Academic Radiology最新文献

{"title":"Neurovascular Coupling and Subclinical White Matter Alterations in Young Adults with Migraine without Aura: A Multimodal MRI and Machine Learning Study.","authors":"Yilei Chen, Yingnan Kong, Jiejing Zhang, Hui Wang, Songhua Zhan, Bo Wang, Wenli Tan, Jun Xu","doi":"10.1016/j.acra.2026.09.070","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.070","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>Young adults with migraine without aura (MwoA) may exhibit subtle white matter abnormalities, but their relationships with neurovascular measures, neurotransmitter distributions, and clinical burden remain unclear. We investigated these multimodal imaging associations in MwoA.</p><p><strong>Materials and methods: </strong>A total of 130 patients with MwoA and 110 healthy controls (HCs) were enrolled. Multimodal magnetic resonance imaging (MRI) was used to quantify white matter hyperintensity volume, free water (FW) fraction, peak width of skeletonized mean diffusivity (PSMD), and global neurovascular coupling (NVC); the regional measure was the cerebral blood flow (CBF)/fractional amplitude of low-frequency fluctuations (fALFF) ratio. Regional CBF/fALFF ratio alterations were spatially correlated with canonical neurotransmitter maps. Correlation, chain mediation, and machine-learning analyses were performed.</p><p><strong>Results: </strong>Relative to HCs, patients with MwoA showed increased global FW fraction and PSMD and reduced global NVC. Decreased regional CBF/fALFF ratios involved sensorimotor, striatal, and cingulate regions. The spatial pattern of NVC abnormalities was positively associated with 5-HT1A, D1, DAT, and GABAergic maps, and negatively associated with D2 and noradrenaline transporter maps. FW fraction, PSMD, and regional CBF/fALFF ratio alterations were associated with migraine-related clinical indices. Random forest yielded the numerically highest area under the curve (0.866). Exploratory mediation analysis indicated that FW and PSMD statistically accounted for part of the association between regional CBF/fALFF ratios and clinical severity.</p><p><strong>Conclusion: </strong>Young adults with MwoA exhibited subtle white matter microstructural abnormalities, reduced global NVC, and altered regional CBF/fALFF ratios. These cross-sectional findings suggest a potentially clinically relevant multimodal imaging pattern but require longitudinal and independent validation.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225399","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":"Super Resolution Reconstruction: A Strategy to Improve the Diagnostic Performance of Radiomics for Predicting Microvascular Invasion in Hepatocellular Carcinoma.","authors":"Xiaojing Liu, Xiaojie Tang, Haichen Zhao, Yuanxiang Gao, Xiaoyuan Zhou, Zhiming Li","doi":"10.1016/j.acra.2026.09.076","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.076","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>Microvascular invasion (MVI) is a key factor associated with high postoperative recurrence rate in patients with hepatocellular carcinoma (HCC). This study aims to evaluate the value of super-resolution (SR) reconstruction based on hepatobiliary phase (HBP) MR images for improving the accuracy of MVI prediction.</p><p><strong>Materials and methods: </strong>A total of 544 patients with HCC from two medical centers between 2018 and 2025 were retrospectively analyzed, including 450 patients in the training set, 51 in the internal validation set, and 43 in the external test set. Based on a lesion-aware 3D Restormer transformer architecture, SR reconstruction was performed using HBP MR images to generate super-resolution MR images. Tumor segmentations were automatically performed using the nnU-Net architecture and manually corrected by an experienced abdominal radiologist before radiomics feature extraction. Radiomics features were then extracted from original and SR images. Radiomics models based on original and SR images were developed to predict MVI status. Three machine learning algorithms, including Logistic Regression, Random Forest, and LightGBM, were used for model development.