Academic Radiology最新文献

Rationale and objectives: Triple rule out CT protocols (TRO-CT) have been advocated as a single test to simultaneously evaluate major causes of acute chest pain, in particular acute myocardial infarction (MI), acute pulmonary embolism (PE), and acute aortic syndrome. However, it is unclear what patient populations would benefit from a such comprehensive exam and current guidelines recommend tailoring CT protocols to the most likely diagnosis.

Methods: We retrospectively reviewed TRO-CT scans performed from the Emergency Department (ED) at our institution from April 2021 to April 2022. Charts were reviewed to calculate clinical risk of MI, PE, and acute aortic syndrome using conventional clinical scoring systems (HEART score, PERC score, ADD-RS). TRO-CT findings and 30-day clinical outcomes were recorded from chart review.

Results: 1279 patients ED patients scanned with TRO-CT were included in the analysis. 831 patients (65.0%) were at-risk for two or more clinical risk scores. At TRO-CT, 381 (29.8%) patients had obstructive CAD. 91 (7.1%) had acute PE. 7 (0.5%) had acute aortic syndrome. At 30-day clinical follow up, 28 patients (2.2%) had the diagnosis of acute MI (95% CI: 1.5-3.2%). 90 patients (7.0%) had the diagnosis of acute PE (95% CI: 5.7-8.6%). 7 patients (0.5%) had the diagnosis acute aortic syndrome (95% CI: 0.2-1.2%). A low-risk HEART score was associated with a 0.3% 30-day clinical diagnosis of acute MI (95% CI: 0.0-1.6%). Low-risk-PERC was associated with a 2.9% 30-day clinical diagnosis of acute PE (95% CI: 0.7-8.7%). Low-risk ADD-RS was associated with a 0.3% 30-day clinical diagnosis of acute aortic syndrome (95% CI: 0.0-1.8%).

Conclusions: We found a high clinical overlap in the presentation of acute MI, acute PE, and acute aortic syndrome based on clinical risk scores. Further studies will be needed to compare a TRO-CT algorithm to a standard-of-care algorithm in patients presenting to the ED.

Rationale and objectives: This study aimed to develop a radiomics model characterized by ⁶⁸Ga-fibroblast activation protein inhibitors (FAPI) positron emission tomography (PET) imaging to predict microvascular invasion (MVI) of hepatocellular carcinoma (HCC). This study also investigated the impact of varying thresholds for maximum standardized uptake value (SUV_max) in semi-automatic delineation methods on the predictions of the model.

Methods: This retrospective study included 84 HCC patients who underwent ⁶⁸Ga-FAPI PET and their MVI results were confirmed by histopathological examination. Volumes of interest (VOIs) for lesions were semi-automatically delineated with four thresholds of 30%, 40%, 50%, and 60% for SUV_max. Extracted shape features, first-, second- and higher-order features. Eight PET radiomics models for predicting MVI were constructed and tested.

Results: In the testing set, the logistic regression (LR) model achieved the highest AUC values for three groups of 30%, 50%, and 60%, with values of 0.785, 0.896, and 0.859, respectively, while the random forest (RF) model in 40% group obtained the highest AUC value of 0.815. The LR model in 50% group and the extreme gradient boosting (XGBoost) model in 60% group achieved the highest accuracy, each at 87.5%. The highest sensitivity was observed in the support vector machine (SVM) model in 30% group, at 100%.

Conclusion: The ⁶⁸Ga-FAPI PET radiomics model has high efficacy in predicting MVI in HCC, which is important for the development of HCC treatment plan and post-treatment evaluation. Different thresholds of SUV_max in semi-automatic delineation methods exert a degree of influence on performance of the radiomics model.

Background: As pancreatic cystic neoplasms (PCN) differ in current standard of care, and these treatments can affect quality of life to varying degrees, a definitive preoperative diagnosis must be reliable. Current diagnostic approaches, specifically traditional cross-sectional imaging techniques, face certain limitations. But radiomics has been shown to have high diagnostic accuracy across a range of diseases. Objective to conduct a comprehensive review of the literature on the use of radiomics to differentiate Mucinous Cystic Neoplasm (MCN) from Serous Cystic Neoplasm (SCN).

