Academic Radiology最新文献

{"title":"Histogram Analysis of Advanced Diffusion-weighted MRI Models for Evaluating the Grade and Proliferative Activity of Meningiomas","authors":"Xiaodan Chen MD ,&nbsp;Yichao Zhang ,&nbsp;Hui Zheng MD ,&nbsp;Zhitao Wu MD ,&nbsp;Danjie Lin MD ,&nbsp;Ye Li ,&nbsp;Sihui Liu MD ,&nbsp;Yizhu Chen MD ,&nbsp;Rufei Zhang MD ,&nbsp;Yang Song ,&nbsp;Yunjing Xue MD ,&nbsp;Lin Lin MD, PhD","doi":"10.1016/j.acra.2024.10.047","DOIUrl":"10.1016/j.acra.2024.10.047","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To explore the value of diffusion tensor imaging (DTI), diffusion kurtosis imaging (DKI), neurite orientation dispersion and density imaging (NODDI), and mean apparent propagator (MAP) magnetic resonance imaging histogram analysis in evaluating the grade and proliferative activity of meningiomas.</div></div><div><h3>Materials and Methods</h3><div>A total of 134 meningioma patients were prospectively included and underwent magnetic resonance diffusion imaging. The whole-tumor histogram parameters were extracted from multiple functional maps. Mann-Whitney U test was used to compare the histogram parameters of high- and low-grade meningiomas. The receiver operating characteristic (ROC) curve and multiple logistic regression analysis were used to evaluate the diagnostic efficacy. The correlation between histogram parameters and the Ki-67 index was analyzed. The diffusion model was further validated with an independently validation set (n = 33).</div></div><div><h3>Results</h3><div>Among single histogram parameters, the variance of NODDI-ISOVF (isotropic volume fraction) showed the highest AUC of 0.829 in grading meningiomas. For the combined models, the DKI model had the best performance in the diagnosis (AUC=0.925). Delong test showed the DKI combined model showed superior diagnostic performance to those of DTI, NODDI and MAP models (<em>P</em> &lt; 0.05 for all). Moreover, moderate to weak correlations were found between various diffusion parameters and the Ki-67 labeling index (rho=0.20–0.45, <em>P</em> &lt; 0.05 for all). In the validation set, the DKI model still showed higher performance (AUC, 0.85) than other diffusion models, thus demonstrating robustness.</div></div><div><h3>Conclusions</h3><div>Whole-tumor histogram analyses of DTI, DKI, NODDI, and MAP are useful for evaluating the grade and cellular proliferation of meningiomas. DKI combined model has higher diagnostic accuracy than DTI, NODDI and MAP in meningioma grading.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2171-2181"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142689497","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":"HRS Improves Active Surveillance for Prostate Cancer by Timely Identification of Progression","authors":"Isabella M. Kimbel ,&nbsp;Veronica Wallaengen ,&nbsp;Evangelia I. Zacharaki ,&nbsp;Adrian L. Breto ,&nbsp;Ahmad Algohary ,&nbsp;Sophia Carbohn ,&nbsp;Sandra M. Gaston ,&nbsp;Nachiketh Soodana-Prakash ,&nbsp;Pedro F.S. Freitas ,&nbsp;Oleksandr N. Kryvenko ,&nbsp;Patricia Castillo ,&nbsp;Matthew C. Abramowitz ,&nbsp;Chad R. Ritch ,&nbsp;Bruno Nahar ,&nbsp;Mark L. Gonzalgo ,&nbsp;Dipen J. Parekh ,&nbsp;Alan Pollack ,&nbsp;Sanoj Punnen ,&nbsp;Radka Stoyanova PhD","doi":"10.1016/j.acra.2024.11.008","DOIUrl":"10.1016/j.acra.2024.11.008","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Active surveillance (AS) is the preferred management strategy for low-risk prostate cancer. This study aimed to evaluate the impact of Habitat Risk Score (HRS), an automated approach for mpMRI analysis, for early detection of progressors in a prospective AS clinical trial (MAST NCT02242773).</div></div><div><h3>Materials and Methods</h3><div>The MAST protocol includes Confirmatory mpMRI ultrasound fusion (MRI-US) biopsy and yearly surveillance MRI-US biopsies for up to 3 years. Clinical and mpMRI data from patients that progressed based on protocol criteria at years 1–3 were reviewed. Patients were classified as “MRI/HRS Progressors” if the PI-RADS lesion(s) had been targeted throughout the surveillance and resulted in positive biopsies, or as \"Missed Progressors\" if the lesion(s) were not identified by PI-RADS (“PI-RADS Miss”) or were missed by the biopsy (“Needle Miss”). HRS maps were generated for each patient and evaluated for association with histopathological progression.</div></div><div><h3>Results</h3><div>Of the 34 patients, 15 were classified as “MRI/HRS Progressors” and 19 as \"Missed Progressors\" (12 \"PI-RADS Miss\", seven \"Needle Miss\"). In all cases, HRS confirmed the PI-RADS assessment. In the \"PI-RADS Miss\" group, HRS identified the lesions in all patients that were not targeted by biopsy and resulted in patient reclassification. HRS volumes showed clear association with tumor evolution both in terms of volume and aggressiveness over time.