Acta radiologica最新文献

Background: Congenital coronary artery to pulmonary artery fistulas (CPAFs) are extremely rare congenital vascular malformations.

Purpose: To give a practical approach and consider technical challenges and pitfalls for endovascular embolization of CPAF.

Material and methods: Anatomic, technical, and pathophysiologic considerations are given and demonstrated for antegrade and retrograde endovascular embolization of CPAF.

Results: Antegrade embolization is easier to perform, saves radiation exposure, and is recommended especially in younger patients. In case of a single dominant feeder, antegrade embolization of this feeder might sufficiently treat the CPAF. Retrograde embolization from the pulmonary orifice is technically more challenging but leads to a complete and definite closure of the fistula in one single step.

Conclusion: Patient age and fistula configuration must be taken into consideration for appropriate treatment approach in CPAF. Prerequisite for successful embolization of CPAF is profound clinical and interventional experience, why we highly recommend to both plan and carry out embolization of CPAF as interdisciplinary procedure.

Background: Patellar instability (PI) or patellar dislocation (PD) is challenging to diagnose accurately based on medical history and clinical manifestations alone. While X-ray, computed tomography (CT), and magnetic resonance imaging (MRI) are commonly employed for detecting PI or PD, computer vision has not yet been widely utilized for this purpose.

Purpose: To explore the feasibility of computer vision, specifically the You Only Look Once (YOLO) algorithm, in identifying patellar instability or dislocation.

Material and methods: A total of 550 patients (190 diagnosed with patellar instability or dislocation) were divided into a training set (n = 360), validation set (n = 90), and external test set (n = 100). Four indicators were measured on transverse knee MRI scans to determine the presence of patellar instability, and 450 images were labeled using Labelme software. YOLO version 8 (YOLOv8) was refined using these labeled images and validated on 100 unlabeled images. The diagnostic accuracy of YOLOv8 was compared with that of a junior radiologist.

Results: The sensitivity, specificity, and accuracy of the refined YOLO model and the junior radiologist were 62%, 97%, and 83%, and 62%, 82%, and 74%, respectively. Although the YOLO model demonstrated slightly higher accuracy, the difference did not reach statistical significance (P = 0.093). The YOLO model required approximately 14.01 ± 10.34 ms to interpret each image, significantly shorter than the 9.55 ± 2.39 s required by the radiologist (P < 0.001).

Conclusion: The refined YOLOv8 model is not inferior to junior radiologists in identifying patellar instability or dislocation and offers a significantly faster interpretation time.

Background: 68Ga-labled fibroblast activating protein inhibitor (68Ga-FAPI) represents a new and exciting positron emission tomography-computed tomography/magnetic resonance (PET-CT/MR) radiotracer.

Purpose: To compare the diagnostic efficacy of 68Ga-FAPI PET CT/MR and 18F-fluorodeoxyglucose (18F-FDG) PET/CT in metastatic lesions of gynecological cancers (GCs).

Material and methods: The PubMed, Embase, and Web of Science databases were thoroughly investigated from inception until 22 December 2023. A head-to-head contrast between 18F-FDG PET/CT as well as 68Ga-FAPI PET CT/MR for the assessment of GCs was presented by the included studies. A random variable model was employed to examine the sensitivity in detection of lymph node (LN) and peritoneal metastases (PM).

Results: The pooled sensitivity for 68Ga-FAPI PET CT/MR and 18F-FDG PET/CT in lymph node metastases (LNM) of GC were 0.98 (95% confidence interval [CI] = 0.86-1) and 0.85 (95% CI = 0.65-0.98), respectively, while the results about peritoneal metastases in ovarian cancer were 0.98 (95% CI = 0.93-1) and 0.71 (95% CI = 0.55-0.86). Compared with 18F-FDG PET/CT, 68Ga-FAPI PET CT/MR exhibited a better sensitivity in peritoneal involvement of ovarian cancer with a relative risk of 0.24 (95% CI = 0.09-0.40) and P = 0.002.

