Contrast-enhanced Computed Tomography Imaging Research Articles

Deep Learning-Based Prediction of Post-treatment Survival in Hepatocellular Carcinoma Patients Using Pre-treatment CT Images and Clinical Data.

The objective of this study was to develop and evaluate a model for predicting post-treatment survival in hepatocellular carcinoma (HCC) patients using their CT images and clinical information, including various treatment information. We collected pre-treatment contrast-enhanced CT images and clinical information including patient-related factors, initial treatment options, and survival status from 692 patients. The patient cohort was divided into a training cohort (n = 507), a testing cohort (n = 146), and an external CT cohort (n = 39), which included patients who underwent CT scans at other institutions. After model training using fivefold cross-validation, model validation was performed on both the testing cohort and the external CT cohort. Our cascaded model employed a 3D convolutional neural network (CNN) to extract features from CT images and derive final survival probabilities. These probabilities were obtained by concatenating previously predicted probabilities for each interval with the patient-related factors and treatment options. We utilized two consecutive fully connected layers for this process, resulting in a number of final outputs corresponding to the number of time intervals, with values representing conditional survival probabilities for each interval. Performance was assessed using the concordance index (C-index), the mean cumulative/dynamic area under the receiver operating characteristics curve (mC/D AUC), and the mean Brier score (mBS), calculated every 3months. Through an ablation study, we found that using DenseNet-121 as the backbone network and setting the prediction interval to 6months optimized the model's performance. The integration of multimodal data resulted in superior predictive capabilities compared to models using only CT images or clinical information (C index 0.824 [95% CI 0.822-0.826], mC/D AUC 0.893 [95% CI 0.891-0.895], and mBS 0.121 [95% CI 0.120-0.123] for internal test cohort; C index 0.750 [95% CI 0.747-0.753], mC/D AUC 0.819 [95% CI 0.816-0.823], and mBS 0.159 [95% CI 0.158-0.161] for external CT cohort, respectively). Our CNN-based discrete-time survival prediction model with CT images and clinical information demonstrated promising results in predicting post-treatment survival of patients with HCC.

Utility of radiomic features in predicting clinical outcomes in stage II-III pancreatic cancer.

74 Background: We identified computed tomography (CT)-derived radiomic features predictive of tumor progression within three months, then examined their ability to prognosticate overall survival (OS) along with clinical features in pancreatic cancer. We evaluated these features in patients with unresected pancreatic cancer who underwent stereotactic body radiation therapy (SBRT) in sequence with chemotherapy, but not surgery. Methods: In this retrospective study, we examined a cohort of 101 patients with stage II-III pancreatic cancer who underwent SBRT with sequential chemotherapy at a single institution (Stanford Health Care) between 1999-2020. From their pre-SBRT contrast-enhanced CT images with segmented tumors, delineating regions-of-interest, we extracted 900 radiomic (quantitative pixel-level imaging characteristic) features. In the first phase, we identified radiomic features that predicted rapid tumor progression within three months following SBRT. We divided the dataset into a training set (n = 53) for model development and a test set (n = 48) for evaluation. Using logistic regression with the Least Absolute Shrinkage and Selection Operator algorithm for feature selection and classification, we built a binary prediction model on the training set to identify patients at risk of progression within three months of SBRT. To fine-tune parameters, we performed five-fold cross-validation (CV) on the training set, repeating each set of parameters five times. We assessed model performance on the test set using the area under the curve (AUC). We selected the model with the best AUC, generating the predictive radiomic feature set. In the second phase, we conducted univariate and multivariate Cox regression analyses to assess the relationship between OS and individual clinical variables (age, sex, stage, vessel involvement, tumor location, performance status, body mass index, biological equivalent dose of radiation) and the radiomic feature set as high versus low risk. Results: Our cohort consisted of 48 men (mean age, 70 years ± 11 [SD]) and 53 women (mean age, 67 years ± 13 [SD]). From the first phase, 32 textural features comprised the radiomic feature set that best predicted rapid tumor progression, with mean AUCs of 0.852 (CV, n=53) and 0.814 (test, n=48). In the univariate Cox model, only the radiomic feature set was predictive of OS (hazard ratio, HR, 1.724, p=0.011). In the multivariate Cox model, radiomic features and age were significant predictors of OS, with HR of 1.819 (p=0.007) and 1.024 (p=0.024), respectively. Conclusions: CT-derived radiomic features predict rapid tumor progression following SBRT, confer nearly a twofold increase in mortality risk, and, along with patient age, enhance the identification of patients with stage II-III pancreatic cancer with poor OS. [Table: see text]

