Abdominal Radiology最新文献

Objectives

Dual-energy CT (DECT)-derived virtual unenhanced (VUE) images have been investigated for adrenal lesion differentiation, yet previously reported thresholds vary, hampering clinical application. We aimed to test previous VUE thresholds for adrenal lesion differentiation in a large retrospective cohort, to provide a cross-validated threshold based on our data, and to investigate the influence of underlying malignancies on differentiation accuracy.

Methods

290 patients with 348 adrenal lesions (169 metastases, 179 adenomas) were included. Dual-layer DECT-derived VUE thresholds from 3 previous studies were retrieved, applied to our cohort and corresponding sensitivity/specificity/accuracy was calculated. Optimal threshold based on our data were determined using ROC-analysis with five-fold cross validation. Moreover, a threshold with similar specificity to the 10 HU threshold in unenhanced images was calculated. Subgroup analysis of adrenal lesion differentiation depending on underlying malignancies was performed.

Results

The previously suggested thresholds were 20, 22 and 29 HU, and corresponding sensitivity/specificity/accuracy was 0.61/0.92/0.76, 0.67/0.91/0.78, and 0.82/0.59/0.71, respectively. The threshold determined from our cohort was 24.7 HU, yielding a sensitivity/specificity/accuracy of 0.76/0.81/0.79. Differentiation in disease-specific subgroups showed similar sensitivity/specificity/accuracy (Melanoma:0.78/0.84/0.79; Lung cancer:0.78/0.8/0.78; RCC:0.78/1/0.79). The VUE threshold to achieve a 0.98 specificity similar to the unenhanced 10 HU cutoff was 17.3 HU, yielding a sensitivity of 0.49.

Conclusion

Previous VUE attenuation thresholds showed a varying accuracy for differentiation between adenomas and metastases. A cross-validated VUE threshold of 24.7 HU yielded a mean accuracy of 0.79, whereas a threshold of 17.3 HU was best for achieving comparable specificity as reported for the 10 HU threshold in unenhanced images.

Graphical abstract

Purpose

To establish and validate a model based on deep learning (DL), integrating radiomic features with relevant clinical features to generate nomogram, for predicting preoperative serosal invasion in gastric cancer (GC).

Methods

This retrospective study included 335 patients from dual centers. T staging (T1-3 or T4) was used to assess serosal invasion. Radiomic features were extracted from primary GC lesions in the venous phase CT, and DL features from 8 transfer learning models were combined to create the Hand-crafted Radiomics and Deep Learning Radiomics (HCR-DLR) model. The Clinical (CL) model was built using clinical features, and both were combined into the Clinical and Radiomics Combined (CRC) model. In total, 15 predictive models were developed using 5 machine learning algorithms. The best-performing models were visualized as nomograms.

Results

The total of 14 radiomic features, 13 DL features, and 2 clinical features were considered valuable through dimensionality reduction and selection. Among the constructed models: CRC model (AUC, training cohort: 0.9212; internal test cohort: 0.8743; external test cohort: 0.8853) than HCR-DLR model (AUC, training cohort: 0.8607; internal test cohort: 0.8543; external test cohort: 0.8824) and CL model (AUC, training cohort: 0.7632; internal test cohort: 0.7219; external test cohort: 0.7294) showed better performance. A nomogram based on the logistic CL model was drawn to facilitate the usage and showed its excellent predictive performance.

Conclusion

The predictive performance of the CRC Model, which integrates clinical features, radiomic features, and DL features, exhibits robust predictive capability and can serve as a simple, non-invasive, and practical tool for predicting the serosal invasion status of GC.

Graphical Abstract

Objectives

Percutaneous nephrostomy (PCN) is a standard procedure to relieve urinary obstruction. Inadvertent catheter dislodgement occurs in up to 15% of cases, and one of the management options is to replace the new catheter through the original tract. Although previous studies have reported the feasibility of reinserting a new catheter through the original tract, few studies have described the factors contributing to technical failure. This study aims to investigate the feasibility and factors affecting the technical failure of reinsertion of dislodged PCN.

Methods

53 procedures of 42 patients who underwent reinsertion of dislodged PCN through the original tract were retrospectively reviewed. All procedures were performed under fluoroscopic guidance using an angiographic catheter and a guidewire to search the original tract. After confirming access to the renal pelvis, a new catheter was inserted. Technical success was defined as the successful placement of a new catheter through the original tract. Patient characteristics, procedure details, and duration since initial PCN placement, the last replacement and dislodgement were statistically compared between technical success and failure groups.

