Journal of Computer Assisted Tomography最新文献

{"title":"Multiparametric MRI Features of Seromucinous and Mucinous Benign and Borderline Ovarian Tumors.","authors":"Safa Özyılmaz, Fatma Kulalı","doi":"10.1097/RCT.0000000000001930","DOIUrl":"https://doi.org/10.1097/RCT.0000000000001930","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate morphologic and quantitative MRI features associated with seromucinous and mucinous ovarian tumors of benign and borderline histology, with emphasis on their complementary role within structured preoperative imaging assessment.</p><p><strong>Methods: </strong>This retrospective single-center study included 76 women with 79 histopathologically confirmed mucinous or seromucinous ovarian tumors of benign or borderline histology who underwent preoperative pelvic MRI between 2011 and 2025. Tumors were classified as mucinous benign, mucinous borderline, seromucinous benign, or seromucinous borderline. MRI evaluation included morphologic features (tumor size, loculation number, septal thickness, and mural nodule presence) and quantitative analyses using signal intensity ratios on T1- and T2-weighted images (SIR-T1, SIR-T2). Enhancement and diffusion metrics were assessed for mural nodules when present. Multinomial logistic regression identified independent imaging predictors, and receiver operating characteristic (ROC) analysis evaluated diagnostic performance.</p><p><strong>Results: </strong>Tumor subtypes demonstrated distinct morphologic and signal-based patterns. Borderline tumors showed a higher prevalence of mural nodules than benign tumors. SIR-T1 differed significantly among subgroups and remained associated with tumor subtype in exploratory multivariable analysis. Overall intergroup differences were observed for tumor size, loculation number, SIR-T1, and mural nodule presence; however, these findings should be interpreted cautiously given the limited subgroup sizes. ROC analysis showed that SIR-T1 differentiated seromucinous borderline from mucinous benign tumors (AUC=0.801), while tumor size differentiated seromucinous benign from mucinous borderline tumors (AUC=0.796). Inter- and intraobserver agreement was good to excellent.</p><p><strong>Conclusions: </strong>Multiparametric MRI assessment integrating morphologic features with selected signal-based metrics may provide complementary information for the preoperative characterization of mucinous and seromucinous ovarian tumors. These findings should be interpreted as exploratory and adjunctive to structured MRI assessment frameworks such as O-RADS MRI rather than as standalone diagnostic criteria.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148990124","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multiparametric Whole-Body MRI Protocols in Contemporary Clinical Practice: Indications, Technical Considerations, and Emerging Applications.","authors":"Jiya Ghei, Ria Chhabra, Angela He, Alexandra K Callan, Amanda Dann, Aimaz Afrough, Avneesh Chhabra","doi":"10.1097/RCT.0000000000001928","DOIUrl":"https://doi.org/10.1097/RCT.0000000000001928","url":null,"abstract":"<p><p>Multiparametric whole-body MRI (WB-MRI) is an advanced cross-sectional imaging technique that integrates anatomic sequences into functional modalities, including diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) mapping and Dixon-based imaging. WB-MRI can evaluate the body from the vertex to the feet in a single session, offering advantages such as generating important biomarkers, robust musculoskeletal tissue assessment, high signal-to-noise ratio, and speed optimization, among others. This review examines current evidence and provides recommendations for multiparametric WB-MRI across a broad spectrum of oncologic and nononcologic indications, including multiple myeloma, lymphoma, melanoma, metastatic prostate cancer, cancer predisposition syndromes, MSK tumors, inflammatory arthritis, and diffuse neuromuscular disorders. Readers will gain an understanding of key technical considerations, including field strength selection, coil configuration, sequence design, and protocol optimization, that radiologists must navigate when implementing WB-MRI in clinical practice. Emerging applications such as AI-assisted lesion segmentation, automated quantitative biomarker analysis, and accelerated image reconstruction are also discussed, which spotlight the evolving role of WB-MRI in disease characterization and the following treatment response. As protocols become more standardized and evidence continues to grow, this review equips radiologists with a practical framework for incorporating multiparametric WB-MRI into contemporary clinical and research settings.