Radiological assessment of the dissection area in supraomohyoid neck dissection

PurposeThe current supraomohyoid neck dissection (SOHND) is performed above the omohyoid muscle to dissect levels I, II, and III in the levels of cervical lymph nodes. However, the anatomical boundary between levels III and IV is the inferior border of the cricoid cartilage. We investigated the anatomical relationship between the omohyoid muscle and cricoid cartilage using contrast-enhanced CT (CE-CT) images to assess the validity of the current SOHND.MethodsCE-CT images of the head and neck regions in patients were reviewed. The patients were divided into two groups: “malignant tumors” and “others”. The vertebral levels corresponding to the positions of anatomical structures such as the intersection of the omohyoid muscle and internal jugular vein (OM-IJ), and the inferior border of the cricoid cartilage (CC), were recorded.ResultsThe OM-IJ was located around the seventh cervical to the first thoracic vertebra. There was a significant difference between the malignant tumor and others groups in females (p = 0.036). The CC was located around the sixth to seventh cervical vertebrae. There was a significant sex difference in each group (malignant tumor: p < 0.0001; others: p = 0.008). Both sexes tended to have lower OM-IJ than CC, and females had significantly lower OM-IJ than males.ConclusionThis study provides clear anatomical evidence showing the difference between the SOHND dissection area and levels I, II, and III. It could be considered that in most cases SOHND invades level IV, not just levels I, II, and III, especially in female patients.

Textural heterogeneity of liver lesions in CT imaging - comparison of colorectal and pancreatic metastases

PurposeTumoral heterogeneity poses a challenge for personalized cancer treatments. Especially in metastasized cancer, it remains a major limitation for successful targeted therapy, often leading to drug resistance due to tumoral escape mechanisms. This work explores a non-invasive radiomics-based approach to capture textural heterogeneity in liver lesions and compare it between colorectal cancer (CRC) and pancreatic cancer (PDAC).Materials and methodsIn this retrospective single-center study 73 subjects (42 CRC, 31 PDAC) with 1291 liver metastases (430 CRC, 861 PDAC) were segmented fully automated on contrast-enhanced CT images by a UNet for medical images. Radiomics features were extracted using the Python package Pyradiomics. The mean coefficient of variation (CV) was calculated patient-wise for each feature to quantify the heterogeneity. An unpaired t-test identified features with significant differences in feature variability between CRC and PDAC metastases.ResultsIn both colorectal and pancreatic liver metastases, interlesional heterogeneity in imaging can be observed using quantitative imaging features. 75 second-order features were extracted to compare the varying textural characteristics. In total, 18 radiomics features showed a significant difference (p < 0.05) in their expression between the two malignancies. Out of these, 16 features showed higher levels of variability within the cohort of pancreatic metastases, which, as illustrated in a radar plot, suggests greater textural heterogeneity for this entity.ConclusionsRadiomics has the potential to identify the interlesional heterogeneity of CT texture among individual liver metastases. In this proof-of-concept study for the quantification and comparison of imaging-related heterogeneity in liver metastases a variation in the extent of heterogeneity levels in CRC and PDAC liver metastases was shown.

