Academic Radiology最新文献

RATIONALE AND OBJECTIVES: This study aims to compare the diagnostic performance of fibroblast activation protein inhibitor (FAPI) and ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) in patients with cervical cancer.

Methods: In this review, we performed a systematic search of studies published in PubMed, Web of Science, and Embase databases up to October 30, 2025. All included studies used radionuclide labeled FAPI and ¹⁸F-FDG to evaluate their performance in patients with cervical cancer.

Results: In patient-based analysis (n=109), FAPI showed higher sensitivity than ¹⁸F-FDG (0.96 vs 0.77) and greater specificity (0.98 vs 0.86). Lesion-based analysis (n=203 lymph nodes) yielded similar sensitivity patterns (0.99 vs 0.72) but may inflate precision due to within-patient clustering. Semi-quantitative parameters (SUVmax) showed variable patterns across studies (FAPI SUVmax range: 15.1-18.3 for primary lesions; FDG SUVmax range: 13.96-17.1). Descriptive comparison suggests FAPI SUVmax was generally comparable to or higher than FDG values, though formal pooled statistical comparison was not performed due to substantial heterogeneity in PET platforms, tracer variants, and acquisition protocols.

Conclusion: Compared with ¹⁸F-FDG, FAPI PET/CT and PET/MR are suggested to be promising imaging modalities with favorable sensitivity for cervical cancer. However, these results should be interpreted cautiously due to heterogeneity among studies and potential verification bias. Larger prospective studies are needed to confirm it in the future.

Rationale and objectives: To quantify changes in procedure times for interventional radiology (IR) attendings post-fellowship.

Materials and methods: Between 8/2010 and 12/2023, we examined 67943 procedures across 269 procedures for 24 IR attendings post-fellowship at a tertiary teaching hospital. Procedure time and months of attending practice were collected to create procedure time learning curves for individual procedures. Procedure times at 1 year and 6-12 years of practice were compared using student's t-test. To create an aggregate learning curve across all procedures, a normalized procedure time (NPT) was calculated by subtracting a faculty's median career procedure time from their individual procedure times across periods of attending practice.

Results: Sirsphere treatment showed the greatest reduction in procedure time between 1 and 6-12 years of practice (232 vs 155 mins, p<0.001), followed by soft tissue ablation (173 vs 120 mins, p<0.001), mapping arteriogram (216 vs 164 mins, p<0.001), embolization tumor hepatic (229 vs 185 mins, p<0.001), bone marrow aspiration (75 vs 50 mins, p<0.001), nephrostomy exchange (71 vs 53 mins, p<0.001) and pleurx catheter placement (87 vs 71 mins, p<0.001). The aggregate IR procedure time learning curve showed decreasing NPT in year 1 (-5.3 mins/procedure/year, p< 0.19), followed by slower steady significantly decreasing NPT between 1 and 6-12 years of practice (-0.7 mins/procedure/year, p<0.0001).

Conclusion: During the first year of practice post-fellowship, IR attendings have the greatest reduction in procedure times for locoregional cancer treatments. Aggregating all procedures during the first 12 years post-fellowship, IR attendings demonstrate a biphasic learning curve.

Rationale and objectives: Coronary artery disease is associated with cognitive and neuropsychiatric dysfunction through mechanisms that remain poorly understood. The glymphatic system-a cardiac-driven brain clearance network-may provide a mechanistic link. This study examined whether perivascular diffusivity DTI-ALPS differs between coronary disease patients and controls, correlates with disease severity and symptoms, and responds to revascularization.

Methods: This cross-sectional study with seven-day post-revascularization follow-up enrolled 90 coronary artery disease patients and 90 matched controls. DTI-ALPS indices were calculated from 3.0 T MRI to assess periventricular water diffusivity. Cognitive function (Mini-Mental State Examination, Montreal Cognitive Assessment), fatigue (Fatigue Severity Scale), and depression (Hamilton Depression Rating Scale) were evaluated. Cardiac disease severity was quantified using myocardial perfusion imaging. Bootstrap mediation analysis examined whether DTI-ALPS statistically accounted for cardiac-symptom associations.

Results: Coronary artery disease patients demonstrated reduced DTI-ALPS indices compared to controls (1.28±0.21 vs 1.38±0.18, P=0.001, Cohen's d=0.52). DTI-ALPS correlated with cognitive screening performance (r=0.298-0.341, P<0.01) and symptom severity (r=-0.278 to -0.325, P<0.01). Mediation analysis revealed that reduced DTI-ALPS statistically accounted for 14.7-16.8% of cardiac-symptom associations (all bootstrap 95% confidence intervals excluded zero, P<0.05). DTI-ALPS demonstrated dose-response relationships with diseased vessels and myocardial perfusion deficits. Following revascularization, DTI-ALPS increased (P=0.013) alongside trends toward mood improvement, although neuropsychological screening measures did not reach statistical significance.