</p><p><strong>Results: </strong>A total of 851 radiomics features were extracted from the original and SR images, respectively. Redundant features were filtered, and 50 radiomics features were selected using LASSO regression with cross-validated. For MVI prediction, the pooled areas under the receiver operating curves (AUCs) of radiomics model based on the original images were 0.710, 0.764, and 0.614 in the training, validation and test sets, respectively. The corresponding AUCs of the radiomics model based on the SR images were 0.855, 0.798, and 0.827, respectively. In the test set, the AUC of the radiomics model based on the SR images was higher than that based on the original images (0.827 vs. 0.614, p <0.001).</p><p><strong>Conclusion: </strong>SR reconstruction can improve the diagnostic performance of radiomics models for predicting MVI in patients with HCC.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225423","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":"Corrigendum to \"Promotion from Associate Professor to Full Professor Should Not Be Monolithic: A National Bibliometric Study by Radiology Subspecialty\" (Acad Radiol 33 [2026] 4207-4217).","authors":"Dhairya A Lakhani, Mahla Radmard, David M Yousem","doi":"10.1016/j.acra.2026.10.009","DOIUrl":"https://doi.org/10.1016/j.acra.2026.10.009","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225406","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":"Late Career: Maintaining Relevance and Identity.","authors":"Tara Catanzano","doi":"10.1016/j.acra.2026.10.001","DOIUrl":"https://doi.org/10.1016/j.acra.2026.10.001","url":null,"abstract":"<p><p>Late career transitions to retirement are challenging and nonlinear. The transition should occur over a period of time and in a thoughtful and reflective manner. This perspective will address some of the logistics and emotions involved in planning for a transition from active radiologist to retiree.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149225379","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":"Hyperdense Renal Lesion Characterization on Three DECT Technologies: Can Iodine Quantification Differentiate Hyperattenuating Renal Cysts From Papillary Renal Cell Cancers?","authors":"Noor Fatima Majeed, Jinjin Cao, Nayla Mroueh, Avinash R Kambadakone","doi":"10.1016/j.acra.2026.09.048","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.048","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>To evaluate the performance of dual-energy computed tomography (DECT) in differentiation of hyperattenuating renal cysts (HCs) from papillary renal cell cancers (pRCCs) on three different DECT technologies (dual-source [dsDECT], rapid kVp-switching [rsDECT], and dual-layer detector-based [dlDECT]).</p><p><strong>Materials and methods: </strong>In this institutional review board-approved retrospective study, we included 85 patients with a total of 108 incidental hyperdense renal lesions detected on contrast-enhanced DECT scans (dsDECT, n = 41; rsDECT, n = 42; and dlDECT, n = 25). The study cohort included HCs (n = 73) and pRCCs (n = 35). Final confirmation was obtained with histopathology for pRCCs and with magnetic resonance imaging within 6 months of the DECT for HCs. Quantitative iodine evaluation was performed with iodine concentration (IC) measured within the lesion and the abdominal aorta. Normalized IC (NIC) was generated for each lesion by normalizing the lesion IC to the aortic IC.</p><p><strong>Results: </strong>Both IC and NIC measurements yielded significant differences between HCs and pRCCs for the dsDECT and rsDECT techniques (P < 0.001) but not for the dlDECT (P > 0.05). Cutoff values with 100% sensitivity for pRCC detection were 0.30 and 0.60 mg/mL for dsDECT and rsDECT, respectively. Normalization to the abdominal aorta allowed creation of a uniform cutoff on both technologies, with an NIC value of 0.09 allowing 100% sensitivity for each.</p><p><strong>Conclusion: </strong>Significant intertechnique differences exist for the differentiation of HCs from pRCCs. Normalization of IC to the aorta provides a uniform cutoff for differentiation of these lesions on dsDECT and rsDECT.