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148990101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Physical Evaluation of Focal Spot Size and Phantom Thickness in Coronary CT Angiography Using Super-Resolution Deep Learning Reconstruction.","authors":"Sho Ogata, Yuji Tachibana, Risa Kanatani, Ryosuke Iwamoto, Tomohiro Nakayama","doi":"10.1097/RCT.0000000000001925","DOIUrl":"https://doi.org/10.1097/RCT.0000000000001925","url":null,"abstract":"<p><strong>Objective: </strong>To compare images with small and large focal spot sizes under clinical tube current limits across different subject thicknesses in the heart region using super-resolution deep learning reconstruction (SRDLR) to elucidate image quality behavior.</p><p><strong>Methods: </strong>We conducted quantitative assessments to compare small and large focal spot size images. The quantitative assessment involved imaging phantoms with 5 different diameters under coronary computed tomography angiography (CCTA) conditions, followed by SRDLR reconstruction. We calculated the noise power spectrum (NPS) to evaluate the frequency response of image noise in uniform phantom material. Spatial resolution was analyzed by measuring the task-based transfer function (TTF) for sections simulating iodine within the phantom.</p><p><strong>Results: </strong>In the NPS, for phantom diameters between 16 and 31 cm, small focal spot size images exhibited higher noise than large focal spot size images. However, for a 36 cm-diameter phantom, the small focal spot size image had less noise in the high-frequency region. TTF values were better for small focal spot size images, but the difference decreased with increasing phantom diameter, becoming negligible at 31 cm.</p><p><strong>Conclusions: </strong>The study clarified how focal spot size affects image quality in CCTA under maximum dose constraints, suggesting that optimal image quality can be achieved by appropriately selecting the focal spot size according to subject thickness.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148976924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Quantitative Analysis of Enhancing and Peritumoral Cerebral Blood Volume for Differentiating Glioblastoma From Central Nervous System Lymphoma.","authors":"Kazuhiro Murayama, Shohei Harada, Shigeo Ohba, Hiroyuki Maki, Yunosuke Kumazawa, Satomu Hanamatsu, Keigo Tamokami, Keitaro Nishikimi, Saki Ishikawa, Hirotaka Ikeda, Kentaro Tamura, Masanori Inoue","doi":"10.1097/RCT.0000000000001871","DOIUrl":"10.1097/RCT.0000000000001871","url":null,"abstract":"<p><strong>Objective: </strong>To quantitatively compare cerebral blood volume (CBV) in contrast-enhancing tumor areas and nonenhancing peritumoral fluid-attenuated inversion recovery (FLAIR) hyperintense regions between glioblastoma and central nervous system lymphoma (CNSL), and to assess the incremental diagnostic value of peritumoral CBV beyond enhancing tumor CBV.</p><p><strong>Methods: </strong>The study included 34 patients with histopathologically confirmed glioblastoma (n=22) or CNSL (n=12) who underwent pretreatment magnetic resonance imaging, including FLAIR, dynamic susceptibility contrast perfusion imaging, and contrast-enhanced T1-weighted imaging. Automated regions of interest (ROIs) were defined for enhancement areas (EAs) and nonenhancing peritumoral FLAIR abnormalities (PFAs). Quantitative indices included the median CBV within EAs (CBV EA ) and the 95th percentile CBV within PFAs (CBV PFA ). Intergroup differences were assessed, and diagnostic performance was evaluated using univariate and multivariate logistic regression models incorporating CBV EA and CBV PFA , receiver operating characteristic (ROC) analysis, and likelihood ratio testing (LRT).