Abdominal radiology (New York)最新文献

{"title":"DCE-MRI histogram analysis for predicting survival in patients with epithelial ovarian carcinoma.","authors":"Juanru Xu, Ruihong Chen, Limei Guo, Jiale Wang, Yifan Wang, Jinliang Niu, Xiao-Li Song","doi":"10.1007/s00261-026-05814-5","DOIUrl":"https://doi.org/10.1007/s00261-026-05814-5","url":null,"abstract":"<p><strong>Purpose: </strong>To evaluate the potential of dynamic contrast-enhanced MRI (DCE-MRI) histogram analysis for predicting overall survival (OS) in epithelial ovarian cancer (EOC) patients.</p><p><strong>Methods: </strong>This retrospective study included 155 patients with EOC who underwent pretreatment DCE-MRI. Histogram features extracted from DCE-MRI parametric maps were selected using least absolute shrinkage and selection operator (LASSO) Cox regression, with selection stability assessed across 200 bootstrap replicates. Clinical, imaging, and combined models were developed using Cox regression and random survival forest (RSF). Apparent concordance indices (C-indices) were calculated and corrected for optimism using 100 bootstrap resamples with fixed predictor sets. Exploratory Kaplan-Meier analyses assessed risk-group separation, and SHapley Additive exPlanations (SHAP) analysis was used for model interpretation.</p><p><strong>Results: </strong>Median OS was 54 months. The 10th percentile of K^trans was selected in 88.0% of bootstrap replicates. Imaging Cox and RSF models achieved apparent C-indices of 0.790 and 0.827, respectively. Combined models incorporating this feature, International Federation of Gynecology and Obstetrics (FIGO) stage, and residual disease status achieved apparent C-indices of 0.806 for Cox regression and 0.853 for RSF, respectively. Bootstrap-corrected C-indices were higher for the combined than the corresponding clinical models, with a statistically significant improvement for Cox regression (0.805 vs. 0.737, P = 0.020) but not for RSF (0.810 vs. 0.745, P = 0.120). The combined models did not differ significantly (P = 0.820). Exploratory Kaplan-Meier analysis using cohort-specific median risk-score cutoffs showed significant OS separation (both P < 0.001). The 10th percentile of K^trans yielded the highest mean absolute SHAP value in the combined Cox model and exhibited the widest distribution of SHAP values in the combined RSF model.</p><p><strong>Conclusion: </strong>DCE-MRI histogram features, particularly the 10th percentile of K^trans, may aid OS prediction in EOC and provide complementary prognostic information when integrated with clinicopathological factors. These exploratory findings require prospective multicenter external validation.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Percutaneous sclerotherapy for simple hepatic cysts: a descriptive systematic review and proportional meta-analysis of single-arm studies.","authors":"Krish Vennam, Ashwani Kumar Sharma","doi":"10.1007/s00261-026-05817-2","DOIUrl":"https://doi.org/10.1007/s00261-026-05817-2","url":null,"abstract":"<p><p>Percutaneous aspiration and sclerotherapy is a widely used treatment for symptomatic simple hepatic cysts, but the optimal sclerosing agent is unsettled and the literature consists almost entirely of single-arm studies, which cannot support valid between-agent comparison. We performed a descriptive systematic review and proportional meta-analysis restricted strictly to simple hepatic cysts (excluding polycystic liver disease), pooling outcomes within each agent rather than comparing agents. PubMed, Embase, and Cochrane CENTRAL were searched (1,237 records), yielding 25 included studies. Proportional outcomes were pooled with random-effects models on the logit scale (DerSimonian-Laird), with restricted maximum likelihood/Hartung-Knapp and continuity-correction-free (logistic-normal) sensitivity analyses. Prediction intervals and I² are reported, and subgroups of fewer than three studies are described rather than pooled. Symptomatic relief was high and consistent for the two adequately powered agents (ethanol 88.2% [95% CI 82.3-92.3; I²=36%; k = 9]; minocycline 90.3% [77.6-96.2; I²=0%; k = 5]) and similarly high in the two polidocanol/lauromacrogol series (88-100%). Complete radiographic obliteration was not summarizable (ethanol 53.7%, I²=95%, prediction interval 1-99%; polidocanol/lauromacrogol and minocycline each rested on two studies spanning roughly 23-94%). Minor