Academic Radiology最新文献

Rationale and objectives: Coronary artery calcification (CAC) can be quantified by computed tomography (CT). It is an important predictive and prognostic imaging marker for cardiovascular disease. The prognostic role for CAC in oncological patients is provided in preliminary studies, especially in lung cancer patients. The aim of the present study was to establish the effect of CAC score on overall survival (OS) in lung cancer patients based on the published literature MATERIALS AND METHODS: Literature databases were screened for papers analyzing the association between CAC and overall survival in lung cancer patients up to June 2024. The primary endpoint of the present systematic review was the OS. Overall, seven studies were suitable for the analysis and were included.

Results: The included studies comprised 2292 patients undergoing curative treatment. The pooled hazard ratio for the association between CAC score and OS was HR= 1.42 (95% CI=(1.19; 1.69), p < 0.0001) in the univariable analysis and HR= 1.56 (95% CI=(1.25; 1.94), p < 0.0001) in the multivariable analysis. The pooled odds ratio for the association between CAC score and major cardiovascular events was OR= 1.97 (95% CI=(1.24; 3.13)], p = 0.004.

Conclusion: CT-defined CAC has a meaningful impact on overall survival and prediction of major cardiovascular events in lung cancer patients undergoing curative treatment. The sole presence of CAC on staging CT should be reported as an important prognostic marker in these patients.

Rationale and objectives: The purpose of this study is the feasibility of using radiomics features, bone morphometry features (BM), and Hounsfield unit (HU) values obtained from routine chest computed tomography (CT) for assessing bone mineral density (BMD) status.

Materials and methods: This retrospective study analyzed 120 patients who underwent routine chest CT and dual-energy X-ray absorptiometry examinations within a month. Whole thoracic vertebral bodies from routine chest CT images were segmented using the GrowCut semi-automatic segmentation method, and radiomics features, BM features, and HU values were extracted. To assess the intra- and inter-observer variability of segmentation, the Dice similarity coefficient (DSC) was utilized. Feature selection was carried out using the intra-class correlation coefficient and the Boruta algorithm. Six machine learning classification models were employed for classification in a three-class manner. The models' performance was evaluated using the area under the receiver operator characteristics curve (AUC). Other evaluation parameters of the models were calculated, including overall accuracy, precision, and sensitivity.

Results: The DSC values showed high similarity by achieving 0.907 ± 0.034 and 0.887 ± 0.048 for intra- and inter-observer segmentation agreement, respectively. After a two-stepwise feature selection, 21 radiomics features were selected. Different combinations of these radiomics features with five BM features and HU values were applied to six classification models for evaluating BMD. The multilayer perceptron (MLP) model based on integration of radiomics features and BM features in a three-class classification approach achieved higher performance compared to other models with an AUC of 0.981 (95% confidence interval (CI): 0.937-0.997) in normal BMD class, an AUC of 0.896 (95% CI: 0.826-0.944) in osteopenia class, and an AUC of 0.927 (95% CI: 0.866-0.967) in osteoporosis class.

Conclusion: Using the MLP classification model based on a combination of radiomics features and BM features in a three-class classification approach can effectively distinguish different BMD conditions.

Rationale and objectives: It is crucial to inform the patient about potential complications and obtain consent before interventional radiology procedures. In this study, we investigated the accuracy, reliability, and readability of the information provided by ChatGPT-4 about potential complications of interventional radiology procedures.

Materials and methods: Potential major and minor complications of 25 different interventional radiology procedures (8 non-vascular, 17 vascular) were asked to ChatGPT-4 chatbot. The responses were evaluated by two experienced interventional radiologists (>25 years and 10 years of experience) using a 5-point Likert scale according to Cardiovascular and Interventional Radiological Society of Europe guidelines. The correlation between the two interventional radiologists' scoring was evaluated by the Wilcoxon signed-rank test, Intraclass Correlation Coefficient (ICC), and Pearson correlation coefficient (PCC). In addition, readability and complexity were quantitatively assessed using the Flesch-Kincaid Grade Level, Flesch Reading Ease scores, and Simple Measure of Gobbledygook (SMOG) index.

Results: Interventional radiologist 1 (IR1) and interventional radiologist 2 (IR2) gave 104 and 109 points, respectively, out of a potential 125 points for the total of all procedures. There was no statistically significant difference between the total scores of the two IRs (p = 0.244). The IRs demonstrated high agreement across all procedure ratings (ICC=0.928). Both IRs scored 34 out of 40 points for the eight non-vascular procedures. 17 vascular procedures received 70 points out of 85 from IR1 and 75 from IR2. The agreement between the two observers' assessments was good, with PCC values of 0.908 and 0.896 for non-vascular and vascular procedures, respectively. Readability levels were overall low. The mean Flesch-Kincaid Grade Level, Flesch Reading Ease scores, and SMOG index were 12.51 ± 1.14 (college level) 30.27 ± 8.38 (college level), and 14.46 ± 0.76 (college level), respectively. There was no statistically significant difference in readability between non-vascular and vascular procedures (p = 0.16).

Conclusion: ChatGPT-4 demonstrated remarkable performance, highlighting its potential to enhance accessibility to information about interventional radiology procedures and support the creation of educational materials for patients. Based on the findings of our study, while ChatGPT provides accurate information and shows no evidence of hallucinations, it is important to emphasize that a high level of education and health literacy are required to fully comprehend its responses.

