. However, such tools have not been available to date. Although arterial-spin-labeling (ASL) MRI is a powerful noninvasive tool to measure perfusion, its application in neonates has encountered obstacles related to low signal-to-noise ratio (SNR), large-vessel contaminations, and lack of technical development studies.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;To systematically develop and optimize ASL perfusion MRI in healthy neonates under 1 week of age.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study type: &lt;/strong&gt;Prospective.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Subjects: &lt;/strong&gt;Thirty-two healthy term neonates (19 female; postnatal age 1.9 ± 0.7 days).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Field strength/sequence: &lt;/strong&gt;3.0 T; T&lt;sub&gt;2&lt;/sub&gt;-weighted half-Fourier single-shot turbo-spin-echo (HASTE) imaging, single-delay and multi-delay 3D gradient-and-spin-echo (GRASE) large-vessel-suppression pseudo-continuous ASL (LVS-pCASL).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Assessment: &lt;/strong&gt;Three studies were conducted. First, an LVS-pCASL MRI sequence was developed to suppress large-vessel spurious signals in neonatal pCASL. Second, multiple post-labeling delays (PLDs) LVS-pCASL were employed to simultaneously estimate normative cerebral blood flow (CBF) and arterial transit time (ATT) in neonates. Third, an enhanced background-suppression (BS) scheme was developed to increase the SNR of neonatal pCASL.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Statistical tests: &lt;/strong&gt;Repeated measure analysis-of-variance, paired t-test, spatial intraclass-correlation-coefficient (ICC), and voxel-wise coefficient-of-variation (CoV). P-value &lt;0.05 was considered significant.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;LVS-pCASL reduced spurious ASL signals, making the CBF images more homogenous and significantly reducing the temporal variation of CBF measurements by 58.0% when compared to the standard pCASL. Multi-PLD ASL yielded ATT and CBF maps showing a longer ATT and lower CBF in the white matter relative to the gray matter. The highest CBF was observed in basal ganglia and thalamus (10.4 ± 1.9 mL/100 g/min). Enhanced BS resulted in significantly higher test-retest reproducibility (ICC = 0.90 ± 0.04, CoV = 8.4 ± 1.2%) when compared to regular BS (ICC = 0.59 ± 0.12, CoV = 23.6 ± 3.8%).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Data conclusion: &lt;/strong&gt;We devised an ASL method that can generate whole-brain CBF images in 4 minutes with a test-retest image ICC of 0.9. This technique holds potential for studying neonatal brain diseases involving perfusion abnormalities.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Plain language summary: &lt;/strong&gt;MR imaging of cerebral blood flow in neonates remains a challenge due to low blood flow rates and confounding factors from large blood vessels. This study systematically developed an advanced MRI technique to enhance the reliability of perfusion measurements in neonates. The proposed method reduced signal artifacts from large blood vessels and improved the signal-to-noi","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143408711","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 \"Diagnosis of Sacroiliitis Through Semi-Supervised Segmentation and Radiomics Feature Analysis of MRI Images\".","authors":"Eros Montin","doi":"10.1002/jmri.29732","DOIUrl":"https://doi.org/10.1002/jmri.29732","url":null,"abstract":"","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143399027","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":"Myocardial MRI Cine Radiomics: A Novel Approach to Risk-Stratification for Major Adverse Cardiovascular Events in Patients With ST-Elevation Myocardial Infarction.","authors":"Ming-Lei Li, Ruo-Yang Shi, Jin-Yu Zheng, Jin-Yi Xiang, Ward Hedges, Julia Liang, Jiani Hu, Jie Chen, Lei Zhao, Lian-Ming Wu","doi":"10.1002/jmri.29739","DOIUrl":"https://doi.org/10.1002/jmri.29739","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;The incremental prognostic value of integrating myocardial cine radiomics into predictive models for major adverse cardiovascular events (MACE) risk in patients with ST-elevation myocardial infarction (STEMI) is unclear.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;To determine if myocardial cine radiomics can improve risk assessment for MACE when combined with clinical information and cardiac MRI parameters in STEMI patients.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study type: &lt;/strong&gt;Retrospective.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Subjects: &lt;/strong&gt;One thousand twenty-four STEMI patients (83% male; mean age 59 ± 11 years) from two centers, divided into training (819 patients) and external testing (205 patients) cohorts.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Field strength/sequence: &lt;/strong&gt;3.0 T/balanced steady-state free precession cine, and phase-sensitive inversion recovery sequences.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Assessment: &lt;/strong&gt;The Rad_score was calculated as a weighted sum of independent radiomic variables derived from the logistic regression model, providing a concise representation of their combined prognostic impact. Six risk models were developed, incorporating varying combinations of MRI parameters, clinical variables, and Rad_score to comprehensively evaluate their prognostic performance. A final risk stratification, integrating left ventricular ejection fraction (LVEF), the extent of late gadolinium enhancement (LGE), and Rad_score, was established and compared with one based on LVEF alone.