Radiologia Medica最新文献

{"title":"Multimodal prediction of major adverse cardiovascular events in hypertensive patients with coronary artery disease: integrating pericoronary fat radiomics, CT-FFR, and clinicoradiological features.","authors":"Qing Zou, Taichun Qiu, Chunxiao Liang, Fang Wang, Yongji Zheng, Jie Li, Xingchen Li, Yudan Li, Zhongyan Lu, Bing Ming","doi":"10.1007/s11547-025-01991-3","DOIUrl":"https://doi.org/10.1007/s11547-025-01991-3","url":null,"abstract":"<p><strong>Purpose: </strong>People with both hypertension and coronary artery disease (CAD) are at a significantly increased risk of major adverse cardiovascular events (MACEs). This study aimed to develop and validate a combination model that integrates radiomics features of pericoronary adipose tissue (PCAT), CT-derived fractional flow reserve (CT-FFR), and clinicoradiological features, which improves MACE prediction within two years.</p><p><strong>Materials and methods: </strong>Coronary-computed tomography angiography data were gathered from 237 patients diagnosed with hypertension and CAD. These patients were randomly categorized into training and testing cohorts at a 7:3 ratio (165:72). The least absolute shrinkage and selection operator logistic regression and linear discriminant analysis method were used to select optimal radiomics characteristics. The predictive performance of the combination model was assessed through receiver operating characteristic curve analysis and validated via calibration, decision, and clinical impact curves.</p><p><strong>Results: </strong>The results reveal that the combination model (Radiomics.</p><p><strong>Clinical: </strong>Imaging) improves the discriminatory ability for predicting MACE. Its predictive efficacy is comparable to that of the Radiomics.Imaging model in both the training (0.886 vs. 0.872) and testing cohorts (0.786 vs. 0.815), but the combination model exhibits significantly improved specificity, accuracy, and precision. Decision and clinical impact curves further confirm the use of the combination prediction model in clinical practice.</p><p><strong>Conclusions: </strong>The combination prediction model, which incorporates clinicoradiological features, CT-FFR, and radiomics features of PCAT, is a potential biomarker for predicting MACE in people with hypertension and CAD.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143674317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Perspectives on MRI sequences and clustering techniques in predicting osteosarcoma treatment response.","authors":"Yu Mori, Naoko Mori","doi":"10.1007/s11547-025-01999-9","DOIUrl":"https://doi.org/10.1007/s11547-025-01999-9","url":null,"abstract":"","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143674319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning reconstruction for accelerated 3-D magnetic resonance cholangiopancreatography.","authors":"Jan M Brendel, Reza Dehdab, Judith Herrmann, Stephan Ursprung, Sebastian Werner, Haidara Almansour, Elisabeth Weiland, Dominik Nickel, Konstantin Nikolaou, Saif Afat, Sebastian Gassenmaier","doi":"10.1007/s11547-025-01987-z","DOIUrl":"https://doi.org/10.1007/s11547-025-01987-z","url":null,"abstract":"<p><strong>Purpose: </strong>This study aimed to compare a conventional three-dimensional (3-D) magnetic resonance cholangiopancreatography (MRCP) sequence with a deep learning (DL)-accelerated MRCP sequence (hereafter, MRCP_DL) regarding acquisition time and image quality.</p><p><strong>Materials and methods: </strong>We conducted a prospective study of consecutive patients referred for MRCP between November 2023 and April 2024 at a single tertiary center. Each participant underwent 1.5T 3-D T2-weighted turbo spin echo MRCP using both a conventional sequence (threefold acceleration) and MRCP_DL (eightfold acceleration). Three blinded readers independently evaluated image quality, including background signal suppression, bile and pancreatic duct visibility, artifact level, and diagnostic confidence on an ordinal four-point scale. Acquisition times were compared using a paired t-test. Image quality parameters were assessed with repeated measures ANOVA. Interreader agreement was analyzed using Fleiss' κ.