Methods: This study was comprehensively searched in Pubmed, Scopus and Web of Science databases for meta-analysis of studies that used radiomics to distinguish MCN from SCN. Risk of bias was assessed using the diagnostic accuracy study quality assessment method and combined with sensitivity, specificity, diagnostic odds ratio, and summary receiver operating characteristic (SROC)curve analysis.

Results: A total of 884 patients from 8 studies were included in this analysis, including 365 MCN and 519 SCN. The Meta-analysis found that radiomics identified MCN and SCN with high sensitivity and specificity, with combined sensitivity and specificity of 0.84(0.82-0.87) and 0.82(0.79-0.84). The positive likelihood ratio (PLR) and the negative likelihood ratio (NLR) are 5.61(3.72, 8.47) and 0.14(0.09-0.26). In addition, the area under the SROC curve (AUC) was drawn at 0.93. No significant risk of publication bias was detected through the funnel plot analysis. The performances of feature extraction from the volume of interest (VOI) or Using AI classifier in the radiomics models were superior to those of protocols employing region of interest (ROI) or absence of AI classifier.

Conclusion: This meta-analysis demonstrates that radiomics exhibits high sensitivity and specificity in distinguishing between MCN and SCN, and has the potential to become a reliable diagnostic tool for their identification.

Rationale and objectives: Multi-parametric MRI (mpMRI) studies of the body are routinely acquired in clinical practice. However, a standardized naming convention for MRI protocols and series does not exist currently. Conflicts in the series descriptions present in the DICOM headers arise due to myriad MRI scanners from various manufacturers used for imaging, wide variations in imaging practices across institutions, and technologist preferences. These conflicts affect the hanging protocol, which dictates the arrangement of sequences for the reading radiologist. At present, clinician supervision is necessary to ensure that the correct sequence is being read and used for diagnosis. This pilot work seeks to classify five different series in mpMRI studies acquired at the levels of the chest, abdomen, and pelvis.

Materials and methods: First, 2D and 3D classification networks were compared using data acquired by Siemens scanners and the optimal network was identified. Then, its performance was analyzed when trained with different training data quantities. The out-of-distribution (OOD) robustness on data acquired by a Philips scanner was also measured. In addition, the effect of data augmentation on model training was studied. The model was also tested with smaller input volumes through downsampling or cropping. Finally, the model was trained on combined data from both Siemens and Philips scanners to bridge the performance gap between different scanners.

Results: Among 2D and 3D networks of ResNet-50, ResNet-101, DenseNet- 121, and EfficientNet-BN0, the 3D DenseNet-121 ensemble achieved an F₁ score of 99.5% when tested on data from the Siemens scanners. The model performed well on OOD data from the Philips scanner and achieved an F₁ score of 86.5%. There was no statistically significant difference between the models trained with and without data augmentation, and between the models trained with original-sized input and with smaller-sized input. When training the model with combined data, the F₁ score improved to 98.8% for the Philips test set and 99.3% for the Siemens test set respectively.

Conclusion: Our pilot work is useful for the classification of MRI sequences in studies acquired at the level of the chest, abdomen, and pelvis. It has the potential for robust automation of hanging protocols and the creation of large-scale data cohorts for pre-clinical research.

Rationale and objectives: This study aimed to identify the factors associated with malignant biopsy results for new lesions within one year after breast cancer surgery.

Materials and methods: This retrospective study included 192 lesions from 186 patients who underwent biopsy for newly developed breast lesions within one year of breast cancer surgery. All patients underwent breast ultrasound (US) at 6 months and breast US with mammography one year after surgery. We analyzed the biopsy results, patient age, characteristics of previous cancers (histologic type, stage, molecular subtype, histologic and nuclear grade, Ki-67 index, extensive intraductal component, lymphovascular invasion (LVI)), history of neoadjuvant chemotherapy (NAC), adjuvant therapy, and characteristics of biopsied lesions (location, mode of detection, imaging features, and Breast Imaging Reporting and Data System category). Multivariate logistic regression was performed to predict malignant results after a biopsy of the new lesion in the early postoperative period.