Anatomy-constrained synthesis for spleen segmentation improvement in unpaired mouse micro-CT scans with 3D CycleGAN

Objective. Auto-segmentation in mouse micro-CT enhances the efficiency and consistency of preclinical experiments but often struggles with low-native-contrast and morphologically complex organs, such as the spleen, resulting in poor segmentation performance. While CT contrast agents can improve organ conspicuity, their use complicates experimental protocols and reduces feasibility. We developed a 3D Cycle Generative Adversarial Network (CycleGAN) incorporating anatomy-constrained U-Net models to leverage contrast-enhanced CT (CECT) insights to improve unenhanced native CT (NACT) segmentation. Approach. We employed a standard CycleGAN with an anatomical loss function to synthesize virtual CECT images from unpaired NACT scans at two different resolutions. Prior to training, two U-Nets were trained to automatically segment six major organs in NACT and CECT datasets, respectively. These pretrained 3D U-Nets were integrated during the CycleGAN training, segmenting synthetic images, and comparing them against ground truth annotations. The compound loss within the CycleGAN maintained anatomical fidelity. Full image processing was achieved for low-resolution datasets, while high-resolution datasets employed a patch-based method due to GPU memory constraints. Automated segmentation was applied to original NACT and synthetic CECT scans to evaluate CycleGAN performance using the Dice Similarity Coefficient (DSC) and the 95th percentile Hausdorff Distance (HD95p). Main results. High-resolution scans showed improved auto-segmentation, with an average DSC increase from 0.728 to 0.773 and a reduced HD95p from 1.19 mm to 0.94 mm. Low-resolution scans benefited more from synthetic contrast, showing a DSC increase from 0.586 to 0.682 and an HD95p reduction from 3.46 mm to 1.24 mm. Significance. Implementing CycleGAN to synthesize CECT scans substantially improved the visibility of the mouse spleen, leading to more precise auto-segmentation. This approach shows the potential in preclinical imaging studies where contrast agent use is impractical.

Exploring bone density analysis on routine CT scans as a tool for opportunistic osteoporosis screening

The primary aim of this study was to evaluate computed tomography (CT)-based bone density analysis at the level of thoracic vertebra 12 (Th12) as a screening method for decreased bone density in patients admitted to the intensive care unit (ICU). Interobserver variability was analyzed. Secondary aims were to assess the prevalence of CT-based low bone density upon ICU admission in a cohort of COVID-19 patients and to assess the potential effect of long-term ICU stay on bone density in these patients. Retrospective single-center cohort study. ICU of the Leiden University Medical Center (LUMC), the Netherlands. Patients admitted to the ICU of the LUMC between March 1st, 2020 and February 1st, 2022 with a diagnosis of COVID-19, and a length of ICU stay of ≥ 21 days. In the included patients both baseline chest CT scans (obtained upon ICU admission) and follow-up chest CT scans (obtained ≥ 21 days after ICU admission) were available for analysis. A total of 118 CT scans in 38 patients were analyzed. There was a good interobserver variability, with an overall mean absolute difference (between measurements of three observers) of 9.7 Hounsfield Units (HU) and an intraclass correlation coefficient (ICC) of 0.93 (95% CI 0.88–0.96). The effect of intravenous contrast administration on bone density measurements was small (+ 7.5 HU (95% CI 3.4–11.5 HU)) higher in contrast enhanced CT images compared to non contrast enhanced CT images). Thirty-seven percent of patients had a bone density < 140 HU, suggestive of osteoporosis. No significant difference was found between bone density upon ICU admission and bone density at follow-up (≥ 21 days after ICU admission). Vertebral CT-based bone density analysis using routine CT scans is an easily applicable method to identify ICU patients with decreased bone density, which could enable enrollment in osteoporosis prevention programs. A high prevalence of low bone density was found in our cohort of ICU patients. There were no changes observed in bone density between baseline and follow-up measurements.

Extrahepatic Portal Venous Gas Is the Strongest Predictor of Mortality in Patients with Portal Venous Gas and Pneumatosis Intestinalis.