Background: For treatment of urolithiasis, the stone composition is of particular interest, as uric acid (UA) stones can be treated by chemolitholysis. In this ex vivo study, we employed an advanced composition analysis approach for urolithiasis utilizing spectral data obtained from a photon-counting detector CT (PCDCT) to differentiate UA and non-UA stones. Our primary objective was to assess the accuracy of this analysis method.

Methods: A total of 148 urinary stones with a known composition that was measured by the standard reference method infrared spectroscopy (reference) were placed in an abdomen phantom and scanned in the PCDCT. Our objectives were to assess the stone detection rates of PCDCT and the accuracy of the prediction of the stone composition in UA vs non-UA compared to the reference.

Results: Automated detection recognized 86.5% of all stones, with best detection rate for stones larger > 5 mm in diameter (95.4%, 88.8% for stones larger than 3 mm, 94.7% for stones larger than 4 mm). Depending on the volume, we found a recognition rate of 92.8% for stones larger than 20 mm³ and 94.0% for stones with more than 30 mm³. Prediction of UA composition showed an overall sensitivity and a positive predictive value of 66.7% and a specificity and negative predictive value of 94.5%. Best diagnostic values volume wise were found by only including stones with a larger volume than 30 mm³, there we found a sensitivity of 91.7%, and a specificity of 92.4%. Sensitivity in dependance of the largest diameter was best for stones larger than 5 mm (85.7%), but specificity decreased with increasing diameter (to 91.3%).

Conclusion: Automated urinary stone composition analysis with PCDCT showed a good automated detection rate of 86.5% up to 95.4% depending on stone diameter. The differentiation between non-UA and UA stones is performed with an NPV of 94.5% and a PPV of 66.7%. The prediction probability of non-UA stones was very good. This means the automatic detection and differentiation algorithm can identify the patients which will not profit from chemolitholysis.

Rationale and objectives: Metabolic dysfunction-associated fatty liver disease (MAFLD) is linked to an increased risk of cardiovascular events. Our study sought to determine the impact of MAFLD on both the anatomy and function of coronary plaques.

Materials and methods: A total of 203 participants (including 728 plaques) with suspected coronary artery disease (CAD) who underwent coronary CT angiography (CCTA) and abdominal ultrasound were prospectively enrolled. Participants were divided into MAFLD and non-MAFLD groups. For each plaque, necrotic core plaque volume and fractional flow reserve derived from CT (FFR_CT) were measured. Obstructive CAD, segment involvement score (SIS) >4, high-risk plaque (HRP) and FFR_CT ≤ 0.8 were assessed.

Results: Compared to non-MAFLD, necrotic core plaque volume was higher in MAFLD at both participant level (p < 0.001) and plaque level (p = 0.001). MAFLD had a higher prevalence of obstructive CAD, SIS >4, HRP and FFR_CT ≤ 0.8 at participant level (obstructive CAD: 35.9% vs 21.6%, p = 0.026; SIS >4: 39.7% vs 17.6%, p < 0.001; HRP: 55.1% vs 29.6%, p < 0.001; FFR_CT ≤0.8: 33.3% vs 15.2%, p = 0.002). In addition, MAFLD predicted the presence of obstructive CAD (adjusted OR: 2.44; 95% CI: 1.22-4.87; p = 0.011), SIS >4 (adjusted OR: 3.64; 95% CI: 1.78-7.46; p < 0.001), HRP (adjusted OR: 2.52; 95% CI: 1.37-4.63; p = 0.003) and FFR_CT ≤ 0.8 (adjusted OR: 3.53; 95% CI: 1.65-7.57; p = 0.001) independent of traditional cardiovascular risk factors.

Conclusion: MAFLD is associated with CCTA derived plaque characteristics, including the severity and extent of CAD, HRP, as well as physiologic status, independent of traditional risk factors.

Rationale and objectives: The early prediction of response to neoadjuvant chemotherapy (NAC) will aid in the development of personalized treatments for patients with breast cancer. This study investigated the value of longitudinal multimodal deep learning (DL) based on breast MR and ultrasound (US) in predicting pathological complete response (pCR) after NAC.

Materials and methods: We retrospectively reviewed the pre-NAC and post-2nd-NAC MR and/or US images of 448 patients enrolled from three centers and extracted DL features from the largest section of the breast tumour using ResNet50. T test, Pearson correlation analysis and least absolute shrinkage and selection operator regression were used to select the most significant DL features for the pre-NAC and post-2nd-NAC MR and US DL models. The stacking model integrates different single-modality DL models and meaningful clinical data. The diagnostic performance of the models was evaluated.

Results: In all the patients, the pCR rate was 36.65%. There was no significant difference in diagnostic performance between the different single-modality DL models (DeLong test, p > 0.05). The stacking model integrating the above four DL models with HER2 status yielded areas under the curves of 0.951-0.979, accuracies of 91.55%-92.65%, sensitivities of 90.63%-93.94%, and specificities of 89.47%-94.44% in the cohorts.

Conclusion: Longitudinal multimodal DL can be useful in predicting pCR. The stacking model can be used as a new tool for the early noninvasive prediction of the response to NAC, as evidenced by its excellent performance, and therefore aid the development of personalized treatment strategies for patients with breast cancer.