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Statistical tests: &lt;/strong&gt;The prognostic implications of the Rad_score were evaluated using univariable and multivariable Cox proportional hazards models. A P value &lt;0.05 was considered significant.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;During a median follow-up of 3.1 years, 139 patients (17%) in the training set and 30 patients (15%) in the testing set experienced MACE. Rad_score was identified as a significant risk factor for MACE, with a hazard ratio of 1.46 (1.38-1.55) (P &lt; 0.01) in univariate Cox analysis. The risk stratification reclassified the risk for 33% of the study population in the training set and 34% in the testing set.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Data conclusion: &lt;/strong&gt;Myocardial cine radiomics are associated with MACE risk in STEMI patients and provide incremental improvement in risk stratification when combined with traditional parameters.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Plain language summary: &lt;/strong&gt;The development of radiomics has introduced new perspectives in both the diagnosis and prognosis of cardiovascular diseases. However, the incremental prognostic value of incorporating myocardial cine radiomics into predictive models for major adverse cardiovascular events (MACE) risk in patients with ST-elevation myocardial infarction (STEMI) remains unclear. This study integrates radiomics with traditional clinical parameters and cardiac magnetic resonance imaging (MRI) to evaluate its added value in assessing MACE risk in STEMI patients. The results demonstrate that radiomics is significantly associat","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143391062","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":"Deep Learning-Based Techniques in Glioma Brain Tumor Segmentation Using Multi-Parametric MRI: A Review on Clinical Applications and Future Outlooks","authors":"Delaram J. Ghadimi MD,&nbsp;Amir M. Vahdani MD,&nbsp;Hanie Karimi MD, MPH,&nbsp;Pouya Ebrahimi MD,&nbsp;Mobina Fathi MD, MPH,&nbsp;Farzan Moodi MD,&nbsp;Adrina Habibzadeh MD,&nbsp;Fereshteh Khodadadi Shoushtari MS,&nbsp;Gelareh Valizadeh PhD,&nbsp;Hanieh Mobarak Salari MS,&nbsp;Hamidreza Saligheh Rad PhD","doi":"10.1002/jmri.29737","DOIUrl":"https://doi.org/10.1002/jmri.29737","url":null,"abstract":"<p>CLINICAL APPLICATIONS OF DEEP LEARNING-BASED SEGMENTATION FOR GLIOMA. MRI IMAGE ADAPTED FROM BRATS 2020. BY GHADIMI ET AL.(1094-1109)\u0000\u0000 <figure>\u0000 <div><picture>\u0000 <source></source></picture><p></p>\u0000 </div>\u0000 </figure>\u0000 </p>","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":"61 3","pages":"spcone"},"PeriodicalIF":3.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jmri.29737","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143362936","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":"MRI Assessment of Geometric Microstructural Changes of White Matter in Infants With Periventricular White Matter Injury and Spastic Cerebral Palsy.","authors":"Miaoyan Wang, Hua Zhu, Tingting Huang, Jingjing Qiao, Bo Peng, Ni Shu, Anqi Qiu, Jian Cheng, Haoxiang Jiang","doi":"10.1002/jmri.29730","DOIUrl":"https://doi.org/10.1002/jmri.29730","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Periventricular white matter injury (PWMI) is a high-risk factor for spastic cerebral palsy (SCP).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;To investigate the geometric microstructural changes in WM in infants with PWMI-SCP using MRI which may facilitate early identification.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study type: &lt;/strong&gt;Retrospective cohort study.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Population: &lt;/strong&gt;Twenty-three healthy infants (aged 6.53-36 months), 25 infants with PWMI-SCP (aged 6-33 months), and 32 infants with PWMI-nonSCP (aged 6-36 months).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Field strength/sequence: &lt;/strong&gt;3.0 T, T1-weighted three-dimensional gradient-echo sequence, and diffusion tensor imaging (DTI) with a single-shot gradient echo planar sequence.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Assessment: &lt;/strong&gt;The brain was automatically segmented, parcellated into major regions of interest according to the Desikan-Killiany atlas and volumes extracted. Fractional anisotropy (FA) and mean diffusivity (MD) of regions were extracted from DTI data. Director field analysis (DFA) was used to assess the geometric microstructural properties of WM. Motor dysfunction was graded from l (mild) to 5 (severe) according to the Gross Motor Function Classification System.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Statistical tests: &lt;/strong&gt;Tests included analysis of variance, correlation analysis, mediation analysis, and receiver operating characteristic analysis. Corrected P-values &lt;0.05 were considered significant. Mediation analysis examined whether DFA metrics mediated the relationship between brain morphological and motor dysfunction. Models were constructed to identify PWMI-SCP.