complications differed numerically (polidocanol/lauromacrogol 7.9%, ethanol 19.8%, minocycline 24.1%) but with wide, overlapping intervals. Major procedure-related events were rare but did occur, including one ethanol intoxication with respiratory failure, one hepatic abscess, and one suspected chemical peritonitis. Recurrence, reported over widely varying follow-up, was low but is time-dependent and is presented descriptively rather than as a durability ranking. Symptom relief, not obliteration, is the reproducible and decision-relevant outcome. Agent differences in tolerability and durability are numerical and unconfirmed, so agent selection can reasonably be individualized by operator experience and patient priorities pending randomized trials.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of large language models for assigning clear cell likelihood score v2.0 from free-text MRI reports: a feasibility study.","authors":"Yanting Duan, Yilei Zhao, Maowei He, Tiefeng Li, Liuqiong Yang, Zhengyu Hu, Minhong Wang, Zhan Feng","doi":"10.1007/s00261-026-05818-1","DOIUrl":"https://doi.org/10.1007/s00261-026-05818-1","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>To evaluate the performance of four large language models (LLMs) for automated feature extraction and clear cell likelihood score version 2.0 (ccLS v2.0) assignment from free-text reports using a multi-step prompting strategy.</p><p><strong>Materials and methods: </strong>This retrospective multicenter study included 1,048 magnetic resonance imaging (MRI) reports of renal masses from three institutions (2020-2026). Four LLMs (DeepSeek-V3.2, Qwen 3.5, GPT-5.3, and Gemini 3.0) were evaluated for automated ccLS v2.0 assignment through a three-stage prompting workflow: (1) feature extraction, (2) rule-based criteria matching, and (3) final categorization. Performance was evaluated against expert radiologist consensus-based ccLS v2.0 assignments. Accordingly, these results primarily reflect agreement in task execution rather than diagnostic accuracy for ccRCC.</p><p><strong>Results: </strong>Gemini 3.0 achieved the highest overall accuracy, correctly assigning ccLS categories in 683 of 802 reports in the internal cohort (85.2%; 95% CI, 82.5%-87.5%) and 211 of 246 reports in the external validation cohort (85.8%; 95% CI, 80.9%-89.6%). All LLMs achieved accuracy greater than 80% for assigning high-risk categories (ccLS 4-5). Gemini 3.0 showed comparatively strong performance in extracting individual imaging features, such as T2-weighted imaging (T2WI) hyperintensity. However, all models showed reduced performance for features requiring multi-step interpretation, particularly the arterial-to-delayed enhancement ratio (ADR), with Qwen 3.5 showing the weakest performance for these complex features. In the pathology subgroup of the internal cohort (n = 237), expert report-based scores yielded an AUC of 0.799 for identifying ccRCC, with LLM-assigned scores showing AUCs of 0.768-0.791 and small AUC differences from the expert standard (-0.031 to -0.008) with overlapping 95% CIs.</p><p><strong>Conclusion: </strong>LLMs employing multi-step prompting, particularly Gemini 3.0, demonstrated favorable performance in assigning ccLS v2.0 categories from free-text reports. Despite limitations in complex feature reasoning, these models showed potential for supporting automated ccLS assessment in radiology workflows; however, in their current implementation, they should be considered decision-support tools that require expert verification rather than autonomous, unsupervised scoring.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preoperative FDG PET/CT topographic patterns and clinical parameters are associated with incomplete cytoreduction in peritoneal carcinomatosis: proposal of the Resectability Score.","authors":"Mutlay Keskin, Fulya Kasırga Çelik","doi":"10.1007/s00261-026-05811-8","DOIUrl":"https://doi.org/10.1007/s00261-026-05811-8","url":null,"abstract":"<p><strong>Purpose: </strong>To explore associations between preoperative FDG PET/CT-derived topographic patterns, clinical variables, and unfavorable surgical outcome (incomplete cytoreduction or intraoperative unresectability) in peritoneal carcinomatosis (PC), and to propose a simple, transparent, preoperatively applicable Resectability Score (RS) as a hypothesis-generating tool for future prospective validation.