Results

Technical success was achieved in 45 of 53 procedures (85%). Univariate analysis revealed that significant factors affecting technical failure were the smaller size of the catheter (P = 0.05), procedure time (P =  < 0.001), and duration since initial PCN placement (P = 0.001). Sex was also a significant factor; however, this may be the error due to the small sample size.

Conclusion

Salvage of dislodged PCN by inserting a catheter through the original tract was feasible. However, technical failure of catheter reinsertion was significantly affected by the shorter duration since initial PCN placement and the smaller size of catheters.

Purpose

To explore pragmatic approaches integrating MRI and PSMA-PET/CT for evaluating extraprostatic extension (EPE) of prostate cancer (PCa).

Methods

Consecutive patients with newly-diagnosed PCa that underwent multiparametric MRI and PSMA-PET/CT, followed by radical prostatectomy in 2021–2024 were included. Imaging parameters assessed on both modalities were: size, length of capsular contact (LCC), Likert scales (MRI EPE grade/PSMA Likert scale), PI-RADS/PRIMARY scores, and SUV_max. Three pragmatic integrated approaches were tested: (1) Integration of Likert scales (positive if either or both MRI and PSMA-PET/CT were positive); (2) P score (framework combining PI-RADS + PRIMARY); and (3) combining MRI morphological information with PSMA-PET/CT functional information (upgrading suspicion of lesions with LCC below cutoff if SUV_max>12). Diagnostic performance was tested with receiver operating characteristic (ROC) curves and compared using DeLong and McNemar tests.

Results

67 men (median age, 66 years) with EPE in 76.1% (51/67) were included. Area under ROC curves (AUC) were 0.61–0.82; MRI-based LCC yielded the highest AUC 0.82 (0.71–0.92) with cutoff of ≥ 1.7 cm. Integrated Likert scale (MRI EPE grade/PSMA Likert scale) showed sensitivity of 80.4% (41/51) and specificity of 31.3% (5/16). P score (PI-RADS/PRIMARY) demonstrated sensitivity of 31.3% (16/51) and specificity of 87.5% (14/16). Combining morphological MRI information with functional PSMA-PET/CT yielded sensitivity and specificity of 80.4% (41/51) and 81.2% (13/16), respectively, which demonstrated significantly higher sensitivity but non-significantly different specificity compared with MRI-based LCC alone (66.7% [34/51, p = 0.02] and 87.5% [14/16, p > 0.99]). This approach upgraded suspicion in 8 patients with LCC < 1.7 cm due to SUV_max>12 among which 87.5% (7/8) were corrected upgraded and had pathological EPE.

Conclusion

Several pragmatic approaches were explored for integrating MRI and PSMA-PET/CT to assess EPE in PCa. Combining morphological information from MRI and PSMA expression on PET/CT demonstrated good diagnostic performance and may be a simple pragmatic integrated method that can be used.

Graphical abstract

Hydatid disease is a relatively common parasitic infection. Incidence can reach more than 50 per 100,000 person-years in endemic areas. It can affect all organs but favors the liver and lungs. Presentation is often asymptomatic. Imaging plays a crucial role in diagnosis in conjunction with serology. Primary evaluation is usually by ultrasound, often followed by CT or MRI, to understand disease extent, including lesion size, internal architecture, multiplicity, location, and adjacent structure involvement. CT better detects lesion calcifications. MRI is advantageous for certain complications such as biliary communication. Cystic lesions with membranes or daughter cysts are highly indicative of hydatid disease. In comparison, diagnosis during the inactive stage can pose challenges at imaging and often also requires tissue sampling. Due to varying appearances at different developmental stages, hydatid disease can mimic various pathologies ranging from cysts to malignancy. Image-guidance aids management. Treatment is based on cyst type, size and location; with uncomplicated cysts typically amenable to PAIR (puncture, aspiration, injection, and re-aspiration) or modified catheterization technique (MoCAT) using scolicidal agents. In contrast, surgical intervention is required for complicated cysts such as those located in a hepatic subcapsular location, those that exhibit biliary communication, or have ruptured.