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The PWMI-SCP group exhibited significantly elevated all four DFA metrics (splay, bend, twist, and distortion), primarily in the corpus callosum, posterior thalamic radiata, and corona radiata, compared to the PWMI-nonSCP group, and was associated with enlarged lateral ventricles, reduced deep nuclear volumes and motor dysfunction. Mediation analysis indicated that increased splay in the corpus callosum partially mediates (mediating effect ratio: 29.74%, 22.46%) the relationship between the lateral ventricles and motor function. The results showed that DFA achieved a higher area under the curve (AUC) than the FA + MD, especially in distinguishing PWMI-nonSCP from PWMI-SCP (AUC = 0.93).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Data conclusion: &lt;/strong&gt;Monitoring fiber-orientational alterations may provide new insights into early identification of PWMI-SCP.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Plain language summary: &lt;/strong&gt;This study utilized directional field analysis (DFA) to systematically examine white matter microstructural changes in three groups: periventricular white matter injury with spastic cerebral palsy (PWMI-SCP), periventricular white matter injury without spastic cerebral palsy (PWMI-nonSCP), and healthy controls. The results revealed significantly abnormal increases in the white matter geometric structure within the sensorimotor circuit in the PWMI-SCP g","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143364644","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":"Diagnosis of Sacroiliitis Through Semi-Supervised Segmentation and Radiomics Feature Analysis of MRI Images.","authors":"Lei Liu, Ruotao Zhong, Yuzhen Zhang, Haoyang Wan, Shuju Chen, Nanfeng Zhang, JingJing Liu, Wei Mei, Ruibin Huang","doi":"10.1002/jmri.29731","DOIUrl":"https://doi.org/10.1002/jmri.29731","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Sacroiliitis is a hallmark of ankylosing spondylitis (AS), and early detection plays an important role in managing the condition effectively. MRI is commonly used for diagnosing sacroiliitis, traditional methods often depend on subjective interpretation or limited automation which can introduce variability in diagnoses. The integration of semi-supervised segmentation and radiomics features may reduce reliance on expert interpretation and the need for large annotated datasets, potentially enhancing diagnostic workflows.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;To develop a diagnostic model for sacroiliitis and bone marrow edema (BME) using semi-supervised segmentation and radiomics analysis of MRI images.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study type: &lt;/strong&gt;Retrospective cohort study.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Population: &lt;/strong&gt;A total of 257 patients (161 males, 93 females; age 11-74 years), including 155 sacroiliitis and 175 BME patients. A total of 514 sacroiliac joint (SIJ) MRI images are analyzed, with 359 used for training and 155 for testing.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Field strength/sequence: &lt;/strong&gt;3.0 T, spin echo T1-weighted imaging (T1WI) and short-tau inversion recovery (STIR).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Assessment: &lt;/strong&gt;SIJ segmentation is automated using the semi-supervised segmentation-based Unimatch framework. Manual delineation of SIJ regions of interest (ROIs) on T1WI images by an experienced radiologist (W.M., 10-year experience) served as the reference standard for segmentation performance evaluation. Radiomics features from T1WI and STIR are used to train machine learning models, including support vector machine (SVM), logistic regression (LR), and light gradient boosting machine (LightGBM), for sacroiliitis and BME detection. Performance is assessed using area under the curve (AUC), sensitivity, specificity, and accuracy. The Dice coefficient is used to assess the performance of the semi-supervised segmentation model on SIJ segmentation.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Statistical tests: &lt;/strong&gt;Performance is evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Result: &lt;/strong&gt;The Unimatch model achieves an average Dice coefficient of 0.859 for SIJ segmentation. AUCs for sacroiliitis detection are 0.84 (LR), 0.86 (SVM), and 0.78 (LightGBM), while for BME detection, AUCs are 0.73 (LR), 0.76 (SVM), and 0.70 (LightGBM).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Data conclusion: &lt;/strong&gt;This study demonstrates that semi-supervised segmentation combined with radiomics features and machine learning models provides a promising approach for diagnosis of sacroiliitis and BME.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Plain language summary: &lt;/strong&gt;This study aimed to improve the diagnosis of sacroiliitis and bone marrow edema in patients with ankylosing spondylitis. The researchers used a method that automatically segments MRI images and analyzes features from those images. By applying machine learning, they created models to help detect sacroiliitis and bone marrow","PeriodicalId":16140,"journal":{"name":"Journal of Magnetic Resonance Imaging","volume":" ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143255863","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}