Conclusion: 68Ga-FAPI PET CT/MR displayed a superior sensitivity over 18F-FDG PET/CT in detecting metastatic lesions of ovarian cancer. However, there was insufficient evidence to favor the superiority of 68Ga-FAPI PET CT/MR in LNM of CC. Further studies are needed for evaluating primary and metastatic lesions of 68Ga-FAPI PET CT/MR in different GC.

Background: The O-RADS scoring has been proposed to standardize the reporting of adnexal lesions using magnetic resonance imaging (MRI).

Purpose: To assess intra- and inter-observer agreement of the O-RADS scoring using non-dynamic MRI and its agreement with pathologic diagnosis, and to provide the pitfalls in the scoring based on discordant ratings.

Material and methods: Adnexal lesions that were diagnosed using non-dynamic MRI at two centers were scored using O-RADS. Intra- and inter-observer agreements were assessed using kappa statistics. Cross-tabulations were made for intra- and inter-observer ratings and for O-RADS scores and pathological findings.

Results: Intra- and inter-observer agreements were assessed for 404 lesions in 339 patients who were admitted to center 1. Intra-observer agreement was almost perfect (97.8%, kappa = 0.963) and inter-observer agreement was substantial (83.2%, kappa = 0.730). The combined data from center 1 and center 2 included 496 patients; of them, 295 (59.5%) were operated. There was no borderline or malignant pathology for the lesions with O-RADS 1 or 2. Of those with an O-RADS score of 3, 3 (4.1%) lesions were borderline and none were malignant. The O-RADS scoring in discriminating borderline/malignant lesions from benign lesions was outstanding (area under the ROC curve 0.950, 95% CI = 0.923-0.971). Sensitivity, specificity, positive, and negative predictive values of O-RADS 4/5 lesions for borderline/malignant lesions were 96.2%, 87.1%, 72.8%, and 98.4%, respectively.

Conclusion: The O-RADS scoring using non-dynamic MRI is a reproducible method and has good discrimination for borderline/malignant lesions. Potential factors that may lead to discordant ratings are provided here.

Background: The recognition of thin marginal spinal syndesmophytes is important, in part due to their association with non-traumatic or mildly traumatic vertebral fractures.

Purpose: To determine a lower limit on the prevalence of marginal spinal syndesmophytes using chest radiographs.

Material and methods: We conducted a retrospective analysis of 500 chest radiographs, assessing the prevalence of thin marginal syndesmophytes, bridging or near-bridging osteophytes, and flowing paravertebral ossifications in the thoracic intervertebral discs among individuals aged 16 years and older in a North American city.

Results: Among the 500 participants, we observed that thin vertical marginal syndesmophytes were present in 17 (3.4%) cases, bridging or near-bridging osteophytes were present in 126 (25.2%) cases, and flowing paravertebral ossifications were present in 37 (7.4%) cases. Out of the 17 participants with thin marginal syndesmophytes, 10 exhibited a bamboo-like spine appearance, defined as the presence of ≥4 contiguous levels of bridging marginal syndesmophytes. Analysis of syndesmophyte distribution per vertebral level indicated a higher frequency of involvement in the mid to lower thoracic spine, maximal at T9/10.

Conclusions: The presence of thin marginal syndesmophytes in the thoracic spine on routine chest radiographs is substantially more prevalent than would be anticipated based on the existing literature. The feasibility of reliably identifying these syndesmophytes in the spine and the impact of this on morbidity should be further investigated due to their association with advanced ankylosing spondylitis and their susceptibility to fractures.

Background: Dynamic susceptibility contrast-enhanced magnetic resonance imaging (DSC-MRI) can reflect the angiogenesis of ischemic stroke.