Conclusion: Reduced periventricular diffusivity assessed by DTI-ALPS is associated with both coronary artery disease and neuropsychiatric symptoms, statistically accounting for a modest proportion of their association. These cross-sectional findings suggest potential cardiac-brain clearance pathway connections that require prospective validation.

Take-home message:

Rationale and objectives: Accurate preoperative evaluation of human epidermal growth factor receptor 2 (HER2) status is crucial for guiding personalized treatment in gastric cancer (GC). This study aimed to evaluate the clinical value of CT-based multi-subregion habitat radiomics for noninvasive preoperative prediction of HER2 expression.

Materials and methods: This retrospective study enrolled 857 pathologically confirmed GC patients from two medical centers. Patients were stratified into HER2-positive and HER2-negative groups according to postoperative pathology. At Center 1, 657 patients were randomly divided into training (n = 460) and testing (n = 197) cohorts (7:3 ratio), while 200 patients from Center 2 served as an external validation (ex-vad) cohort. Radiomics features were extracted from contrast-enhanced venous-phase CT images, and a radiomics score (Radscore) was subsequently generated. Tumor regions of interest underwent unsupervised clustering to define habitat subregions, forming models based on two, three, and four subregions. The optimal habitat model was selected based on predictive accuracy. A combined model integrating the optimal habitat model and key clinical features was subsequently constructed.

Results: Six models were developed: clinical, Radscore, Habitat2, Habitat3, Habitat4, and combined (Habitat3 plus clinical variables). In the training cohort, area under the receiver operating characteristic (ROC) curve values were 0.68 (95% confidence intervals (CI), 0.62-0.75), 0.78 (0.73-0.83), 0.75 (0.69-0.81), 0.90 (0.87-0.94), 0.78 (0.73-0.84), and 0.94 (0.91-0.97), respectively. The combined model achieved superior predictive performance in both training, test and ex-vad cohorts, followed by Habitat3.

Conclusion: The Habitat3 model robustly predicts HER2 expression in GC. The combined model further improves predictive accuracy.

Summary statement: This study developed CT-based radiomics habitat models for predicting HER2 expression in gastric cancer. Habitat3 achieved optimal performance, and its combination with clinical variables yielded superior efficacy and clinical applicability, providing a robust non-invasive tool for preoperative HER2 assessment in gastric cancer.

Rationale and objectives: This study aimed to develop and validate a deep learning-based brain metastasis detection model (BMDM) in magnetic resonance images for diagnosing brain metastases (BMs).

Materials and methods: We retrospectively collected data from 950 patients serving as the training and test sets for developing BMDM and from an additional 423 patients as the validation set. Three reading modes were compared: radiologists only (10 total, four with ≤3 years of experience and six with >3 years of experience), BMDM only, and radiologists assisted by the BMDM. The alternative free-response receiver operating characteristic (AFROC) method was used for evaluation.

Results: The reading time was reduced by 30.87%, AFROC-area under the curve improved from 0.837 to 0.954, and sensitivity increased from 0.685 to 0.916 with BMDM assistance. The improvement in sensitivity was more pronounced among less experienced radiologists (24.59% vs 22.03%). The detection sensitivity improved by 33.45% for lesions ≤3 mm and by 43.00% for insular lesions.

Conclusion: The results demonstrated that BMDM significantly enhanced time efficiency and diagnostic performance for BM detection, providing clinical benefits.

Rationale and objectives: Cholangiocarcinoma (CCA) is an aggressive malignancy often diagnosed at an advanced stage, precluding curative resection. In such patients, palliative interventions to relieve biliary obstruction are critical. Endobiliary radiofrequency ablation (RFA) using the Habib™ EndoHPB probe has been explored as an adjunct to improve biliary patency, though its percutaneous application and complication rate remain underexplored.

Materials and methods: We retrospectively reviewed eight patients (mean age: 59.9 ± 6.2 years; four males, four females) with unresectable intrahepatic/perihilar CCA who underwent percutaneous endobiliary RFA between January 2022 and June 2025. Outcomes included technical and clinical success, laboratory trends, adverse events, and interventional radiology (IR) service utilization.