Very few studies have examined the association between contrast-enhanced computed tomography (CT) findings observed in portal venous gas (PVG) and pneumatosis intestinalis (PI) and the underlying diseases in these conditions. In this study, we analyzed this association and report the findings for predicting mortality. Overall, 50 patients diagnosed with PVG or PI, observed on contrast-enhanced CT, underwent treatment at our hospital. Based on the underlying disease, we divided the patients into three groups, those with ischemic disease, infectious disease, or gastrointestinal dilatation. Furthermore, cases that underwent surgical treatment or needed surgery but were inoperable were assigned to the high risk group (n=16) and patients who received conservative treatment were assigned to the low risk group (n=34). We reviewed the patients' medical charts, laboratory data, and CT images retrospectively, and analyzed the relationship between CT findings, underlying disease, and association with the high risk or low risk group in each case. Poor enhancement of the intestinal wall, mesenteric fat stranding, extrahepatic PVG, advanced age, and renal disease were significantly associated with ischemic disease (p=0.02, p=0.02, p=0.005, p=0.008 and p=0.049, respectively). PI alone was strongly associated with gastrointestinal dilatation (p=0.009). Patients in the low risk group had more favorable outcomes with conservative treatment. In multivariate analysis, extrahepatic PVG was the only factor associated with the high risk group (p=0.002). Extrahepatic PVG associated with ischemic disease was the strongest predictive factor of mortality. Other CT findings, though useful in diagnosing the underlying disease, were not significant predictive factors.

Bone metastasis prediction in non-small-cell lung cancer: primary CT-based radiomics signature and clinical feature

BackgroundRadiomics provided opportunities to quantify the tumor phenotype non-invasively. This study extracted contrast-enhanced computed tomography (CECT) radiomic signatures and evaluated clinical features of bone metastasis in non-small-cell lung cancer (NSCLC). With the combination of the revealed radiomics and clinical features, the predictive modeling on bone metastasis in NSCLC was established.MethodsA total of 318 patients with NSCLC at the Tianjin Medical University Cancer Institute & Hospital was enrolled between January 2009 and December 2019, which included a feature-learning cohort (n = 223) and a validation cohort (n = 95). We trained a radiomics model in 318 CECT images from feature-learning cohort to extract the radiomics features of bone metastasis in NSCLC. The Kruskal-Wallis and the least absolute shrinkage and selection operator regression (LASSO) were used to select bone metastasis-related features and construct the CT radiomics score (Rad-score). Multivariate logistic regression was performed with the combination of the Rad-score and clinical data. A predictive nomogram was subsequently developed.ResultsRadiomics models using CECT scans were significant on bone metastasis prediction in NSCLC. Model performance was enhanced with each information into the model. The radiomics nomogram achieved an AUC of 0.745 (95% confidence interval [CI]: 0.68,0.80) on predicting bone metastasis in the training set and an AUC of 0.808(95% confidence interval [CI]: 0.71,0.88) in the validation set.ConclusionThe revealed invisible image features were of significance on guiding bone metastasis prediction in NSCLC. Based on the combination of the image features and clinical characteristics, the predictive nomogram was established. Such nomogram can be used for the auxiliary screening of bone metastasis in NSCLC.

Translating contrast enhanced computed tomography images to liver radioembolization dose distribution for more comprehensively indicating patients

Objective. Contrast-enhanced computed tomography (CECT) is commonly used in the pre-treatment evaluation of liver Y-90 radioembolization feasibility. CECT provides detailed imaging of the liver and surrounding structures, allowing healthcare providers to assess the size, location, and characteristics of liver tumors prior to the treatment. Here we propose a method for translating CECT images to an expected dose distribution for tumor(s) and normal liver tissue. Approach. A pre-procedure CECT is used to obtain an iodine arterial-phase distribution by subtracting the non-contrast CT from the late arterial phase. The liver segments surrounding the targeted tumor are selected using Couinaud’s method. The resolution of the resulting images is then degraded to match the resolution of the positron emission tomography (PET) images, which can image the Y-90 activity distribution post-treatment. The resulting images are then used in the same way as PET images to compute doses using the local deposition method. CECT images from three patients were used to test this method retrospectively and were compared with Y-90 PET-based dose distributions through dose volume histograms. Main results. Results show a concordance between predicted and delivered Y-90 dose distributions with less than 10% difference in terms of mean dose, for doses greater than 10% of the 98th percentile (D2%). Significance. CECT-derived predictions of Y-90 radioembolization dose distributions seem promising as a supplementary tool for physicians when assessing treatment feasibility. This dosimetry prediction method could provide a more comprehensive pre-treatment evaluation—offering greater insights than a basic assessment of tumor opacification on CT images.

Segmentation Guided Crossing Dual Decoding Generative Adversarial Network for Synthesizing Contrast-Enhanced Computed Tomography Images.