</div></div><div><h3>Conclusion</h3><div>HRS volumes can serve as a quantitative biomarker for early detection of progression and lead to timely conversion to treatment, thereby improving patient outcomes and reducing the burden of unnecessary surveillance.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2081-2089"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142856167","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":"Evaluating Neoadjuvant Immunochemotherapeutic Response for Bladder Carcinoma Using Amide Proton Transfer-Weighted MRI","authors":"Lingmin Kong MD ,&nbsp;Bei Weng MD ,&nbsp;Qian Cai MD ,&nbsp;Ling Ma MD ,&nbsp;Wenxin Cao MD ,&nbsp;Yanling Chen MD ,&nbsp;Long Qian MD, PhD ,&nbsp;Yan Guo MD, PhD ,&nbsp;Junxing Chen MD, PhD ,&nbsp;Huanjun Wang MD, PhD","doi":"10.1016/j.acra.2024.11.060","DOIUrl":"10.1016/j.acra.2024.11.060","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To investigate the feasibility of amide proton transfer-weighted (APTw) and diffusion-weighted MRI in evaluating the response of bladder cancer (BCa) to neoadjuvant immunochemotherapy.</div></div><div><h3>Materials and Methods</h3><div>From June 2021 to July 2023, participants with pathologically confirmed BCa were prospectively recruited to undergo MRI examinations, including APTw and diffusion-weighted MRI before and after neoadjuvant immunochemotherapy. Histogram analysis features (mean, median, and entropy) were extracted from pre- and post-treatment APTw and apparent diffusion coefficient (ADC) maps, respectively. Participants were categorized into pCR (pathological complete response, no residual tumor) and non-pCR groups based on histologic evaluation of post-treatment cystectomy specimens. The diagnostic efficacy of parameters in predicting tumor responsiveness was evaluated by calculating the area under receiver operating characteristic curve (AUC).</div></div><div><h3>Results</h3><div>Significant differences were found in several imaging biomarkers derived from pre-treatment APTw and diffusion-weighted MRI (<em>P<!--> </em>&lt;<!--> <!-->0.05 for all). The baseline APTw mean values yielded the highest diagnostic performance, with an AUC of 0.85 (AUC: 0.75–0.93), for evaluating tumor responsiveness. For the pCR group, APTw values markedly decreased while ADC values noticeably increased at post-treatment MRI (<em>P<!--> </em>&lt;<!--> <!-->0.05 for all). However, the parameter changes in non-pCR group were not significant (<em>P<!--> </em>&gt;<!--> <!-->0.05 for all).</div></div><div><h3>Conclusion</h3><div>MRI parametrics derived from APTw and diffusion-weighted MRI can both serve as valuable noninvasive imaging biomarkers for evaluating the efficacy of immunochemotherapy and may be used to guide personalized precision therapy.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2090-2098"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142967131","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":"The Scourge of Workplace Bullying","authors":"Richard B. Gunderman MD, PhD","doi":"10.1016/j.acra.2025.02.023","DOIUrl":"10.1016/j.acra.2025.02.023","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2378-2379"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143517198","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":"Generative AI as Both Threat to and Tool for Maintaining Radiology Research Propriety","authors":"Jeffrey P. Guenette MD, MPH","doi":"10.1016/j.acra.2025.02.008","DOIUrl":"10.1016/j.acra.2025.02.008","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 1791-1792"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143411434","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 with Multiphase CTA and CTP Images for Predicting Hemorrhagic Transformation in Acute Ischemic Stroke Patients","authors":"Hongyu Qu ,&nbsp;Zhilin Zhang","doi":"10.1016/j.acra.2025.02.014","DOIUrl":"10.1016/j.acra.2025.02.014","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2150-2151"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143450888","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":"Prediction of Radiation Therapy Induced Cardiovascular Toxicity from Pretreatment CT Images in Patients with Thoracic Malignancy via an Optimal Biomarker Approach","authors":"Mujun Long ,&nbsp;Mostafa Alnoury ,&nbsp;Jayaram K. Udupa ,&nbsp;Yubing Tong ,&nbsp;Caiyun Wu ,&nbsp;Nicholas Poole ,&nbsp;Sutirth Mannikeri ,&nbsp;Bonnie Ky ,&nbsp;Steven J. Feigenberg ,&nbsp;Jennifer W. Zou ,&nbsp;Shannon O’Reilly ,&nbsp;Drew A. Torigian","doi":"10.1016/j.acra.2025.01.012","DOIUrl":"10.1016/j.acra.2025.01.012","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Cardiovascular toxicity is a well-known complication of thoracic radiation therapy (RT), leading to increased morbidity and mortality, but existing techniques to predict cardiovascular toxicity have limitations. Predictive biomarkers of cardiovascular toxicity may help to maximize patient outcomes.