Purpose: To investigate the value of DSC-MRI with ultrasmall superparamagnetic particles of iron oxides (USPIO) in evaluating angiogenesis in the peri-infarction zones in subacute ischemic stroke in a permanent middle cerebral artery occlusion (pMCAO) rat model.

Material and methods: A total of 21 Sprague-Dawley rats were randomly divided into the pMCAO and sham operation groups. Every rat in each group underwent DSC-MRI with USPIO at 3, 5, and 7 days. DSC-MRI parameters of the relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), relative mean transit time (rMTT), and relative time to peak (rTTP) were measured, calculated, and compared among the different times. Sequential correlations were analyzed among the histopathological indexes with the microvascular density (MVD) and percentage of vascular area (%VA), the serum factors with vascular endothelial growth factor (VEGF), vascular cell adhesion molecule 1 (VCAM-1), and perfusion parameters, respectively.

Results: The rCBV and rCBF in the peri-infarction area of pMCAO rats were significantly higher on day 7 than on day 3, whereas no significant changes in rMTT and rTTP were observed at 3, 5, and 7 days. Significantly positive correlations were found between rCBV and MVD, %VA, VEGF, VCAM-1, between rCBF and MVD, %VA, VEGF, and VCAM-1 at 3, 5, and 7 days in the pMCAO group.

Conclusion: The rCBV and rCBF deriving from USPIO-DSC may be potentially useful for evaluating the angiogenesis of the peri-infarction zones in the subacute phase of ischemic stroke.

Background: The accurate evaluation of tumor response after locoregional therapy is crucial for adjusting therapeutic strategy and guiding individualized follow-up.

Purpose: To determine the inter-reader agreement of the LR-TR algorithm for hepatocellular carcinoma treated with locoregional therapy among radiologists with different seniority.

Material and methods: A total of 275 treated observations on 249 MRI scans from 99 patients were retrospectively collected. Three readers of different seniorities (senior, intermediate, and junior with 10, 6, and 2 years of experience in hepatic imaging, respectively) analyzed the presence or absence of features (arterial-phase hyperenhancement and washout) and evaluated LR-TR category.

Results: There were substantial inter-reader agreements for overall LR-TR categorization (kappa = 0.704), LR-TR viable (kappa = 0.715), and LR-TR non-viable (kappa = 0.737), but fair inter-reader agreement for LR-TR equivocal (kappa = 0.231) among three readers. The inter-reader agreement was substantial for arterial-phase hyperenhancement (kappa = 0.725), but moderate for washout (kappa = 0.443) among three readers. The inter-reader agreements between two readers were substantial for overall LR-TR categorization (kappa = 0.734, 0.727, 0.652), LR-TR viable (kappa = 0.719, 0.752, 0.678), and LR-TR non-viable (kappa = 0.758, 0.760, 0.694), which were at the same level as the inter-reader agreements among three readers. In addition, the inter-reader agreements between two readers were substantial for arterial-phase hyperenhancement (kappa = 0.733, 0.766, 0.678), but moderate for washout (kappa = 0.473, 0.422, 0.446), which were at the same level as the inter-reader agreements among three readers.

Conclusion: LR-TR algorithm demonstrated overall substantial inter-reader agreement among radiologists with different seniority.

Background: Cervical cancer is a major cause of morbidity and mortality among gynecological malignancies. Diagnostic imaging of lymph node (LN) metastasis for prognosis and staging is used; however, the accuracy in classifying the stage needs to improve.

Purpose: To examine the accuracy of AI-based radiomics in diagnosis, prognosis assessment and predicting the diagnostic value of radiomics for pelvic LN metastasis in cervical cancer patients.

Material and methods: The study included 118 female patients with 660 LNs and 118 merged LNs. Four imaging histology models-decision tree, random forest, logistic regression, and support vector machine (SVM)-were created in this study. The imaging histology features were extracted from both the independent and merged LN groups. The AUC values for the test sets and the training sets of the four imaging histology models were compared for the independent LN group and the merged LN group. The DeLong test was used to compare the models.

