Cancer Imaging最新文献

{"title":"CT-based renal artery caliber change as a potential imaging marker of treatment response in renal cell carcinoma; an exploratory imaging study.","authors":"Himawari Asanuma, Teppei Okamoto, Tadashi Yoshizawa, Takuya Oishi, Tomoko Hamaya, Hirotake Kodama, Takuma Narita, Jotaro Mikami, Naoki Fujita, Tohru Yoneyama, Hayato Yamamoto, Atsushi Imai, Chikara Ohyama, Shingo Hatakeyama","doi":"10.1186/s40644-026-01023-x","DOIUrl":"10.1186/s40644-026-01023-x","url":null,"abstract":"<p><strong>Background: </strong>To evaluate whether treatment-induced changes in tumor-bearing renal artery caliber on contrast-enhanced CT are associated with pathological response and pathological downstaging after systemic therapy in locally advanced or metastatic renal cell carcinoma (RCC), and to explore the clinical relevance of this vascular parameter.</p><p><strong>Methods: </strong>We retrospectively analyzed 43 patients with locally advanced or metastatic RCC who underwent systemic therapy followed by nephrectomy. Renal artery inner diameter was measured on arterial-phase CT before and after treatment using curved planar reformation. Interobserver reproducibility of the measurement was excellent (intraclass correlation coefficient [ICC] = 0.992). A subset of 33 patients had adequate bilateral imaging for tumor-bearing versus contralateral comparison. Tumor size was measured as the longest axial diameter on contrast-enhanced CT, and tumor shrinkage was expressed as percentage change from baseline according to RECIST 1.1 measurement principles. Choi criteria incorporated tumor size and attenuation changes. Pathological response was defined as the proportion of treatment-induced necrosis and fibrosis. Associations between arterial caliber change, pathological response, and pathological downstaging were assessed. Multivariable regression analysis identified independent predictors of pathological response. Discriminative ability for downstaging was evaluated using receiver operating characteristic analysis.</p><p><strong>Results: </strong>At baseline, the tumor-bearing renal artery had a larger inner diameter than the contralateral artery. After systemic therapy, its diameter significantly decreased (6.35 to 5.65 mm, p < 0.001). In the paired subset (n = 33), significant reduction occurred only in tumor-bearing arteries. Percentage diameter reduction correlated with pathological response (r = 0.314, p = 0.040). Patients with pathological downstaging showed greater diameter reduction than those without downstaging (16.3% vs 6.83%, p = 0.021). Tumor size change based on RECIST principles and Choi criteria were not significantly associated with downstaging. Multivariable analysis confirmed renal artery diameter change rate as an independent predictor of pathological response. ROC analysis showed moderate discrimination for downstaging (AUC 0.71, 95% CI: 0.55-0.87).</p><p><strong>Conclusion: </strong>Reduction in tumor-bearing renal artery caliber was independently associated with pathological response and downstaging. This vascular parameter may complement conventional tumor-centered imaging criteria in selected patients.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13147648/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147527061","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":"[¹⁸F]AlF-NOTA-QHY-04 uptake in spleen before anti-tumor therapy may predict degree of myelosuppression for patients with small cell lung cancer.","authors":"Siyu Cai, Jingru Liu, Kai Cheng, Xinzhi Zhao, Shengnan Xu, Jinli Pei, Shijie Wang, Kunlong Zhao, Jinming Yu, Wanhu Li, Jie Liu","doi":"10.1186/s40644-026-01024-w","DOIUrl":"10.1186/s40644-026-01024-w","url":null,"abstract":"<p><p>BACKGROUND: The aim of this study was to assess the value of C-X-C motif chemokine receptor 4 (CXCR4) targeted positron emission tomography/computed tomography ([18F]AlF-NOTA-QHY-04 PET/CT) imaging in spleen for prediction of degree of myelosuppression in patients with small cell lung cancer (SCLC) before therapy. METHODS: 59 patients with SCLC underwent [18F]AlF-NOTA-QHY-04 PET/CT scanning before treatment. The maximum, mean, and peak standard uptake value (SUVmax, SUVmean, SUVpeak), metabolic tumor volume (MTV) and total lesion CXCR4 expression (TLC) of spleen at baseline were acquired. Chemotherapy-based treatment was administered for limited-stage and extensive-stage SCLC. The most severe degree of myelosuppression during the treatment was recorded. Patients were identified as mild and severe myelosuppression. Potential predictors were identified and validated through intergroup difference analysiis, ROC curve assessment, and ordinal regression analysis. RESULTS: Patients with severe myelosuppression exhibited