</p><p><strong>Results: </strong>Both CBV EA and CBV PFA were significantly higher in the glioblastoma group than in the CNSL group [CBV EA : 4.90 (4.14-5.58) vs. 2.84 (2.04-3.21) mL/100 g, P <0.0001, CBV PFA : 5.67 (3.75-8.08) vs. 4.10 (3.00-5.23) mL/100 g; P =0.029]. CBV EA showed the highest discriminatory performance in univariate analysis, whereas CBV PFA demonstrated a more modest association with tumor type. Although the AUC of the combined model was not significantly different from that of the CBV EA model alone [CBV EA : 0.913 (0.816-1.000) vs. combined model: 0.932 (0.848-1.000), P =0.602], the combined model showed a significant improvement in model fit according to LRT (χ 2 =4.1, P =0.043).</p><p><strong>Conclusions: </strong>Automated ROI-based analysis demonstrated significant differences in peritumoral CBV between glioblastoma and CNSL, and adding peritumoral CBV to the logistic regression model significantly improved overall model fit for differentiating between these entities beyond enhancing tumor CBV alone.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147772844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-driven Abdominal Aortic Calcification Extracted From Contrast-enhanced CT Is Predictive of All-cause Mortality and Cardiovascular Events in a Large Adult Population.","authors":"Louis A Hinshaw, Matthew H Lee, Ryan Zea, John W Garrett, Ronald M Summers, Perry J Pickhardt","doi":"10.1097/RCT.0000000000001876","DOIUrl":"10.1097/RCT.0000000000001876","url":null,"abstract":"<p><strong>Objective: </strong>To assess the utility of AI-driven quantification of abdominal aortic calcification (AAC) extracted from contrast-enhanced CT (CECT) scans using a fully automated AI tool for predicting all-cause mortality and cardiovascular events (CVEs).</p><p><strong>Materials and methods: </strong>In this retrospective cohort study, a fully automated deep learning tool for quantifying AAC from the aortic hiatus to iliac bifurcation was applied to abdominal CT scans in a large, adult population undergoing CT between January 2001 and February 2021. The aortic tool was applied to noncontrast CT (NCCT) and CECT scans. Validated linear adjustments were applied to CECT scans to derive NCCT-equivalent (NCE) measures. Death and cardiovascular events following CT were documented from the electronic health record. ROC curve and time-to-event analyses were performed to generate 5 and 10-year AUCs, HRs comparing the highest and lowest risk quartiles, and Kaplan-Meier survival curves.</p><p><strong>Results: </strong>A total of 123,500 adults (mean age, 51; M:F, 58,442: 65,058) underwent abdominal CT over the study interval (43,455 NCCT; 79,885 CECT). 21,937 patients died, and 16,599 patients had a documented CVE over the study interval (median follow-up: 59 mo). Similar 10-year AUCs were observed for mortality [0.744 (NCCT), 0.732 (CECT), 0.732 (NCE)] and CVE[0.716 (NCCT), 0.718 (CECT), and 0.718 (NCE)] prediction. Comparing the highest and lowest risk quartiles, higher AAC was associated with increased mortality risk: HR of 2.08 (CI: 1.94, 2.24) for NCCT, HR of 1.8 (CI: 1.70, 1.91) for CECT ( P <0.001 for all). Higher AAC was also associated with increased CVE risk: HR of 1.73 (CI: 1.61,1.86) for NCCT, HR of 1.45 (CI: 1.35,1.55) for CECT ( P <0.001 for all). No differences were observed after applying adjustments for IV contrast.</p><p><strong>Conclusions: </strong>AAC extracted from clinical CECT scans is predictive of mortality and CVD risk, with similar performance to NCCT. AI-driven quantification of AAC from CECT is feasible and can expand the scope and impact of population-level \"opportunistic screening.\"</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147868734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of Brain Computed Tomography Attenuation Values Between Deep-Learning and Conventional Reconstruction Methods: A Bias-Free Approach.","authors":"Ryo Yamakuni, Hirofumi Sekino, Masaki Saito, Takeyasu Kakamu, Katsuhiro Takahashi, Yasuhiro Abe, Anna Yamaki, Daichi Kuroiwa, Junko Hara, Hiroki Suenaga, Shiro Ishii, Kenji Fukushima, Hiroshi Ito","doi":"10.1097/RCT.0000000000001882","DOIUrl":"10.1097/RCT.0000000000001882","url":null,"abstract":"<p><strong>Objectives: </strong>To investigate the differences in brain computed tomography (CT) attenuation with deep-learning reconstruction (DLR) versus conventional reconstruction methods using a bias-free approach.