</p><p><strong>Results: </strong>Out of 419 consecutive patients, 30 participants were evaluated (mean age, 63 ± 15 years; 16 men, 14 women). The mean acquisition time was 10:30 ± 03:04 min for conventional MRCP and 3:57 ± 01:13 min for MRCP_DL, P < 0.001. MRCP_DL reduced acquisition time by 62.4%. Artifact levels were rated at 3.17 ± 0.77 for conventional MRCP and 3.56 ± 0.66 for MRCP_DL (P = 0.041). Background signal suppression, bile duct visibility, pancreatic duct visibility, and diagnostic confidence did not differ significantly (P > 0.05). Interreader agreement was substantial to almost perfect (κ: 0.64-87).</p><p><strong>Conclusions: </strong>Deep learning-accelerated 3-D MRCP reduced acquisition time by 62%, minimized artifacts, and preserved bile and pancreatic duct visibility, supporting its adoption in routine clinical practice.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143658511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 segmentation of ultra-low-dose CT images using an optimized nnU-Net model.","authors":"Yazdan Salimi, Zahra Mansouri, Chang Sun, Amirhossein Sanaat, Mohammadhossein Yazdanpanah, Hossein Shooli, René Nkoulou, Sana Boudabbous, Habib Zaidi","doi":"10.1007/s11547-025-01989-x","DOIUrl":"https://doi.org/10.1007/s11547-025-01989-x","url":null,"abstract":"<p><strong>Purpose: </strong>Low-dose CT protocols are widely used for emergency imaging, follow-ups, and attenuation correction in hybrid PET/CT and SPECT/CT imaging. However, low-dose CT images often suffer from reduced quality depending on acquisition and patient attenuation parameters. Deep learning (DL)-based organ segmentation models are typically trained on high-quality images, with limited dedicated models for noisy CT images. This study aimed to develop a DL pipeline for organ segmentation on ultra-low-dose CT images.</p><p><strong>Materials and methods: </strong>274 CT raw datasets were reconstructed using Siemens ReconCT software with ADMIRE iterative algorithm, generating full-dose (FD-CT) and simulated low-dose (LD-CT) images at 1%, 2%, 5%, and 10% of the original tube current. Existing FD-nnU-Net models segmented 22 organs on FD-CT images, serving as reference masks for training new LD-nnU-Net models using LD-CT images. Three models were trained for bony tissue (6 organs), soft-tissue (15 organs), and body contour segmentation. The segmented masks from LD-CT were compared to FD-CT as standard of reference. External datasets with actual LD-CT images were also segmented and compared.</p><p><strong>Results: </strong>FD-nnU-Net performance declined with reduced radiation dose, especially below 10% (5 mAs). LD-nnU-Net achieved average Dice scores of 0.937 ± 0.049 (bony tissues), 0.905 ± 0.117 (soft-tissues), and 0.984 ± 0.023 (body contour). LD models outperformed FD models on external datasets.</p><p><strong>Conclusion: </strong>Conventional FD-nnU-Net models performed poorly on LD-CT images. Dedicated LD-nnU-Net models demonstrated superior performance across cross-validation and external evaluations, enabling accurate segmentation of ultra-low-dose CT images. The trained models are available on our GitHub page.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143658514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transvenous pulmonary chemoembolization and microwave ablation for lung metastases from breast cancer: a propensity score matching analysis.","authors":"Hamzah Adwan, Lars Hammann, John Bielfeldt, Sven Becker, Thomas J Vogl","doi":"10.1007/s11547-025-01966-4","DOIUrl":"https://doi.org/10.1007/s11547-025-01966-4","url":null,"abstract":"<p><strong>Purpose: </strong>To compare the outcomes of patients with pulmonary metastases from breast cancer, who were treated with transvenous pulmonary chemoembolization (TPCE) and consecutive microwave ablation (MWA) with patients treated by TPCE alone.</p><p><strong>Material and methods: </strong>This retrospective single-center study included patients with unresectable and/or non-responsive to systemic chemotherapy pulmonary metastases originating from breast cancer, treated by TPCE followed by MWA, in case of adequate response to TPCE, or by TPCE alone. The groups of patients were balanced using propensity score matching (PSM).