Result: The imaging histology prediction model developed in the merged LN group outperformed the independent LN group in terms of test set AUC (0.668 vs. 0.535 for decision tree, 0.841 vs. 0.627 for logistic regression, 0.785 vs. 0.637 for random forest, 0.85 vs. 0.648 for SVM) and accuracy (0.754 vs. 0.676 for decision tree, 0.780 vs. 0.671 for random forest, 0.848 vs. 0.685 for logistic regression, 0.822 vs. 0.657 for SVM).

Conclusion: The constructed SVM imaging histology model for the merged LN group might be advantageous in predicting pelvic LN metastasis in cervical cancer.

Background: Breast cancer in pediatric patients is rare, but ultrasound (US) is widely utilized for symptomatic cases.

Purpose: To determine biopsy and cancer detection rates of pediatric patients and to assess if breast US can be omitted.

Material and methods: A retrospective review of a 5-year period was conducted of single-center breast US performed in patients aged <19 years. Data regarding presentation, clinical opinion (P1-5 score), and US (U1-5 score) were collected. If biopsy or surgery was performed, pathology was reviewed (B1-5 score).

Results: In total, 579 patients were included (19 boys, 560 girls; mean age=16.2±1.9 years; age range=0-18 years). Clinical examination was normal or benign (P1/P2) in all boys (100%) and 557/560 (99.5%) girls, and P3 in 3 (0.5%) girls. Of US, 52% demonstrated normal findings (U1) for both sexes (300/579); in the remaining cases, the most frequent findings were gynecomastia in 12/19 boys and well-defined breast masses in 208/560 girls. Of the 560 girls, 6 (1%) underwent US-guided biopsy, with final histology of fibroadenoma (B2) in all cases, while 27 (5%) had a surgical excision, with final histology of fibroadenoma (22/27, 81.5%), hamartoma (2/27, 7.4%), benign phyllodes tumor (2/27, 7.4%), and angiomyxoma skin lesion (1/27, 3.7%). No malignant lesions were diagnosed at the time of clinical referral or during the 18-month follow-up in patients with a well-defined mass on US.

Conclusion: Breast malignancy is extremely rare in pediatric population. US can be safely omitted if clinical examination is normal; this approach would have avoided breast US in 52% of patients in this study.

Background: Myocardial fibrosis is often detected in patients with hypertrophic cardiomyopathy (HCM), which causes left ventricular (LV) dysfunction and tachyarrhythmias.

Purpose: To evaluate the potential value of a machine learning (ML) approach that uses radiomic features from late gadolinium enhancement (LGE) and cine images for the prediction of ventricular tachyarrhythmia (VT) in patients with HCM.

Material and methods: Hyperenhancing areas of LV myocardium on LGE images were manually segmented, and the segmentation was propagated to corresponding areas on cine images. Radiomic features were extracted using the PyRadiomics library. The least absolute shrinkage and selection operator (LASSO) method was employed for radiomic feature selection. Our model development employed the TabPFN algorithm, an adapted Prior-Data Fitted Network design. Model performance was evaluated graphically and numerically over five-repeat fivefold cross-validation. SHapley Additive exPlanations (SHAP) were employed to determine the relative importance of selected radiomic features.

Results: Our cohort consisted of 60 patients with HCM (73.3% male; median age = 51.5 years), among whom 17 had documented VT during the follow-up. A total of 1612 radiomic features were extracted for each patient. The LASSO algorithm led to a final selection of 18 radiomic features. The model achieved a mean area under the receiver operating characteristic curve of 0.877, demonstrating good discrimination, and a mean Brier score of 0.119, demonstrating good calibration.

Conclusion: Radiomics-based ML models are promising for predicting VT in patients with HCM during the follow-up period. Developing predictive models as clinically useful decision-making tools may significantly improve risk assessment and prognosis.