</p><p><strong>Methods: </strong>Unenhanced brain CT scans of 27 participants (11 males and 16 females; mean age: 46.4 y) without brain abnormalities were included. The scans were reconstructed using 4 methods: hybrid iterative reconstruction (Hybrid-IR), DLR_mild, DLR_standard, and DLR_strong. All CT images were normalized to Montreal Neurological Institute coordinates, and CT attenuation values and signal-to-noise ratios (SNRs) were measured using 22 white and 10 gray-matter regions of interest. The edge rise distance (ERD) was measured in the right and left lateral ventricles.</p><p><strong>Results: </strong>The mean attenuation between Hybrid-IR and DLR ROIs showed strong correlations in both white ( R =0.953 to 0.964) and gray ( R =0.914 to 0.943) matter. The Bland-Altman analysis showed greater attenuation with DLR than with Hybrid-IR, with mean biases of 2.27 to 2.40 in white matter and 1.78 to 1.84 in gray matter. The median SNRs were highest for DLR_strong, followed by DLR_standard, DLR_mild, and Hybrid-IR. The median ERD of DLR_mild [2.86 (pixels)] was the lowest, followed by DLR_standard (3.27), Hybrid-IR (3.27), and DLR_strong (3.55). The Freedman test revealed significant differences based on the reconstruction method ( P =0.0080).</p><p><strong>Conclusions: </strong>The attenuation values for Hybrid-IR and DLR were strongly correlated, with DLR showing higher attenuation.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148143591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Low-Field Musculoskeletal MRI: Recent Advances, Challenges, and Practice Considerations.","authors":"Paul R Wojack, Iman Khodarahmi","doi":"10.1097/RCT.0000000000001879","DOIUrl":"10.1097/RCT.0000000000001879","url":null,"abstract":"<p><p>Magnetic resonance imaging (MRI) is well established as a cornerstone of musculoskeletal imaging. Although most clinical musculoskeletal MRI examinations are performed at 1.5 T and 3 T, recent advancements in low-field MRI have enabled imaging at 0.55 T with diagnostic utility approaching that of higher field strengths. Advances in transmit-receive chain design enhance signal efficiency, and deep learning-based image reconstruction algorithms substantially improve apparent signal-to-noise ratios. Low-field MRI also offers potential advantages, including reduced system and installation costs, decreased susceptibility to metal artifacts, and improved patient comfort. However, limitations remain, most notably the difficulty in achieving robust fat suppression and the inherently lower signal-to-noise ratio. This review discusses the technical principles, image quality considerations, economic factors, and clinical challenges associated with low-field MRI, with a particular focus on applications in the musculoskeletal system.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147868666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Commentary on Mid and Low-Field MR Imaging Systems: What Does the Future Hold?","authors":"Hersh Chandarana, Luke Ginocchio, Daniel K Sodickson","doi":"10.1097/RCT.0000000000001858","DOIUrl":"10.1097/RCT.0000000000001858","url":null,"abstract":"","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147377325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Preliminary Study of a Machine Learning Prediction of Poorly Differentiated Hepatocellular Carcinoma Based on a Comprehensive Parameter Analysis Using Dual-Energy Computed Tomography.","authors":"Atsushi Takamatsu, Norihide Yoneda, Fumihito Toshima, Taichi Kitagawa, Takahiro Komori, Dai Inoue, Azusa Kitao, Kazuto Kozaka, Osamu Matsui, Satoshi Kobayashi","doi":"10.1097/RCT.0000000000001867","DOIUrl":"10.1097/RCT.0000000000001867","url":null,"abstract":"<p><strong>Objective: </strong>To develop and evaluate the performance of a predictive machine learning model for poorly differentiated hepatocellular carcinoma (p-HCC) using comprehensive quantitative parameters from dual-energy computed tomography (DECT).