</p><p><strong>Methods: </strong>In this single-center retrospective cohort study (May 2018-April 2026), 39 patients with histopathologically confirmed PC who underwent surgical exploration were included. Univariable associations between PET/CT variables (PET-estimated Peritoneal Cancer Index [PCI] category, mesenteric root uptake, retroperitoneal uptake) and clinical factors and the composite endpoint were assessed. Given the limited number of outcome events, no multivariable model was fitted. The RS was proposed conceptually, with weights reflecting the strength of the observed associations, and its performance was deliberately not assessed in this dataset.</p><p><strong>Results: </strong>Eighteen of 39 patients (46.2%) had an unfavorable outcome (CC-2/3, n = 12; intraoperative unresectability, n = 6). PET-estimated PCI > 20 (OR 9.4, 95% CI 2.0-44.3; p = 0.003), mesenteric root FDG uptake (0/21 vs. 6/18; p = 0.006), and retroperitoneal FDG uptake (0/21 vs. 6/18; p = 0.006) were significantly associated with unfavorable outcome; gastric primary tumor showed a non-significant trend (OR 4.8, 95% CI 1.0-22.3; p = 0.07). Among 33 patients who underwent CRS + HIPEC, complete cytoreduction rates by PET-estimated PCI category were 85.7% (PCI ≤ 10), 63.6% (PCI 11-20), and 25.0% (PCI > 20) (p = 0.017). All six unresectable patients had shown mesenteric root or retroperitoneal FDG uptake preoperatively.</p><p><strong>Conclusion: </strong>Preoperative FDG PET/CT topographic patterns in particular high disease burden and mesenteric root or retroperitoneal uptake are associated with unfavorable surgical outcomes in PC. The proposed RS is a hypothesis-generating tool only; formal derivation, internal validation, and external validation are required before any clinical application.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"When TIME is kidney: an imaging-based diagnostic algorithm for iatrogenic ureteral injuries.","authors":"Emanuele Barabino, Pietro Corazza, Giacomo Costa, Edoardo Nizzi, Marta Verda, Nicola Camisassi, Francesco Petrocelli, Giulio Bovio, Jeries Paolo Zawaideh, Francesca Lacelli, Giuseppe Cittadini, Veronica Giasotto","doi":"10.1007/s00261-026-05819-0","DOIUrl":"https://doi.org/10.1007/s00261-026-05819-0","url":null,"abstract":"<p><p>Iatrogenic ureteral injuries are uncommon but potentially severe complications of abdominal, pelvic, and endourological procedures. Because many injuries are not recognized intraoperatively and may initially produce nonspecific clinical or laboratory findings, imaging frequently provides the first indication of ureteral damage. Computed tomography urography is the principal cross-sectional technique in clinically stable adults, while retrograde and antegrade pyelography provide direct luminal assessment when urinary excretion is absent, CT findings remain indeterminate, or treatment is planned. Imaging should establish ureteral continuity, identify urinary extravasation, assess drainage, localize the lesion, and detect associated collections and infection. We propose the TIME framework-Transection, Isolated obstruction, Mixed injury, and Extravasation-only-as a practical imaging-based description of the principal patterns of iatrogenic ureteral injury. The framework is applied hierarchically. Demonstrated complete loss of native ureteral continuity defines TIME-T. TIME-I denotes obstruction without demonstrated leakage; TIME-M denotes concomitant leakage with upstream collecting-system dilatation and at least partially preserved continuity; and TIME-E denotes leakage without upstream collecting-system dilatation and with preserved or partially preserved continuity. This pictorial review describes the multimodality imaging findings of iatrogenic ureteral injury, introduces the TIME framework, highlights common diagnostic pitfalls, and presents a time-sensitive algorithm intended to support structured radiological assessment and multidisciplinary communication.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Photon-counting detector CT for follow-up after TACE in hepatocellular carcinoma: effects of reconstruction kernel and sharpness.","authors":"Meng Yuan, Dandan Wang, Zhongxing Shi, Zhongqi Sun, Yingzhu Zhao, Tiancai Yan, Yu Yuan, Zheng Xu, Huijie Jiang","doi":"10.1007/s00261-026-05810-9","DOIUrl":"https://doi.org/10.1007/s00261-026-05810-9","url":null,"abstract":"<p><strong>Purpose: </strong>To compare the performance of different reconstruction kernels and sharpness levels in photon-counting detector CT (PCD-CT) for post-transarterial chemoembolization (TACE) imaging of hepatocellular carcinoma (HCC), with particular emphasis on residual viable tumor conspicuity, and to identify reconstruction settings that may improve clinical follow-up assessment.