Purpose

This study aimed to investigate the value of integrating computed tomography (CT)-based tumor radiomics features with clinical parameters for preoperative prediction of microvascular invasion (MVI) in clear cell renal cell carcinoma (ccRCC).

Methods

We retrospectively analyzed data from a single-center cohort of ccRCC patients. Radiomics features were extracted from preoperative multiphasic CT scans (unenhanced, corticomedullary, and nephrographic phases). Following dimensionality reduction and feature selection, eight machine learning algorithms were evaluated to identify the optimal radiomics model. Independent clinical predictors were determined through univariate and multivariate analyses. A nomogram integrating the radiomics signature (rad-score) with significant clinical parameters was subsequently developed. Model performance was assessed using the area under the curve (AUC), decision curve analysis (DCA), and calibration curve analysis (CAC).

Results

Of 143 initially enrolled patients, 110 met inclusion criteria after screening, with 5502 radiomics features extracted. The support vector classifier (SVM) model demonstrated the highest discriminative ability, achieving mean AUCs of 0.976 (training cohort) and 0.892 (test cohort), significantly outperforming the clinical model (training AUC = 0.935, test AUC = 0.933). The nomogram showed superior diagnostic performance, with AUCs of 0.958 (test). DCA and CAC confirmed its clinical utility and robustness.

Conclusions

Multiparametric CT radiomics models enable non-invasive prediction of MVI status in ccRCC, with the SVM-based algorithm showing optimal performance. The integrated nomogram provides excellent and consistent diagnostic accuracy, offering a valuable preoperative tool for clinical decision-making.

Purpose

To evaluate the efficacy of deep learning reconstruction (DLR) in diffusion-weighted imaging (DWI) with single-shot echo-planar imaging (SSEPI) for endometrial cancer, compared to multiplexed sensitivity-encoding (MUSE) DWI.

Methods

We retrospectively reviewed 31 women with surgically confirmed endometrial cancer who underwent preoperative pelvic magnetic resonance imaging (MRI) including DWI. Qualitative analysis including overall image quality, susceptibility artifacts, sharpness of the uterine edge, and lesion conspicuity were compared among conventional SSEPI (SSEPI-C), SSEPI with DLR (SSEPI-DL), and MUSE using the Friedman’s test. Quantitative analysis including the apparent diffusion coefficient (ADC) values, noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were also compared among three DWI sequences using the Friedman’s test. In addition, the diagnostic accuracy for deep myometrial invasion was compared to three DWI sequences using Cochran’s Q test.

Results

The scores of overall image quality, sharpness of the uterine edge, and lesion conspicuity in SSEPI-DL were higher than SSEPI-C (p < 0.001) with no significant difference compared to MUSE (p > 0.05). Noise in SSEPI-DL was lower than SSEPI-C (p < 0.001), with no significant difference compared to MUSE (p > 0.05). SNR and CNR in SSEPI-DL were also superior to SSEPI-C (p < 0.001), and comparable to MUSE (p > 0.05). The diagnostic accuracy for detecting deep myometrial invasion showed no significant difference among SSEPI-C, SSEPI-DL and MUSE (p > 0.05).

Conclusion

DLR improves the image quality of DWI in endometrial cancer, demonstrating image quality equivalent to that of SSEPI-DL and MUSE. SSEPI-DL can be an alternative to MUSE in female pelvic MRI, with the benefit of significantly shortened scan time.

Graphical Abstract

Purpose

To assess the repeatability and reproducibility of quantitative MRCP-derived metrics generated from MRCP + software, designed for assessing biliary tree health.

Methods

Metric accuracy was assessed using a 3D-printed phantom containing 20 tubes with sinusoidally-varying diameters, simulating strictures and dilatations along ducts. Data from 80 participants (60 healthy volunteers and 20 with liver disease) was analysed in total. Repeatability and reproducibility of the quantitative metrics were assessed on Siemens, GE and Philips scanners at both 1.5T and 3T. All subjects were scanned on a Siemens Prisma 3T scanner which acted as the reference scanner. A subset of these participants also underwent scanning on the remaining scanners. Data from healthy volunteers was used to estimate the natural range of measured values (reference ranges). The reproducibility coefficient (RC) of 7 commonly reported quantitative metrics were compared between healthy controls and published values in primary sclerosing cholangitis (PSC) patients.