Although contrast-enhanced computed tomography (CE-CT) images significantly improve the accuracy of diagnosing focal liver lesions (FLLs), the administration of contrast agents imposes a considerable physical burden on patients. The utilization of generative models to synthesize CE-CT images from non-contrasted CT images offers a promising solution. However, existing image synthesis models tend to overlook the importance of critical regions, inevitably reducing their effectiveness in downstream tasks. To overcome this challenge, we propose an innovative CE-CT image synthesis model called Segmentation Guided Crossing Dual Decoding Generative Adversarial Network (SGCDD-GAN). Specifically, the SGCDD-GAN involves a crossing dual decoding generator including an attention decoder and an improved transformation decoder. The attention decoder is designed to highlight some critical regions within the abdominal cavity, while the improved transformation decoder is responsible for synthesizing CE-CT images. These two decoders are interconnected using a crossing technique to enhance each other's capabilities. Furthermore, we employ a multi-task learning strategy to guide the generator to focus more on the lesion area. To evaluate the performance of proposed SGCDD-GAN, we test it on an in-house CE-CT dataset. In both CE-CT image synthesis tasks-namely, synthesizing ART images and synthesizing PV images-the proposed SGCDD-GAN demonstrates superior performance metrics across the entire image and liver region, including SSIM, PSNR, MSE, and PCC scores. Furthermore, CE-CT images synthetized from our SGCDD-GAN achieve remarkable accuracy rates of 82.68%, 94.11%, and 94.11% in a deep learning-based FLLs classification task, along with a pilot assessment conducted by two radiologists.

Clinical feasibility of deep learning based synthetic contrast enhanced abdominal CT in patients undergoing non enhanced CT scans

Our objective was to develop and evaluate the clinical feasibility of deep-learning-based synthetic contrast-enhanced computed tomography (DL-SynCCT) in patients designated for nonenhanced CT (NECT). We proposed a weakly supervised learning with the utilization of virtual non-contrast CT (VNC) for the development of DL-SynCCT. Training and internal validations were performed with 2202 pairs of retrospectively collected contrast-enhanced CT (CECT) images with the corresponding VNC images acquired from dual-energy CT. Clinical validation was performed using an external validation set including 398 patients designated for true nonenhanced CT (NECT), from multiple vendors at three institutes. Detection of lesions was performed by three radiologists with only NECT in the first session and an additionally provided DL-SynCCT in the second session. The mean peak signal-to-noise ratio (PSNR) and structural similarity index map (SSIM) of the DL-SynCCT compared to CECT were 43.25 ± 0.41 and 0.92 ± 0.01, respectively. With DL-SynCCT, the pooled sensitivity for lesion detection (72.0% to 76.4%, P < 0.001) and level of diagnostic confidence (3.0 to 3.6, P < 0.001) significantly increased. In conclusion, DL-SynCCT generated by weakly supervised learning showed significant benefit in terms of sensitivity in detecting abnormal findings when added to NECT in patients designated for nonenhanced CT scans.

Comprehensive analysis of clinical data and radiomic features from contrast enhanced CT for differentiating benign and malignant pancreatic intraductal papillary mucinous neoplasms

The primary aim of this investigation was to leverage radiomics features derived from contrast-enhanced abdominal computed tomography (CT) scans to devise a predictive model to discern the benign and malignant nature of intraductal papillary mucinous neoplasms (IPMNs). Radiomic signatures were meticulously crafted to delineate benign from malignant IPMNs by extracting pertinent features from contrast-enhanced CT images within a designated training cohort (n = 84). Subsequent validation was conducted with data from an independent test cohort (n = 37). The discriminative ability of the model was quantitatively evaluated through receiver operating characteristic (ROC) curve analysis, with the integration of carefully selected clinical features to improve the comparative analysis. Arterial-phase images were utilized to construct a model comprising 8 features for distinguishing between benign and malignant cases. The model achieved an accuracy of 0.891 [95% confidence interval (95% CI), 0.816–0.996] in the cross-validation set and 0.553 (95% CI 0.360–0.745) in the test set. Conversely, employing 9 features from the venous-phase resulted in a model with a cross-validation accuracy of 0.862 (95%CI 0.777–0.946) and a test set accuracy of 0.801 (95% CI 0.653–0.950).Integrating the identified clinical features with imaging features yielded a model with a cross-validation accuracy of 0.934 (95% CI 0.879–0.990) and a test set accuracy of 0.904 (95% CI 0.808–0.999), thereby further improving its discriminatory ability. Our findings distinctly illustrate that venous-phase radiomics features eclipse arterial-phase radiomic features in terms of predictive accuracy regarding the nature of IPMNs. Furthermore, the synthesis and meticulous screening of clinical features with radiomic data significantly increased the diagnostic efficacy of our model, underscoring the pivotal importance of a comprehensive and integrated approach for accurate risk stratification in IPMN management.