</p><p><strong>Methods: </strong>In this retrospective study, 72 patients with HCC undergoing TACE were included. Triple-phase contrast-enhanced CT images were reconstructed using body regular (Br) and quantitative regular (Qr) kernels at four different sharpness levels (36, 40, 44, and 48). Quantitative analysis included image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and edge sharpness. Subjective image quality was assessed by two readers using a 5-point Likert scale.</p><p><strong>Results: </strong>Increasing kernel sharpness significantly increased image noise and decreased SNR and CNR across all contrast-enhanced phases (e.g., arterial phase, Br kernel, noise 16.8 ± 2.1 to 36.1 ± 4.2; CNR_VT-L, median 2.4 [IQR: 1.3-4.2] to 1.2 [IQR: 0.6-1.9]). At matched sharpness levels, Qr kernels exhibited slightly lower image noise and higher viable tumor SNR than Br kernels, particularly at low-to-moderate sharpness. Edge sharpness improved with increasing kernel sharpness for both Br and Qr, with Br demonstrating higher edge clarity for viable tumor-lipiodol deposition. Subjective assessment showed very good inter-reader agreement (κ = 0.81-0.95). Overall image quality decreased with higher kernel sharpness, whereas task-specific evaluations revealed kernel-dependent differences: Br44 provided superior vessel and viable tumor-lipiodol visualization, while Qr40 achieved better residual viable tumor clarity.</p><p><strong>Conclusion: </strong>Reconstruction kernel type and sharpness level significantly influence quantitative and subjective image quality in post-TACE PCD-CT of HCC. Optimized reconstruction kernel and sharpness settings may improve residual viable tumor conspicuity and facilitate lesion assessment.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Imaging of mass-forming intrahepatic cholangiocarcinoma: from detection to personalized treatments.","authors":"Cesare Maino, Paolo Niccolò Franco, Antonio Vizzuso, Maria Vittoria Bazzocchi, Francesco Cortese, Giuseppe Di Giovanni, Davide Ippolito, Matteo Renzulli, Riccardo Inchingolo","doi":"10.1007/s00261-026-05773-x","DOIUrl":"https://doi.org/10.1007/s00261-026-05773-x","url":null,"abstract":"<p><p>Mass-forming intrahepatic cholangiocarcinoma (MF-iCCA) represents the most frequent phenotype of intrahepatic CCA in Europe, yet its insidious clinical onset and tendency toward late presentation translate into a poor prognosis, with only 20-30% of patients eligible for curative resection. In this context, imaging plays a pivotal role, spanning from initial lesion detection and characterization to staging, treatment selection, and response assessment. Contrast-enhanced computed tomography (CT) and magnetic resonance imaging (MRI) constitute the backbone of diagnostic workup, with complementary strengths: CT offers superior spatial resolution for vascular assessment, while MRI, enhanced by diffusion-weighted imaging (DWI) and hepato-biliary phase sequences, provides higher sensitivity for intrahepatic staging and tumor delineation. Contrast-enhanced ultrasound (CEUS) serves as a valuable adjunct for differential diagnosis. Accurate resectability assessment requires careful evaluation of vascular involvement, future liver remnant (FLR) volume, and metastatic spread, with 18 F-FDG-PET playing an incremental role in nodal staging. In the era of personalized therapy, including chemo-immunotherapy, targeted agents, and locoregional treatments, functional response criteria beyond RECIST 1.1 are increasingly essential. A thorough knowledge of expected post-treatment imaging appearances is critical to avoid misclassification of treatment effects as disease progression. This narrative review aims to provide a comprehensive, imaging-focused overview of MF-iCCA, covering its radiological features and differential diagnosis, the role of imaging in staging and resectability assessment, imaging-based criteria guiding personalized treatment selection, and expected post-treatment imaging findings.