</p><p><strong>Results: </strong>A total of 97 patients met the inclusion criteria for this study. After PSM, 23 patients were included in the combination therapy group (Group 1) and 42 patients were included in the monotherapy group (Group 2). The median overall survival (OS) time was 33.6 months for Group 1 with a 2-year OS rate of 62%, and 20.2 months for Group 2 with a 2-year OS rate of 43%. There was no significant difference between the two groups regarding OS (p value: 0.429). The rate of progressive/recurrent disease was 17.4% (4/23) in Group 1 and 23.8% (10/42) in Group 2 (p value: 0.754). The number of metastases was the only significant factor for OS in all patients after PSM (p value: 0.032, HR: 1.016, 95% CI 1.001-1.031).</p><p><strong>Conclusion: </strong>TPCE is an effective potential treatment for lung metastases of breast cancer, which can be performed alone or combined with MWA. Patients who responded to TPCE and received subsequent MWA demonstrated non-significant better OS and local tumor control.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143658515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive cohort study: computer tomography-guided high-dose rate brachytherapy as metastasis-directed therapy for liver metastases from colorectal cancer in repeat oligoprogression.","authors":"Mateusz Bilski, Katarzyna Korab, Magdalena Orzechowska, Julia Ponikowska, Paweł Cisek, Barbara Alicja Jereczek-Fossa, Jacek Fijuth, Łukasz Kuncman","doi":"10.1007/s11547-025-01988-y","DOIUrl":"https://doi.org/10.1007/s11547-025-01988-y","url":null,"abstract":"<p><strong>Purpose: </strong>The standard treatment for oligometastatic colorectal cancer includes systemic therapy, with surgery and metastasis-directed therapy as options. The optimal strategy, especially for repeat oligoprogression (rOP), remains unclear. We report outcomes of liver computer tomography-guided high-dose rate brachytherapy (CT-BRT) in this setting.</p><p><strong>Methods: </strong>This retrospective cohort study included colorectal cancer patients with liver-only oligoprogression during systemic therapy, meeting criteria of up to 5 liver metastases, CT-BRT eligibility, and ECOG status ≤ 2. Patients were followed for local response, progression-free survival (PFS), overall survival (OS), and toxicity. Response, according to RECIST 1.1, was initiated 6 months post-CT-BRT.</p><p><strong>Results: </strong>A total of 262 metastases were treated in 127 patients, with 67.7% receiving third-line or later systemic therapies. One to four liver metastases were found in 29.1%, 42.5%, 21.2%, and 7.1% of patients, respectively, with a median volume of 128 cm³. A median of 3 applicators was used, with CT-BRT doses of 15 Gy, 20 Gy, and 25 Gy given to 29.9%, 41.7%, and 28.3% of patients. At 6 months complete response occurred in 3.1%, progressive disease in 23.6%, partial response in 19.7%, and stable disease in 53.5%. Median PFS was 9 months, median OS was 16 months, with 1-year and 2-year OS rates of 65% and 16%, respectively. Liver-only metastases and objective response were associated with longer PFS. The G3 toxicity was 4.0%; no events > G3 were reported.</p><p><strong>Conclusions: </strong>This largest study documents favorable outcomes of liver CT-BRT for rOP, establishing this method as a viable option in this indication.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143625712","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stereotactic body radiotherapy combined with immunotherapy or targeted therapy: a screenshot from Italy on behalf of the Italian Association of Clinical Oncology and Radiotherapy (AIRO).","authors":"Giulia Corrao, Giulia Marvaso, Mattia Zaffaroni, Maria Giulia Vincini, Serena Badellino, Paolo Borghetti, Francesco Cuccia, Manuela Federico, Giampaolo Montesi, Antonio Pontoriero, Ciro Franzese, Mauro Loi, Barbara Alicja Jereczek-Fossa, Marta Scorsetti","doi":"10.1007/s11547-025-01977-1","DOIUrl":"https://doi.org/10.1007/s11547-025-01977-1","url":null,"abstract":"<p><strong>Aim: </strong>Despite the widespread use of immunotherapy (IO) and targeted therapy (TT) in clinical practice, data on toxicity in combination with SBRT are lacking, largely based on retrospective studies and case reports. The present survey, conducted within the AIRO Oligometastatic Study Group, was developed for radiation oncologists to investigate the current clinical practice in Italy regarding hypofractionated SBRT (defined as a dose/fraction ≥ 5 Gy) in cancer patients using IO and TT.</p><p><strong>Methods: </strong>The online survey, composed of 19 questions, was developed using the cloud-based platform SurveyMonkey® and was sent to all registered AIRO members using the association's mailing list and was administered online and in anonymous form.