Results

The phantom analysis confirmed measurement accuracy with absolute bias of 0.0-0.1 for strictures and 0.1–0.2 for dilatations across all scanners (95% limits of agreement within ± 1.0). In vivo, RCs for the quantitative MRCP-derived metrics across the scanners ranged from: 12.4–25.4 for total number of ducts; 4.9–7.9 for number of dilatations; 3.3–6.5 for number of strictures; 4.6–9.8 mm for total length of dilatations; 26.5–51.7 mm for total length of strictures; and 4.4–6.8 for number of ducts with a stricture or dilatation. Repeatability on the same scanner was generally better than comparisons across scanners. Six metrics demonstrated sufficient cross-scanner reproducibility to distinguish healthy volunteers from PSC patients.

Conclusion

The precision of quantitative MRCP-derived metrics were sufficient to differentiate PSC and healthy subjects and should be well suited for multi-centre trials and assessment of biliary tree health.

Purpose

To compare the image quality and diagnostic performance of periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER), reduced field-of-view (rFOV), and conventional diffusion-weighted imaging (cDWI) combined with deep learning reconstruction (DLR) for evaluating rectal tumors.

Methods

This prospective study included 42 MRI examinations of 38 patients with rectal tumors who underwent initial staging and/or restaging MRI. PROPELLER-DWI, rFOV-DWI, and cDWI obtained with DLR were reviewed by two radiologists and compared for image quality and diagnostic performance for local tumor extent at staging and restaging and response to chemoradiotherapy at restaging.

Results

PROPELLER-DWI had significantly the least artifacts and distortions, but the worst perceptive noise, while rFOV-DWI had significantly the best sharpness for both readers (P < 0.01). For overall image quality and rectal/tumor conspicuity, PROPELLER-DWI and rFOV-DWI were significantly superior to cDWI in both readers (P < 0.01). The incidence of suboptimal image quality was significantly lower with PROPELLER-DWI and rFOV-DWI than with cDWI (5 and 1 patients with PROPELLER-DWI, 14 and 6 with rFOV-DWI, and 29 and 25 with cDWI by the 2 readers, P < 0.01). Although there were no significant differences in the accuracy of staging and restaging among the 3 types of DWI, inter-reader agreement was highest for PROPELLER-DWI (weighted kappa, 0.62–0.71) compared with cDWI (weighted kappa, 0.38–0.52) and rFOV-DWI (weighted kappa, 0.47–0.61).

Conclusions

PROPELLER-DWI and rFOV-DWI with DLR may improve the image quality of rectal DWI by reducing artifacts and distortions or increasing sharpness, although the impact on diagnostic accuracy was not significant.

Graphical abstract

Purpose

Metabolic associated steatotic liver disease (MASLD) is the most prevalent chronic liver disease, and non-invasive fibrosis staging still represents a challenge. Our main objective was to estimate the degree of liver fibrosis in these patients using T1 mapping and the extracellular volume (ECV) by MRI in comparison with liver stiffness assessed by MR elastography (MRE).

Methods

In a single-center cross-sectional study, patients with MASLD were prospectively enrolled and underwent MRI with liver T1 mapping and ECV calculations. Groups with and without significant liver fibrosis assessed by MRE were compared with the Mann–Whitney test, chi-square test, and Fisher's exact test. Correlation analysis was conducted using Spearman’s test, and a receiver operating characteristic (ROC) curve was generated to assess the ability of ECV to differentiate between groups.

Results

This study evaluated 54 patients, 37% were men, with a mean age of 58.0 ± 12.0 years. Mild liver fibrosis (F0–F1) was present in 38 patients, and significant fibrosis (F2–F4) was detected in 16 patients. Patients with significant fibrosis presented higher native T1 (954 ± 126 vs. 820 ± 123; p < 0.001) and ECV (37.9% vs. 29.1%; p < 0.001) values than those with no/mild fibrosis. Liver stiffness was correlated with native T1 (r = 0.512, p < 0.001) and ECV (r = 0.443, p < 0.001). The native liver T1 and ECV differentiated patients with and without significant liver fibrosis on MRE (AUC = 0.85 and 0.84, respectively).

Conclusion

Native T1 and ECV show potential as an alternative method for the non-invasive staging of fibrosis in patients with MASLD, although further validation in larger cohorts is needed.

Graphical abstract