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149122417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Interventional management of acute mesenteric venous thrombosis: a systematic review of current evidence, emerging techniques and future perspectives.","authors":"Michael Köhler, Ziad Maksoud, Gala Nacul Mora, Gesa Helen Pöhler","doi":"10.1007/s00261-026-05806-5","DOIUrl":"https://doi.org/10.1007/s00261-026-05806-5","url":null,"abstract":"","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149121788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lateral pelvic lymph nodes in rectal cancer: a narrative review.","authors":"Daniel Linhares Cardoso, Cesar Gabriel Rocha da Costa Paloschi, Emanuela Mendes Junqueira de Barros, Anna Gabriela Dos Santos Souza, João Miranda, Zhaohui Jin, Yasamin Sharifzadeh, Ryan W Kendziora, Eric J Dozois, Natally Horvat","doi":"10.1007/s00261-026-05805-6","DOIUrl":"https://doi.org/10.1007/s00261-026-05805-6","url":null,"abstract":"<p><p>Lateral pelvic lymph node (LPLN) involvement represents a critical determinant of oncologic outcomes in rectal cancer, directly influencing the risks of local recurrence, distant metastasis, and overall survival. High-resolution rectal magnetic resonance imaging (MRI) is the modality of choice for both initial staging and post-treatment restaging. A comprehensive approach integrating nodal size, morphology and primary tumor characteristics and treatment response is indicated for accurate evaluation. The optimal management of LPLNs differs between Eastern and Western practices, although a convergence toward selective, risk-adapted strategies is increasingly recognized. Familiarity with the imaging anatomy of the lateral pelvis and current high-risk criteria are essential for radiologists involved in the multidisciplinary care of rectal cancer patients.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149066980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A rapid automated segmentation method for total kidney volume measurement on unenhanced computed tomography in autosomal dominant polycystic kidney disease.","authors":"Yingying Hu, Tianpei Zhao, Yunpeng Gao, Peiyao Zhang, Tongxi Liu, Yue Lin, Shuya Gao, Tongyin Zhang, Yinghao Xu, Bing Li, Yu Zhang, Sheng Xie, Li Zhou, Hongliang Sun","doi":"10.1007/s00261-026-05799-1","DOIUrl":"https://doi.org/10.1007/s00261-026-05799-1","url":null,"abstract":"<p><strong>Purpose: </strong>To develop and validate a fully automated deep learning-based method for total kidney volume (TKV) measurement from unenhanced computed tomography (CT) images in patients with autosomal dominant polycystic kidney disease (ADPKD).</p><p><strong>Methods: </strong>A fully automated kidney segmentation model based on the nnU-Net framework was developed using unenhanced CT images from 236 patients, including 111 patients with ADPKD. The development dataset was divided into a training cohort and an internal testing cohort. An independent cohort of 70 patients with ADPKD was used for validation, including longitudinal scans from 20 patients. The reference standard consisted of semi-automated kidney segmentation followed by slice-by-slice manual correction performed independently by two experienced radiologists. Automated and reference segmentations were compared using the Dice similarity coefficient (DSC), intraclass correlation coefficient (ICC), linear regression and Bland-Altman analysis.</p><p><strong>Results: </strong>Automated segmentation exhibited excellent spatial agreement with the reference standard, with a mean DSC of 0.96 ± 0.01 for TKV. Volumetric agreement was near perfect (ICC = 0.999; 95% confidence interval, 0.998-0.999), with minimal bias (-0.14%) and high precision (±2.09%). Longitudinal analysis demonstrated robust agreement for kidney growth assessment (ICC = 0.99; bias, - 0.64%; precision, ± 4.81%). The automated method required approximately 5 s per case, compared with 30 min for expert-assisted reference segmentation.</p><p><strong>Conclusion: </strong>This automated segmentation method enables fast, accurate, and reproducible TKV quantification from unenhanced CT images in patients with ADPKD, offering a clinically feasible alternative to manual volumetric analysis.</p>","PeriodicalId":520541,"journal":{"name":"Abdominal radiology (New York)","volume":" ","pages":""},"PeriodicalIF":2.2,"publicationDate":"2026-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149067001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}