Results: The mean patient age was 49.0 (range, 28-82) years. During follow-up, 137 lesions developed in the ipsilateral remnant breast or mastectomy bed, and 55 lesions developed in the contralateral breast. In total, 37 (19.3%) of the biopsied lesions were malignant, and the following conditions were associated with malignant results in the newly detected lesions: irregularly shaped hypoechoic mass with increased vascularity, presence of previous LVI, history of NAC, and no history of adjuvant radiotherapy or hormone therapy in the indicated patients.

Conclusion: Active biopsy may be warranted for new lesions with suspicious imaging findings in the breast or operation bed of patients with LVI, a history of NAC, and no history of adjuvant radiotherapy or hormone therapy, even within one year of breast cancer surgery.

Rationale and objectives: New vertebral compression fractures (NVCF) are very common in patients following percutaneous vertebroplasty (PVP) or kyphoplasty (PKP). The study aims to evaluate the comparative predictive performance of vertebral bone quality (VBQ) score and Hounsfield units (HU) value in forecasting NVCF after surgery.

Materials and methods: This study retrospectively analyzed patients who underwent PVP/PKP at our institution between 2020 and 2021. The VBQ score and HU value were obtained from preoperative magnetic resonance imaging (MRI) and computed tomography (CT) scans, respectively. Subsequently, the forecasting capabilities of these two parameters were assessed by contrasting their receiver operating characteristic (ROC) curve.

Results: A total of 303 eligible patients (56 with NVCF and 247 without) were identified in the study. Six relevant literature factors were identified and included in the multivariate analysis revealed that lower HU value (OR = 0.967, 95% CI = 0.953-0.981, P < 0.001) and higher VBQ score (OR = 3.964, 95% CI = 2.369-6.631, P < 0.001) emerged as independent predictors of NCVF occurrence. Compared to the ROC curve of the HU value, demonstrating a diagnostic accuracy of 83.2% (95% CI = 77.5%-88.9%, P < 0.001), the VBQ score was 85.8%. And, a statistically significant negative correlation was observed between the VBQ score and the T-score (r = -0.529, P < 0.001).

Conclusion: In patients undergoing PVP/PKP, VBQ score, and HU value are independently associated with the occurrence of NVCF. Assessing the HU value and the VBQ score could play an effective role in planning PVP/PKP operations.

Rationale and objectives: To evaluate the performance of virtual MR elastography (vMRE) for predicting microvascular invasion (MVI) in Barcelona Clinic Liver Cancer (BCLC) stage A (≤ 5.0 cm) hepatocellular carcinoma (HCC) and to construct a combined nomogram based on vMRE, multi-b-value DWI models, and clinical-radiological (CR) features.

Methods: Consecutive patients with suspected HCC who underwent multi-b-value DWI examinations were prospectively collected. Quantitative parameters from vMRE, mono-exponential, intravoxel incoherent motion, and diffusion kurtosis imaging models were obtained. Multivariate logistic regression was used to identify independent MVI predictors and build prediction models. A combined MRI_Score was constructed using independent quantitative parameters. A visualized nomogram was built based on significant CR features and MRI_Score. The predictive performance of quantitative parameters and models was evaluated.

Results: The study included 103 patients (median age: 56 years; range: 35-70 years; 87 males and 16 females). Diffusion-based shear modulus (μ_Diff) exhibited a predictive performance for MVI with area under the curve (AUC) of 0.735. The MRI_Score was developed employing true diffusion coefficient (D), mean kurtosis (MK), and μ_Diff. CR model and MRI_Score achieved AUCs of 0.787 and 0.840, respectively. The combined nomogram based on AFP, corona enhancement, tumor capsule, TTPVI, and MRI_Score significantly improved the predictive performance to an AUC of 0.931 (Delong test p < 0.05).

Conclusion: vMRE exhibited great potential for predicting MVI in BCLC stage A HCC. The combined nomogram integrating CR features, vMRE, and quantitative diffusion parameters significantly improved the predictive accuracy and could potentially assist clinicians in identifying appropriate treatment options.