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217306","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":"Differential Cerebral Perfusion Patterns Across Migraine Phenotypes: An Exploratory ASL MRI Study.","authors":"Yixuan Hu, Mei Huang, Chenyu Dai, Xiaoqing Liu, Zi Tao, Xiaojie Zhang, Xinhao Li, Yuxuan Wang, Kangyue Chen, Lu Han, Yan Li, Fangwang Fu, Xiaozheng Liu, Keyang Chen, Yungang Cao, Zhihan Yan, Kun Liu","doi":"10.1016/j.acra.2026.09.073","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.073","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>This study aims to characterize cerebral blood flow (CBF) alterations across migraine phenotypes and determine whether perfusion measures provide additional information for phenotype differentiation.</p><p><strong>Materials and methods: </strong>The final cohort comprised 143 patients with migraine and 49 healthy controls who underwent three-dimensional pseudo-continuous arterial spin labeling (ASL) MRI. Voxel-wise analyses of normalized CBF (nCBF) were performed to identify phenotype-related perfusion alterations. Sensitivity analyses repeated the voxel-wise comparisons with phenotype-appropriate clinical covariates. Exploratory correlation analyses examined associations between regional perfusion and clinical characteristics. Logistic regression was further used to assess whether perfusion measures provided additional information for differentiating migraine with aura (MA) from migraine without aura (MO).</p><p><strong>Results: </strong>Compared with healthy controls, migraine patients with depressive symptoms showed reduced nCBF in a right hippocampus-parahippocampal cluster (Cluster 1) in the primary voxel-wise analysis; however, this finding did not survive additional adjustment for sleep quality. Lower Cluster 1 nCBF was associated with greater headache frequency. Compared with MO, patients with MA showed increased nCBF in a right cerebellar-occipitotemporal cluster peaking in the right cerebellar Crus I (Cluster 2), and this difference remained significant after additional adjustment for sleep quality, medication-overuse headache status, and preventive medication exposure. Within the MA group, higher Cluster 2 nCBF was associated with greater headache intensity. In exploratory logistic regression analysis, the combined clinical-imaging model showed a higher apparent area under the curve (AUC) than the clinical model alone.</p><p><strong>Conclusion: </strong>Clinically relevant migraine phenotypes showed distinct patterns of regional cerebral perfusion. These findings highlight the importance of considering clinical heterogeneity in migraine neuroimaging and suggest that ASL-derived perfusion measures may provide complementary biological information for phenotype characterization, particularly with respect to aura status.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217327","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":"MRI Features in Children with Pituitary Stalk Interruption Syndrome: Associations with Hormonal Phenotype and Intracranial Anomalies.","authors":"Elif Yasar, Betul Demircan Coskun, Gozde Akin Kagizmanli, Fatma Ceren Sarioglu, Gokay Gedik, Ayhan Abaci, Handan Guleryuz","doi":"10.1016/j.acra.2026.09.057","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.057","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>Pituitary stalk interruption syndrome (PSIS) is a rare congenital disorder with variable endocrine and magnetic resonance imaging (MRI) phenotypes, and the relationship between MRI features, hormonal phenotype, and associated intracranial anomalies remains incompletely understood. This study evaluated MRI features associated with hormonal phenotype and intracranial anomalies in pediatric PSIS.</p><p><strong>Materials and methods: </strong>This retrospective study included 51 pediatric patients with biochemically confirmed pituitary hormone deficiency and MRI findings of PSIS. Pituitary stalk morphology was classified as absent, interrupted, or thin. Anterior pituitary volume was assessed by two- and three-dimensional methods, and two-dimensional volumes were converted to age- and sex-specific z-scores. We analyzed associations between MRI findings, hormonal phenotype, and intracranial anomalies using univariable and multivariable methods.