Results: Technical success was achieved in all patients (100%). Clinical success, defined as ≥50% bilirubin reduction from baseline within one month, and either symptom resolution within 14 days or improvement in biliary obstruction-related clinical parameters, was observed in seven patients (87.5%). No major procedure-related adverse events were noted, including vascular complications, despite endobiliary ablation. Following RFA, mean total bilirubin decreased from 5.5 ± 7.6 mg/dL to 2.7 ± 2.7 mg/dL at one month (p = 0.187). Aspartate aminotransferase (AST) declined from 85.9 ± 48.1 U/L to 59.7 ± 39.5 U/L (p = 0.005), and alanine aminotransferase (ALT) decreased from 104.7 ± 76.2 U/L to 56.1 ± 50.8 U/L (p = 0.054). The frequency of drain exchanges improved, with intervals extending from 1-3 weeks pre-RFA to 4-8 weeks post-RFA. Over a mean follow-up of 398.6 ± 562.3 days, three patients (37.5%) achieved drain internalization, while two deaths were attributed to disease progression.

Conclusion: Percutaneous endobiliary RFA with the Habib probe is safe, technically feasible, and associated with favorable laboratory and clinical outcomes in unresectable CCA.

Purpose: This study assesses whether automated segmentation allows evaluation of subtle changes in volume and intrinsic tissue characteristics in an obese patient population undergoing metabolic surgery.

Materials and methods: Multiorgan segmentation and intrinsic Hounsfield Unit (HU) analyses were performed in 1877 preoperative/postoperative computed tomography (CT) examinations, processing nine compartments, organ systems, and musculoskeletal structures as follows: subcutaneous, peritoneal/retroperitoneal fat, skeletal muscle and liver, spleen, kidneys, pancreas, and first lumbar vertebral body. Statistical assessments utilized Wilcoxon signed-rank test and generalized linear mixed model comparisons successively using volumes and HU as dependent values, preoperative/postoperative state as fixed factor, and gender, age as covariates.

Results: Metabolic surgery led to significant reductions of subcutaneous, retroperitoneal/peritoneal fat volumes averaging 25% and 29%, respectively, p<0.001. Subcutaneous and retroperitoneal/peritoneal fat compartments differed significantly in HU, p<0.001. Postoperative reduction in fat tissue was accompanied by a significant increase in corresponding median HU, increasing by 6% and 8%, respectively, p<0.001. Volumes of liver, spleen, pancreas, and kidneys were significantly impacted by metabolic surgery with intraindividual shrinkages of 7-11% postoperatively, p<0.001. Core striated musculature showed significant decreases in volume, averaging 4%, and a simultaneous significant increase in HU following metabolic surgery, averaging 4%. Throughout, gender was a significant covariate with larger volumes preoperatively, a more substantial volume loss postoperatively, as well as a higher HU increase, all observed in males.

Conclusion: Simultaneously, operator-independent and automated compartmental quantitative segmentation allows for assessment of subtle changes in volume and intrinsic structural tissue characteristics over time in patients undergoing metabolic surgery. Our results underline the complex transformation of fat, organs, and muscles after surgery-induced weight loss.

Rationale and objectives: This study aims to develop a comprehensive nomogram for predicting the 3-year recurrence risk of patients with soft-tissue sarcoma (STS) undergoing surgical resection based on preoperative MRI images and clinical-radiological factors.

Materials and methods: 202 patients with STS of the extremities and trunk who had undergone surgical resection were included from two centers. We extracted tumor and peritumoral radiomics features from contrast-enhanced T1-weighted imaging (CE-T1WI) and fat-saturated T2-weighted imaging (FS-T2WI) sequences to construct corresponding models, and used pre-trained VGG11 and ResNet18 networks to build sequence-specific deep learning models. A clinical-radiological model was built using selected clinical-radiological features. Finally, deep learning, tumor and peritumoral radiomics, and clinical-radiological analysis results were integrated to construct a comprehensive nomogram for systematic evaluation and analysis from multiple perspectives.

Results: Among all STS patients, the 3-year postoperative recurrence rate was 47.52% (96/202). The nomogram showed excellent predictive performance, with AUC values of 0.874(95% confidence interval [CI]: 0.761-0.987) and 0.822 (95% CI: 0.707-0.938) in internal and external validation sets, respectively; its concordance index for 3-year recurrence risk prediction was 0.746 and 0.690 in the two sets. Kaplan-Meier curves demonstrated significant prognostic differences in patient stratification across all cohorts (log-rank test, all p < 0.01).

Conclusions: The nomogram can predict the 3-year recurrence risk of patients, identify high-risk patients, and support personalized treatment.