</p><p><strong>Materials and methods: </strong>We retrospectively analyzed 181 surgically resected, pathologically proven HCCs in 170 patients who had undergone preoperative DECT between April 2019 and November 2025. After propensity-score matching on age, sex, alpha-fetoprotein (AFP), tumor size on CT, and LI-RADS categories, the dataset was divided into a training set including 51 HCCs (17 p-HCCs and 34 non-p-HCCs) from 2019 to 2022 and a testing set including 33 HCCs (11 p-HCCs, 22 non-p-HCCs) from 2023 to 2025. Overall, 3516 DECT parameters were extracted from precontrast images and three contrast-enhanced images [arterial phase (AP), portal venous phase (PVP), and delayed phase (DP)] for each case. These parameters included virtual monochromatic imaging (VMI) CT values, effective atomic numbers (Effective-Z), material density, spectral curve slopes (Slope), and interphase differences (Diff). Clinical data, including etiology of liver disease, serum AFP level, and lesion size, were collected. A machine learning model based on an extra trees classifier was trained, and a Shapley additive explanations (SHAP) analysis was performed for each selected parameter following data preprocessing. Model performance was evaluated using accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the receiver operating characteristic curve (AUC).</p><p><strong>Results: </strong>SHAP analysis revealed that the most highly influential features were delayed phase-related, with the average 40-keV VMI CT value in the delayed phase contributing most strongly. In addition, several Effective-Z-related features and selected material density parameters also showed substantial importance. In the testing set, the model demonstrated an accuracy, sensitivity, specificity, PPV, NPV, and AUC of 0.606, 0.818, 0.500, 0.450, 0.846, and 0.800, respectively.</p><p><strong>Conclusions: </strong>The predictive machine learning model for p-HCC showed acceptable diagnostic performance and, although still preliminary, may contribute to a noninvasive assessment of HCC differentiation grades, guiding clinical decision-making.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147690297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiomics Nomogram Based on Multiparametric MRI for Predicting the Hormone Receptor Status of HER2-Low Expression Breast Cancer.","authors":"Weishu Hou, Qun Wang, Hongli Pan, Shuhai Zhang, Yongqiang Yu","doi":"10.1097/RCT.0000000000001857","DOIUrl":"10.1097/RCT.0000000000001857","url":null,"abstract":"<p><strong>Objective: </strong>To develop and validate a multiparametric magnetic resonance imaging (mpMRI)-based radiomics nomogram to predict hormone receptor (HR) status in HER2-low breast cancer.</p><p><strong>Methods: </strong>A total of 198 HER2-low expression breast cancer patients who underwent mpMRI in The First Affiliated Hospital of Anhui Medical University Hospital from January 2019 to January 2025 were retrospectively analyzed. 69.2% (n=137) of patients were HR-positive, and 30.8% (n=61) were HR-negative. Patients were divided into a training set (n=138) and a testing set (n=60) in a 7:3 ratio. Radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced (DCE) images separately, and the radiomics score (radscore) was calculated. The clinical-radiological model (CM), single radiomics model (RM), and mpMRI RM were constructed, and a nomogram integrating radscore with clinical-radiological characteristics was developed. The predictive performance of the models was evaluated by receiver operating characteristic (ROC) curve analysis.</p><p><strong>Results: </strong>The area under the curve (AUC) of mpMRI RM in the training set and the testing set was 0.940 and 0.897, respectively, which were superior to that of single-modality models. The nomogram incorporating radscore and clinical-radiological characteristics, including ADC value, T2SI radio, and enhancement pattern, demonstrated higher AUC in both the training set (AUC=0.957) and testing set (AUC=0.891) than other RMs in predicting HR status of HER2-low expression breast cancer.</p><p><strong>Conclusion: </strong>An mpMRI-based nomogram incorporating radscore and clinical-radiological characteristics showed good predictive efficacy for assessing the HR status of HER2-low expression breast cancer.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147673822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}