</p><p><strong>Results: </strong>Growth hormone deficiency was present in 49 patients (96.1%) and multiple pituitary hormone deficiency (MPHD) in 23 (45.1%). Intracranial anomalies were detected in 31 patients (60.8%), most commonly corpus callosum dysgenesis. MPHD was more frequent with an absent stalk (18/23, 78.3%) than with an interrupted (3/8, 37.5%) or thin stalk (2/20, 10.0%) (p < 0.001), and intracranial anomalies followed the same gradient (p < 0.001). After adjustment for age and intracranial anomalies, absent or interrupted stalk morphology remained independently associated with MPHD (odds ratio 16.8, 95% confidence interval 2.2-130.1; p = 0.007), driven primarily by the absent-stalk group. Neither two- nor three-dimensional anterior pituitary volume nor volume z-scores differed significantly between the isolated deficiency and MPHD groups; sella morphology also showed no significant association with hormonal phenotype.</p><p><strong>Conclusion: </strong>Pituitary stalk morphology, particularly an absent stalk, showed the strongest association with hormonal phenotype and intracranial anomalies in pediatric PSIS.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217294","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":"A Hybrid Human-AI Workflow for Ultrasound Report Quality Control: A Multicenter Performance and Efficiency Evaluation.","authors":"Xuan Liang, Hui He, Chunzheng Liu, Rui Sang, Jialin Zhu, Xiaoqing Wang, Wenjing Hou, Deyi Li, Junguang Wang, Yuanyuan Li, Jiayu Su, Xiaoyi Wang, Xi Wei","doi":"10.1016/j.acra.2026.09.069","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.069","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>This study evaluated a hybrid human-AI workflow for ultrasound report quality assurance using multicenter retrospective data.</p><p><strong>Materials and methods: </strong>786 anonymized ultrasound reports covering 22 examination categories were collected from four hospitals. Reports containing naturally occurring or implanted errors were matched across arms. GPT-4o and DeepSeek-v3 performed initial screening before review by six human reviewers (HRs). HR with blinded expert over-reading (HR-ex) served as control. Accuracy, false-positive (FP) and false-negative (FN) rates, and review time were compared. Supplementary validation compared HR and DeepSeek-assisted review using 1680 consecutive reports without implanted errors.</p><p><strong>Results: </strong>Hybrid workflow (HYB) accuracy was comparable to HR (4o-HYB 80.2%, v3-HYB 82.5%, HR 82.3%) but lower than HR-ex (87.3%). Both HYBs showed higher FN rates than HR-ex (4o-HYB 34.2% and v3-HYB 27.3% vs HR-ex 16.9%, both p < 0.001). Severity assessment did not indicate preferential omission of clinically significant errors. Human verification reduced LLM-generated FNs and FPs. Median review time decreased substantially from 51.0 s (HR-ex) to 11.0 s (4o-HYB) and 12.0 s (v3-HYB). In supplementary validation (error prevalence, 9.6%), v3.2-HYB and HR showed FN rates of 22.6% and 24.3%, FP rates of 7.7% and 10.0%, and median review times of 15.4 and 24.3 s, respectively.</p><p><strong>Conclusion: </strong>The LLM-assisted HYB achieved accuracy comparable to single-physician review while substantially reducing audit time. These findings require prospective real-world validation before routine clinical implementation.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217340","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":"Ultra-High Gradient Thin-Slice Oblique Coronal Zoomed DWI for Perihilar and Distal Cholangiocarcinomas: Improved Tumor Depiction for Precise Surgical Planning.","authors":"Lingsong Zhou, Beixuan Zheng, Xianling Qian, Xuhao Song, Jili Chen, Caixia Fu, Houbao Liu, Ruofan Sheng, Mengsu Zeng","doi":"10.1016/j.acra.2026.09.064","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.064","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>Accurate preoperative assessment of extrahepatic cholangiocarcinoma (eCCA) remains challenging with conventional imaging. This study aimed to investigate the value of thin-slice oblique coronal zoomed diffusion-weighted imaging (OC-Z-DWI) on an ultra-high gradient system for eCCA (including both perihilar and distal subtypes) assessment.