</p><p><strong>Results: </strong>Sixty-eight AIRO members from different Italian regions completed the proposed survey. 59% stated that there is sufficient knowledge within their institute regarding the potential interaction between SBRT and TT/IO. According to 76% of the pool, a multidisciplinary protocol is available in their institute. However, 50% of participants report that interdisciplinary consultation with a radiation oncologist did not always take place. Only 6% of the radiation oncologists stated that they lacked sufficient knowledge to adequately consider the treatment of these patients, while 81% of them felt confident in deciding whether or not to combine SBRT and IO/TT. The 26% and the 37% answered that TT and IO, respectively, should be stopped during RT administration. Regarding which drug types would be of concern when patients are referred for SBRT, the majority of the answers regarded ALK inhibitors (21%), BRAF inhibitors (25%), EGFR inhibitors (26%), immune checkpoint inhibitors (25%) and PARP inhibitors (25%). Only 5% did not consider any of these classes of drugs. Particularly, many radiation oncologists (19%) did not expect a real risk of tumor flare upon discontinuation of TT or IO.</p><p><strong>Conclusion: </strong>The findings from the present survey underscore significant variability in clinical practice regarding the combination of SBRT with IO and TT across Italy. Therefore, there is an urgent need for prospective clinical studies to evaluate the safety and efficacy of combining SBRT with IO/TT. These studies should aim to generate robust data that can inform the development of comprehensive, evidence-based guidelines.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143617143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hepatic arterial infusion chemotherapy combined with lenvatinib and immune checkpoint inhibitor versus lenvatinib for advanced hepatocellular carcinoma: a multicenter study with propensity score and coarsened exact matching.","authors":"Qunfang Zhou, Hui Li, Ye Liang, Ruixia Li, Xiaohui Wang, Wei Wang, Mingyu Liu, Feng Duan, Zhimei Huang","doi":"10.1007/s11547-025-01975-3","DOIUrl":"https://doi.org/10.1007/s11547-025-01975-3","url":null,"abstract":"<p><strong>Purpose: </strong>Hepatic arterial infusion chemotherapy (HAIC) combined with lenvatinib (Len) and immune checkpoint inhibitor (ICI) in treating advanced hepatocellular carcinoma (HCC) still needs further confirmation. We aimed to evaluate the efficacy of HAIC combined with Len and ICI (HAIC + Len + ICI) versus Len alone in advanced HCC.</p><p><strong>Methods: </strong>A total of 290 patients in Len group and 349 patients in HAIC + Len + ICI group were analysed. Propensity score matching (PSM), inverse probability treatment weighting (IPTW), and coarsened exact matching (CEM) analyses were used to balance the bias between two groups. Mediation analysis of treatment type in survival was performed for analysis.</p><p><strong>Results: </strong>The median progression-free survival (PFS) was 5.9 ± 0.2 months in Len group and 9.2 ± 0.5 months in HAIC + Len + ICI group. The HAIC + Len + ICI group demonstrated significantly better PFS than the Len group across the entire cohort (hazard ratio [HR], 0.50; 95% CI 0.43-0.60; P < 0.001). This advantage in PFS was sustained in the PSM, IPTW, and CEM cohorts. HAIC + Len + ICI group also showed better overall survival (OS) than the Len group (HR, 0.38; 95% CI 0.31-0.46; P < 0.001). The OS was also superior in the PSM, IPTW, and CEM cohorts. The objective response rate (ORR) in HAIC + Len + ICI group was twice as high as that in Len group. Further mediation analysis showed tumor response at 3 and 6 months had different mediation effect on survival.</p><p><strong>Conclusions: </strong>HAIC combined with Len and ICI showed improved better OS and PFS than Len alone. This triple therapy could be considered as a first-line treatment for advanced HCC.