</div></div><div><h3>Methods</h3><div>The machine learning optimal biomarker (OBM) method was employed to predict development of cardiotoxicity (based on serial echocardiographic measurements of left ventricular ejection fraction and longitudinal strain) from computed tomography (CT) images in patients with thoracic malignancy undergoing RT. Manual segmentations of 10 cardiovascular objects of interest were performed on pre-treatment non-contrast-enhanced CT simulation images in 125 patients with thoracic malignancy (41 who developed cardiotoxicity and 84 who did not after RT). 1078 features describing morphology, image intensity, and texture for each of these objects were extracted and the top 5 features among them that were most uncorrelated and showed the best ability to discriminate between cardiotoxicity/ no cardiotoxicity were determined. The best combination among all possible combinations among these 5 features that yielded the highest accuracy of prediction on a training data set was selected, an SVM classifier was then trained on this combination, and tested for prediction accuracy on an independent data set. Prediction accuracy was quantified for the optimal features derived from each object.</div></div><div><h3>Results</h3><div>The best feature combination based on 5 CT-based features derived from the left ventricle had the highest testing prediction accuracy of 0.88 among all objects. Prediction accuracies over all objects ranged from 0.76–0.88. Heart, Left Atrium, Aortic Arch, Thoracic Aorta, and Descending Thoracic Aorta showed the next best accuracy of 0.84. Most optimal features were texture properties based on the co-occurrence matrix.</div></div><div><h3>Conclusion</h3><div>It is feasible to predict future cardiotoxicity following RT with high accuracy in individual patients with thoracic malignancy from available pre-treatment CT images via machine learning techniques.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 1895-1905"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143054125","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":"Accuracy of Contrast-enhanced Ultrasonography with Perfluorobutane for Diagnosing Subpleural Lung Lesions","authors":"Wuxi Chen ,&nbsp;Qing Tang ,&nbsp;Guosheng Liang ,&nbsp;Liantu He ,&nbsp;Shiyu Zhang ,&nbsp;Jiaxin Tang ,&nbsp;Haixing Liao ,&nbsp;Yuxin Zhang","doi":"10.1016/j.acra.2024.09.033","DOIUrl":"10.1016/j.acra.2024.09.033","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To investigate the diagnostic value of perfluorobutane-enhanced ultrasound (US) examinations for differentiating benign from malignant subpleural lung lesions.</div></div><div><h3>Methods</h3><div>This single-center, retrospective study enrolled consecutive patients with subpleural lung lesions between January 2022 and March 2023. The cause of the lung lesions was confirmed by biopsy and follow-up examinations. The lesions were continuously evaluated using perfluorobutane-enhanced US for 0–180 s, and washout (WT) was observed after 3, 5, and 10 min. Univariate and multivariate analyses were used to identify significant US features, which were evaluated for their diagnostic performance. The diagnostic performance of combining several features for predicting malignant lung lesions was also assessed by multivariate logistic regression analysis.</div></div><div><h3>Results</h3><div>Seventy cases were included (17 benign lesions [13 men, 4 women; mean age: 57.5 ± 12.2 years] and 53 malignant lesions [41 men, 12 women; mean age: 63.3 ± 11.6 years]). Peak intensity (PI), arrival time (AT), and WT after 10 min significantly differed between malignant and benign lesions. The sensitivity and accuracy were significantly higher for 10-minute WT than for AT (both p &lt; 0.05). The area under the curve of the combined diagnostic evaluation with AT, PI, and 10-minute WT was 0.897 (95% [CI]: 0.806–0.988), which was significantly higher than that of AT or PI alone.</div></div><div><h3>Conclusion</h3><div>Perfluorobutane-enhanced US can differentiate benign from malignant lung lesions, and combining AT, PI, and 10-minute WT for diagnostic purposes performed better than a single feature.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2272-2280"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142569849","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":"Quantitative Parameters of Intravoxel Incoherent Movement Imaging and Dynamic Contrast Enhancement MRI for the Prediction of HER2-Zero, -Low, and -Positive Breast Cancers","authors":"Siqi Zhao ,&nbsp;Shiyu Wang ,&nbsp;Yuanfei Li ,&nbsp;Yueqi Wu ,&nbsp;Moyun Zhang ,&nbsp;Ning Ning ,&nbsp;Hongbing Liang ,&nbsp;Deshuo Dong ,&nbsp;Jie Yang ,&nbsp;Xue Gao ,&nbsp;Haonan Guan ,&nbsp;Lina Zhang MD","doi":"10.1016/j.acra.2024.11.011","DOIUrl":"10.1016/j.acra.2024.11.011","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To explore the predictive value of quantitative parameters from intravoxel incoherent movement (IVIM) imging and dynamic contrast enhancement MRI (DCE-MRI) for HER2 expression in breast cancer.