significantly higher splenic uptake values (SUVmax, SUVmean, SUVpeak and TLC) compared to those with mild myelosuppression (all p < 0.05). Regions of interest (ROI) curve analysis identified that higher metabolic parameters, including SUVmax (AUC = 0.853, p < 0.0001), SUVmean (AUC = 0.885, p < 0.0001), SUVpeak (AUC = 0.821, p < 0.0001), TLC (AUC = 0.753, p = 0.0016), were significantly associated with more severe myelosuppression. Multivariate analysis identified SUVmax (p = 0.007) and age (p = 0.005) as significant independent predictors of treatment-related myelosuppression severity. CONCLUSIONS: Our findings suggest that baseline splenic [18F]AlF-NOTA-QHY-04 PET/CT parameters, particularly SUVmax, may help identify patients with small cell lung cancer who are at high risk of severe treatment-related myelosuppression, with older age serving as an additional clinical risk factor. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13147742/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147527016","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":"RADS classification systems for bone tumors: current status and where do we go from here?","authors":"Vu Pham Thao Vy, Shivani Ahlawat, Hillary W Garner, Jim S Wu","doi":"10.1186/s40644-026-01021-z","DOIUrl":"10.1186/s40644-026-01021-z","url":null,"abstract":"<p><p>The assessment and management of bone tumors can be challenging as there are numerous neoplastic and non-neoplastic lesions. It can be difficult to understand which lesions can be ignored and which lesions require treatment. To standardize the reporting and management of bone tumors, four bone-lesion related reporting and data systems (RADS) have been published since 2021. The Bone Tumor Imaging Reporting and Data System (BTI-RADS), Oncologic Tumor Reporting and Data System (OT-RADS), Society of Skeletal Radiology Bone Reporting and Data System (SSR Bone-RADS), and American College of Radiology Bone Reporting and Data System (ACR Bone-RADS) are based in part on Breast Imaging Reporting & Data System (BI-RADS). These bone-lesion related RADS aim to improve communication, simplify management, and improve diagnostic consistency by linking imaging features to specific risks and evidence-based action recommendations. Although these systems show high sensitivity for determination of malignancy and moderate consistency for intra-observer readings, their coexistence leads to fragmentation and a lack of a unifying system. This review will discuss the main features of these 4 RADS systems and discuss future directions.</p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13141412/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147519869","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 dual-layer spectral CT for differentiating metastatic and non-metastatic lymph nodes in colorectal cancer.","authors":"Liyan Tang, Guoqing Liu, Xia Lv, Weicui Chen, Xian Liu","doi":"10.1186/s40644-026-01019-7","DOIUrl":"10.1186/s40644-026-01019-7","url":null,"abstract":"<p><p>BACKGROUND: To investigate the value of quantitative parameters obtained from dual layer spectrum CT (DLSCT) in distinguishing metastatic and non-metastatic lymph nodes in colorectal cancer (CRC). METHODS: This retrospective study enrolled 211 LNs from 66 CRC patients, including 78 metastatic LNs and 133 non-metastatic LNs. The morphological characteristics and quantitative DLSCT parameters of each identified lymph node were evaluated and compared. Univariate and multivariable logistic regression analyses were used to identify the independent factors for predicting LN metastasis. Receiver operating characteristic curve (ROC) and decision curve analysis (DCA) were employed to evaluate the diagnostic performance and clinical usefulness of the models. RESULTS: All morphological features differed between metastatic and non-metastatic LNs. Metastatic LNs showed significantly lower iodine concentration (IC) and normalized iodine concentration (NIC) in the arterial phase, while exhibiting higher effective atomic number (Zeff), normalized effective atomic number (nZeff) in the venous phase, and slope of the spectral Hounsfield unit curve (λHu) values in both phases. A positive correlation was found between λHu values of LNs and tumors in both the arterial and venous phases within the LN metastasis group. For differentiating metastatic and non-metastatic LNs, combined morphological features yielded an AUC of 0.695, with 66.7% sensitivity and 68.4% specificity, while a multivariate model incorporating DLSCT parameters improved this to 0.901, with 82.1% sensitivity and 87.2% specificity. The DeLong test and decision curve analysis (DCA) showed that quantitative multiparameter model offered better predictive performance and benefits for LN metastasis than the morphological features. CONCLUSION: Quantitative parameters from DLSCT provide significant value in distinguishing lymph node metastasis in colorectal cancer. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13127020/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147493694","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":"Radiomics-based differentiation between glioblastoma and primary central nervous system lymphoma: CT vs MRI.","authors":"Feifei Yu, Junhui Yuan, Xiaoye Lin, Feng Wang, Lan Yu, Shujie Yu, Youquan Zhu, Yang Song, Dairong Cao, Jieyun Chen, Zhen Xing","doi":"10.1186/s40644-026-01018-8","DOIUrl":"10.1186/s40644-026-01018-8","url":null,"abstract":"<p><p>BACKGROUND: To systematically evaluate and compare the diagnostic efficacy of radiomics models derived from noncontrast CT (NCCT) versus multiparametric MRI in differentiating glioblastoma (GBM) from primary central nervous system lymphoma (PCNSL). METHODS: In this retrospective, multicenter study, 543 patients with pathologically confirmed GBM (n = 401) or PCNSL (n = 142) were divided into 3 cohorts. 1084 quantitative features were extracted from contrast-enhancing (CE) and non-enhancing (NE) regions across NCCT and five MRI sequences (T2WI, T1WI, ADC, FLAIR, and CE-T1WI). Feature selection employed ANOVA, Kruskal-Wallis test, and recursive feature elimination, followed by nested cross-validation (5-fold outer, 3-fold inner) to construct four machine learning classifiers: support vector machine, linear discriminant analysis, logistic regression, and decision tree. Model performance was rigorously assessed through AUC, accuracy, sensitivity, specificity with bootstrap-derived 95% confidence intervals. The Shapley Additive Explanation (SHAP) analysis was employed to explore the interpretability of models. RESULTS: The CE-T1WI radiomics model demonstrated superior diagnostic capability, with its AUCs of train/internal test/external test in CE regions and NE regions were 0.962/0.963/0.907 and 0.966/0.892/0.867, respectively. Notably, the CT-based model was not significantly different from other MRI models except for CE-T1WI model. The AUCs of train/internal test/external test for CT model in CE and NE regions were 0.941/0.906/0.822 and 0.902/0.891 /0.782, respectively. CONCLUSIONS: Both NCCT and multiparametric MRI are valuable in identifying GBM and PCNSL. The CE-T1WI radiomics model has the best diagnostic efficacy. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13104201/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147466830","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":"The result prediction of fluorescence in situ hybridization for breast cancer patients based on machine learning and deep learning models: a multicenter study.","authors":"Cong Jiang, Dong Chen, Xiao Yu, Shentao Zhang, Yuting Xiu, Ningbin Luo, Jingjing Wu, Xuefang Zhang, Man Chen, Dechun Yang, Ziyu Zhu, Yuanxi Huang, Shipeng Ning, Shicong Tang","doi":"10.1186/s40644-026-01015-x","DOIUrl":"10.1186/s40644-026-01015-x","url":null,"abstract":"<p><p>OBJECTIVE: To develop and validate a multicenter ultrasound-based predictive model for fluorescence in situ hybridization (FISH) results in HER2 (2+) breast cancer patients, aiming to provide a convenient and cost-effective tool to support clinical decision-making. MATERIALS AND METHODS: In this retrospective multicenter study, 5,888 breast cancer patients from six institutions were included. Radiomics features were extracted from ultrasound images using PyRadiomics, and deep learning features were obtained using a Vision Transformer (ViT). Feature selection was performed using least absolute shrinkage and selection operator (LASSO) regression. Multiple machine learning models were developed, and their performance was evaluated with the area under the curve (AUC). The DeLong test was used for model comparison. RESULTS: The proportion of FISH-positive cases ranged from 9.7% to 20.0% across the six centers. The fusion model combining ViT and radiomics signatures consistently outperformed the individual models in the training, test, and all external validation cohorts. The AUCs of the fusion model were 0.887 in the training cohort, 0.799 in the test cohort, and 0.763, 0.796, 0.734, and 0.632 in the four external validation cohorts, respectively (all P < 0.05). CONCLUSION: The proposed ultrasound-based fusion model enables accurate prediction of FISH assay results in HER2 (2+) breast cancer patients and may serve as a reliable decision-support tool to reduce unnecessary FISH testing in clinical practice. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13097979/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147455180","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":"Response assessment in rectal cancer undergoing total neoadjuvant therapy: decision tree analysis integrating contrast enhanced T1-weighted MRI.","authors":"An-Hsin Chen, Eric Yi-Liang Shen, Chih-Ying Deng, Jeng-Fu You, Tai-Di Chen, Chien-Ming Chen, Gigin Lin, Ying-Chieh Lai","doi":"10.1186/s40644-026-01016-w","DOIUrl":"10.1186/s40644-026-01016-w","url":null,"abstract":"<p><p>BACKGROUND: This study assesses the role of contrast enhanced T1-weighted imaging (CE-T1WI) in improving response assessment for rectal cancer patients undergoing total neoadjuvant therapy (TNT) and its integration with T2-weighted imaging (T2WI) and DWI using decision tree analysis. METHODS: A retrospective analysis was conducted on rectal cancer patients between January 2020 and December 2022 who underwent TNT followed by MRI-based response assessment, including T2WI, DWI, and CE-T1WI. Complete response (CR) was defined as pathologic CR or sustained clinical CR for ≥2 years. Diagnostic performance was assessed using receiver operating characteristic analysis. Logistic regression was performed to identify predictors of CR. Finally, a decision tree model was constructed to determine the optimal combination of sequences for predicting CR. RESULTS: A total of 124 rectal cancer patients (mean age 58 ± 12 years; 89 men) were analyzed, of whom 45 (36%) achieved CR. CE-T1WI demonstrated superior predictive performance (AUC = 82.3% [75.0−88.7]) compared to DWI (78.6% [71.1−85.9]) and T2WI (74.5% [66.5−82.4]). In the two sequences combinations, the combination of DWI and CE-T1WI achieved the highest AUC of 82.9% (75.7−89.8), with an accuracy of 85.5% (79.0−91.9), sensitivity of 73.4% (60.0−86.0) and specificity of 92.4% (85.9−97.6), outperforming the other combinations. Finally, the decision tree model integrating three sequences achieved an AUC of 90.1% (83.7−95.6), with an accuracy of 86.4% (80.6−91.9), a sensitivity of 77.6% (62.5−91.8) and a specificity of 91.2% (81.7−97.6). CONCLUSION: Integrating CE-T1WI with other sequences enhances CR prediction in rectal cancer patients undergoing TNT, supporting its inclusion in routine MRI for improved response assessment. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13094038/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147430779","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":"Precision medicine in myeloma demands precision imaging: is whole body MRI the solution?","authors":"Christina Messiou, Martin Kaiser","doi":"10.1186/s40644-026-01017-9","DOIUrl":"10.1186/s40644-026-01017-9","url":null,"abstract":"<p><p>Outcomes for patients with multiple myeloma have improved substantially over the past two decades, with median overall survival more than doubling in many cohorts. Management of smouldering myeloma has also shifted, with randomised trials demonstrating that early intervention in high-risk disease delays progression and improves survival compared with observation alone. Imaging diagnostics must therefore evolve in parallel to provide minimally invasive, sensitive and quantitative insights into disease for patients who will live with myeloma for many years. Here we review the latest evidence for WB-MRI in myeloma, outline current guideline recommendations, and consider future directions.</p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":"26 1","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12977864/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147430744","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":"Prediction of MYC/BCL-2 co-expression in diffuse large B-cell lymphoma using a multimodal fusion model: a retrospective study based on PET/CT habitat radiomics and deep learning.","authors":"Yu He, Shirong Chen, Xinyang Li, Jingkai Yi, Dan Wang, Kailin Qi, Xiao Jiang, Ping Wu, Meng Zhao, Hao Lu, Ying Kou, Yutang Yao, Zhuzhong Cheng","doi":"10.1186/s40644-026-01014-y","DOIUrl":"10.1186/s40644-026-01014-y","url":null,"abstract":"<p><p>BACKGROUND: The co-expression of MYC and BCL-2 proteins in diffuse large B-cell lymphoma (DLBCL) is linked to poor prognosis and resistance to standard therapies. Thus, a non-invasive and accurate method to detect this co-expression before treatment is essential for pre-treatment risk stratification and assisting in personalized patient management. METHODS: This retrospective study included DLBCL patients who underwent baseline 18F-FDG PET/CT between December 2018 and August 2024. Clinical data were collected. Habitat radiomics features were extracted by segmenting tumors into distinct subregions, and 3D deep learning features were obtained using convolutional neural networks, both derived from PET/CT images. Two individual models were built: A habitat radiomics model and a 3D deep learning model. A multimodal fusion model was also constructed by integrating dimensionally reduced features from habitat radiomics, 3D deep learning, clinical data, and PET-derived metabolic parameters. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). DeLong’s test was used to compare AUCs, and net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were calculated to assess net benefit. RESULTS: A total of 242 patients were enrolled (95 DEL-positive [39.3%] and 147 DEL-negative [60.7%]) and were stratified-randomly split by DEL status into a training set (n = 193) and test set (n = 49) in an 8:2 ratio. This was a retrospective single-center study with an internal hold-out test cohort. All feature selection and model development were performed in the training cohort only, and the test cohort was used solely for final evaluation. The habitat radiomics model showed better performance than the deep learning model, with AUCs of 0.869 (95% CI: 0.820–0.918) and 0.812 (95% CI: 0.661–0.964) vs. 0.844 (95% CI: 0.787–0.902) and 0.715 (95% CI: 0.562–0.869) in the training and test sets, respectively. The fusion model outperformed both, achieving AUCs of 0.946 (95% CI: 0.917–0.974) in the training and 0.890 (95% CI: 0.793–0.987) in the test set. Calibration curves demonstrated strong agreement between predicted and observed outcomes. DCA confirmed higher clinical benefit for the fusion model. DeLong’s test showed the fusion model significantly outperformed both individual models in the training set and the deep learning model in the test set (P < 0.05). NRI and IDI further supported improved discrimination, suggesting potential incremental value. CONCLUSIONS: The multimodal fusion model based on 18F-FDG PET/CT and clinical data provides a non-invasive and reliable tool for predicting MYC/BCL-2 co-expression in DLBCL, providing complementary prognostic information to assist personalized treatment planning. TRIAL REGISTRATION: This study was retrospectively registered. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13101366/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147430767","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":"Deep learning prediction model based on multi-modal transvaginal ultrasound scan images for endometrial cancer.","authors":"Xingzhe Liu, Yuanjia Wen, Jiahao Liu, Wenzhi Lv, Yuan Wu, Yue Gao, Shaoqing Zeng, Guannan Li, Yu Xia, Shennan Shi, Qiuyang Xu, Xiaofei Jiao, Wenjian Gong, Ding Ma, Guang-Nian Zhao, Yong Fang, Xiaodong Cheng, Xiao-Yan Xu, Dan Liu, Qinglei Gao","doi":"10.1186/s40644-026-01013-z","DOIUrl":"10.1186/s40644-026-01013-z","url":null,"abstract":"<p><p>BACKGROUND: Ultrasound-based deep learning (DL) models for the precise diagnosis of endometrial cancer are insufficient. Our aim is to develop and validate an automatic multi-modal ultrasound DL prediction model for the accurate identification of benign versus malignant endometrial diseases. METHODS: The retrospective dataset of patients with endometrial diseases from two hospitals was segregated into an internal set (n = 696) and an external set (n = 78). All patients underwent grayscale, color Doppler, and 3D reconstructed ultrasound scans. We established ResNet-18 DL models using individual sequences (DLgray, DLCDFI, DL3D) and multi-modal sequences (DLfusion), with comparative performance evaluation through area under the receiver operating characteristic curve (AUC) analysis. We compared the best-performing model with 12 radiologists and assessed the diagnostic performance of radiologists with model assistance. RESULTS: The DLfusion model exhibits precise discrimination capabilities for benign and malignant endometrial diseases, achieving an AUC of 0.92. This performance significantly surpasses that of DLgray (0.78, p < 0.01), DLCDFI (0.79, p = 0.01), and assessments by radiologists (0.64, p < 0.001). The DLfusion model is more accurate than radiologists (0.86 vs. 0.78, p < 0.001) at detecting endometrial cancer. With the assistance of the DLfusion model, the average diagnostic accuracy of twelve radiologists significantly improves from 0.77 to 0.85 (p < 0.001). CONCLUSIONS: The DLfusion model based on multi-modal ultrasonic images exhibits a better capacity in diagnosing endometrial cancer compared with radiologists and DL models based on single modal, and significantly improves the diagnostic accuracy of radiologists, offering valuable support for clinical decision-making. </p>","PeriodicalId":9548,"journal":{"name":"Cancer Imaging","volume":" ","pages":""},"PeriodicalIF":4.8,"publicationDate":"2026-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13049860/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147321433","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}