</p><p><strong>Materials and methods: </strong>This prospective study enrolled 93 patients with pathologically diagnosed eCCA (54 males, 66.8 ± 9.0 years; 62 perihilar and 31 distal) and 43 patients without eCCA (19 males, 62.9 ± 11.8 years). All patients underwent conventional magnetic resonance imaging and OC-Z-DWI on an ultra-high gradient system (200 mT/m amplitude). Image quality of OC-Z-DWI was evaluated, and diagnostic performance and lesion sizing were compared between the conventional protocol and one incorporating OC-Z-DWI.</p><p><strong>Results: </strong>OC-Z-DWI showed good scoring in noise, artifacts, and overall image quality (mean 3.83 ± 0.41, 3.65 ± 0.56, and 3.80 ± 0.43, respectively). Adding OC-Z-DWI to the conventional protocol significantly improved lesion edge sharpness (3.89 ± 0.31 vs 3.60 ± 0.53, P < 0.001), lesion conspicuity (3.94 ± 0.25 vs 3.75 ± 0.46, P < 0.001), diagnostic confidence (3.86 ± 0.38 vs 3.58 ± 0.65, P < 0.001), and diagnostic performance (AUC 0.917 vs 0.869, P = 0.018). Lesion size measurements with the OC-Z-DWI-based protocol demonstrated strong pathological correlation (r = 0.822), with lower pathological-imaging diameter discrepancy than the conventional protocol (median 4.0 mm vs 7.0 mm, P < 0.001). These improvements were consistent across both perihilar and distal subgroups. Furthermore, R0 resection patients exhibited a stronger imaging-pathology sizing correlation than R1 patients (r = 0.891 vs 0.528, P = 0.021).</p><p><strong>Conclusion: </strong>Ultra-high gradient thin-slice OC-Z-DWI provides high-quality images for eCCA characterization. Adding OC-Z-DWI to the conventional magnetic resonance protocol enhances lesion delineation, diagnostic performance, and tumor sizing accuracy, delivering more precise preoperative information to optimize surgical planning.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149208120","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":"Retrospective Lung-RADS v2022 Characterization of Fatal Lung Cancer Cases in the National Lung Screening Trial.","authors":"Ashita Tanwar, Sayedomid Ebrahimzadeh, Maya Vella, Hannah Ahn, Alexander Beagle, Tician Schnitzler, Masha Bondarenko, Adam Yala, James Brown, Jae Ho Sohn","doi":"10.1016/j.acra.2026.09.065","DOIUrl":"https://doi.org/10.1016/j.acra.2026.09.065","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>Persistent lung cancer mortality despite screening is multifactorial. Characterizing computed tomography (CT) findings preceding fatal lung cancer may identify imaging features and screening challenges that warrant further study.</p><p><strong>Materials and methods: </strong>We conducted a retrospective, outcome-enriched case series nested within the 26,722-participant low-dose computed tomography arm of the National Lung Screening Trial. Among 359 participants who died of lung cancer within the prespecified interval, exclusions for positive baseline screening (n = 186), incomplete screening (n = 28), withdrawal/declined screening (n = 19), unavailable CTs (n = 6), or three negative screens followed by death 2+ years after the final screen (n = 29) yielded 91 participants. Screening CTs were retrospectively assessed using the Lung Imaging Reporting and Data System (Lung-RADS) v2022. Four thoracic radiologists independently reviewed visible-in-retrospect nodules and adjudicated whether they met criteria for suspicion.</p><p><strong>Results: </strong>Of 91 participants, 22 (24%; 95% confidence interval (CI), 17%-34%) had visible-in-retrospect nodules considered suspicious under Lung-RADS v2022 (\"missed\"), 16 (18%; 95% CI, 11%-27%) had visible nodules not meeting suspicion criteria (\"not callable\"), 50 (55%; 95% CI, 45%-65%) had no visible-in-retrospect nodule, and 3 (3.3%) were inconclusive. Missed nodules were frequently juxtapleural (41%), obscured (50%), or irregularly bordered (68%); not-callable nodules were frequently <6 mm (75%), juxtapleural (56%), or irregularly bordered (56%).</p><p><strong>Conclusion: </strong>In this selected fatal-case series, small and juxtapleural nodules were common among visible nodules not meeting Lung-RADS v2022 suspicion criteria. These findings warrant evaluation in representative screening populations and should not be interpreted as estimates of population-level Lung-RADS performance.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149208088","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}