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143617061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accuracy of abbreviated magnetic resonance compared to 3-dimensional mammography and ultrasound in early detection of breast cancer.","authors":"Sara El Garwany, Azza Abdel-Hamid Gad, Sahar Mahmoud Mansour, Mohammad Ahmed Al-Shatouri, Taghreed Alshafeiy, Ahmed Fathy AlSerafi","doi":"10.1007/s11547-025-01983-3","DOIUrl":"https://doi.org/10.1007/s11547-025-01983-3","url":null,"abstract":"<p><strong>Background: </strong>The purpose of this study is to assess the usefulness of the novel abbreviated MR (AB-MR) protocol in the screening of women with an intermediate risk of breast cancer. Sixty women with a Tyrer-Cuzick model-determined intermediate risk of breast cancer underwent AB-MR, mammography, and tomosynthesis examinations; as an auxiliary procedure, ultrasound imaging was carried out. Every modality was allocated a final BI-RADS category. Time spent on acquisition and interpretation was also noted. Pathological confirmation was obtained in all cases exhibiting malignant findings. The difference in sensitivity and specificity between the two modalities was evaluated using McNemar's test.</p><p><strong>Results: </strong>When compared to traditional screening methods, AB-MR demonstrated 100% NPV, 98% specificity, 66.7% PPV, and 100% sensitivity in women with intermediate risk of breast cancer. Comparing mammography/ultrasound to positive malignancies verified by biopsy, the results indicated 100% sensitivity, 96.5% specificity, 60% PPV, and 100% NPV. Complete agreement was observed between abbreviated MR and malignant biopsies (100% sensitivity, specificity, NPV, and PPV). For AB-MR and mammography, the average reading time was 4 min and 5 min, respectively. The average acquisition time for AB-MRI was around 10 min, whereas the average time for complete MR imaging is 17 min.</p><p><strong>Conclusion: </strong>AB-MR has better sensitivity, specificity, PPV, and NPV in screening of intermediate- and high-risk breast cancer. Acquisition time was shorter than full MR protocol. Reading time was decreased in respect of mammography. MRI screening ought to be more practical with the AB-MR protocol.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143617054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiomics for differential diagnosis of ischemic and dilated cardiomyopathy using non-contrast-enhanced cine cardiac magnetic resonance imaging.","authors":"Jaravee Lasode, Warath Chantaksinopas, Sararas Khongwirotphan, Pairoj Chattranukulchai, Yongkasem Vorasettakarnkij, Sira Sriswasdi, Monravee Tumkosit, Yothin Rakvongthai","doi":"10.1007/s11547-025-01979-z","DOIUrl":"https://doi.org/10.1007/s11547-025-01979-z","url":null,"abstract":"<p><strong>Objective: </strong>Our study investigated the effectiveness of using radiomics and non-gadolinium cine cardiac magnetic resonance (CMR) imaging for differential diagnosis between ischemic and dilated cardiomyopathy (ICM vs. DCM) and detecting myocardial scar without relying on gadolinium-based contrast agents.</p><p><strong>Materials and methods: </strong>We retrospectively enrolled 200 patients with age <math><mo>≥</mo></math> 18 who were diagnosed with ICM and DCM, both with and without myocardial scars, and had complete short-axis cine CMR images without artifacts over the heart area. Radiomic features were extracted from the myocardium and blood pools in short-axis cine CMR images. Feature selection was performed using intraclass correlation coefficient and recursive feature elimination with cross-validation (RFECV). Multivariate analysis with logistic regression was utilized to construct 3 radiomic models (ICM vs DCM, Scar vs NoScar in ICM and in DCM). Model performance was evaluated using area under the receiver operating characteristic (ROC) curve (AUC) and compared with radiologist's performance using cine CMR images.</p><p><strong>Results: </strong>For each of the three models, 21, 14, and 19 radiomic features were selected. The AUC values of each model were 0.964 ± 0.008, 0.989 ± 0.004, and 0.996 ± 0.004, respectively, in the training set and 0.918 ± 0.040, 0.955 ± 0.045, and 0.935 ± 0.052, respectively, in the test set (p < 0.0001). The radiomic models outperformed the radiologist as shown by the ROC curve.</p><p><strong>Conclusions: </strong>Radiomics showed promise in differentiating between ICM and DCM and detecting myocardial scar with cine CMR images, which offered an alternative for ICM and DCM differentiation in patients with gadolinium contraindication.</p>","PeriodicalId":20817,"journal":{"name":"Radiologia Medica","volume":" ","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143597868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}