</div></div><div><h3>Materials and Methods</h3><div>This retrospective study included 167 women with breast cancer who underwent MRI from December 2019 to December 2023, categorized into 48 HER2-positive, 78 HER2-low and 41 HER2-zero cancers. All patients underwent IVIM imaging and DCE-MRI. Statistical analyses, including one-way ANOVA, Kruskal-Wallis test and χ2 test, were employed to compare clinical data, MRI features, and MRI quantitative parameters including standard ADC(ADC), pure diffusion coefficient(D), perfusion-related diffusion coefficient(D*), perfusion fraction(f), volume transfer constant(K^trans), extravascular extracellular interstitial volume ratio(V_e) and rate constant(K_ep) between the three groups. Multivariable logistic regression was used to identify independent predictors for distinguishing HER2 expressions. The diagnostic efficacy of significant IVIM and DCE parameters for different HER2 expressions was analyzed using receiver operator characteristic (ROC) curves.</div></div><div><h3>Results</h3><div>Peritumoral edema, histological grade and K_ep achieved an AUC of 0.86(95%CI:0.78,0.91) in distinguishing HER2-positive tumors from HER2-low expressing tumors and were independent predictors for differentiating these two groups. Among HER2-positive and -zero breast cancers, the combined model of D*, K^trans and K_ep had an AUC of 0.74(95%CI:0.63,0.82) for the prediction of HER2-positive versus HER2-zero cancers, and its prediction efficiency was not improved compared with that of a single parameter(<em>P</em> &gt; .05).</div></div><div><h3>Conclusion</h3><div>Quantitative parameters from intravoxel incoherent movement imaging and dynamic contrast enhancement MRI can predict different HER2 expressions in breast cancer from different perspectives, with implications for therapy.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 1851-1860"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142734345","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":"Radiomics Model Based on Contrast-enhanced CT Intratumoral and Peritumoral Features for Predicting Lymphovascular Invasion in Hypopharyngeal Squamous Cell Carcinoma","authors":"Chenyang Xu ,&nbsp;Yifan Ju ,&nbsp;Zhiwei Liu ,&nbsp;Changling Li ,&nbsp;Shengda Cao ,&nbsp;Tongliang Xia ,&nbsp;Dongmin Wei ,&nbsp;Wenming Li ,&nbsp;Ye Qian ,&nbsp;Dapeng Lei","doi":"10.1016/j.acra.2024.11.017","DOIUrl":"10.1016/j.acra.2024.11.017","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Patients with Hypopharyngeal Squamous Cell Carcinoma (HSCC) exhibiting lymphovascular invasion (LVI) frequently demonstrate a poor prognosis. We aim to determine whether contrast-enhanced computed tomography (CECT)-derived intratumoral and peritumoral radiomic features could predict the LVI status of HSCC patients.</div></div><div><h3>Materials and Methods</h3><div>166 patients with pathologically confirmed HSCC were included in this study, 47 of whom were LVI positive. Preoperative CECT data were randomly divided into a training dataset and a validation dataset in an 8:2 ratio. A total of 1648 radiomics features were extracted from the total tumor volume (GTV) and the surrounding 1- to 5-mm-wide tumor margins (labeled as Peri1V-5V). A deep learning model based on the GTV was also constructed. Radiomics nomograms were established by integrating deep learning model features and clinical features. Receiver operating characteristic curve (ROC), calibration curve, and decision curve analysis (DCA) were utilized to evaluate and compare the predictive performance of all models.</div></div><div><h3>Results</h3><div>Peri1V-Radscore showed the best prediction efficiency in the validation dataset among all peritumoral models. Among the clinical variables, the upper tumor boundaries and clinical N stage were independent predictors. Compared with the clinical predictor model, Peri1V-Radscore, deep learn model and Nomogram model can improve prediction efficiency in LVI status. Their respective AUC values were 0.94, 0.84, and 0.96. The results of DCA showed that a good net benefit could be obtained from the Peri1V-Radscore model.</div></div><div><h3>Conclusion</h3><div>Intratumoral combined peritumoral radiomics model based on CECT can superior predict LVI status in HSCC patients and may have significant potential for future applications in clinical practice.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 4","pages":"Pages 2099-2110"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142792767","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}