Prediction of clear cell renal cell carcinoma ≤ 4cm: visual assessment of ultrasound characteristics versus ultrasonographic radiomics analysis.

To investigate the diagnostic efficacy of the clinical ultrasound imaging model, ultrasonographic radiomics model, and comprehensive model based on ultrasonographic radiomics for the differentiation of small clear cell Renal Cell Carcinoma (ccRCC) and Renal Angiomyolipoma (RAML). The clinical, ultrasound, and contrast-enhanced CT(CECT) imaging data of 302 small renal tumors (maximum diameter ≤ 4cm) patients in Tianjin Medical University Cancer Institute and Hospital from June 2018 to June 2022 were retrospectively analyzed, with 182 patients of ccRCC and 120 patients of RAML. The ultrasound images of the largest diameter of renal tumors were manually segmented by ITK-SNAP software, and Pyradiomics (v3.0.1) module in Python 3.8.7 was applied to extract ultrasonographic radiomics features from ROI segmented images. The patients were randomly divided into training and internal validation cohorts in the ratio of 7:3. The Random Forest algorithm of the Sklearn module was applied to construct the clinical ultrasound imaging model, ultrasonographic radiomics model, and comprehensive model. The efficacy of the prediction models was verified in an independent external validation cohort consisting of 69 patients, from 230 small renal tumor patients in two different institutions. The Delong test compared the predictive ability of three models and CECT. Calibration Curve and clinical Decision Curve Analysis were applied to evaluate the model and determine the net benefit to patients. 491 ultrasonographic radiomics features were extracted from 302 small renal tumor patients, and 9 ultrasonographic radiomics features were finally retained for modeling after regression and dimensionality reduction. In the internal validation cohort, the area under the curve (AUC), sensitivity, specificity, and accuracy of the clinical ultrasound imaging model, ultrasonographic radiomics model, comprehensive model, and CECT were 0.75, 76.7%, 60.0%, 70.0%; 0.80, 85.6%, 61.7%, 76.0%; 0.88, 90.6%, 76.7%, 85.0% and 0.90, 92.6%, 88.9%, 91.1%, respectively. In the external validation cohort, AUC, sensitivity, specificity, and accuracy of the three models and CECT were 0.73, 67.5%, 69.1%, 68.3%; 0.89, 86.7%, 80.0%, 83.5%; 0.90, 85.0%, 85.5%, 85.2% and 0.91, 94.6%, 88.3%, 91.3%, respectively. The DeLong test showed no significant difference between the clinical ultrasound imaging model and the ultrasonographic radiomics model (Z=-1.287, P=0.198). The comprehensive model showed superior diagnostic performance than the ultrasonographic radiomics model (Z=4. 394, P<0.001) and the clinical ultrasound imaging model (Z=4. 732, P<0.001). Moreover, there was no significant difference in AUC between the comprehensive model and CECT (Z=-0.252, P=0.801). Both in the internal and external validation cohort, the Calibration Curve and Decision Curve Analysis showed a better performance of the comprehensive model. It is feasible to construct an ultrasonographic radiomics model for distinguishing small ccRCC and RAML based on ultrasound images, and the diagnostic performance of the comprehensive model is superior to the clinical ultrasound imaging model and ultrasonographic radiomics model, similar to that of CECT.

Differentiation of granulomatous nodules with lobulation and spiculation signs from solid lung adenocarcinomas using a CT deep learning model

BackgroundThe diagnosis of solitary pulmonary nodules has always been a difficult and important point in clinical research, especially granulomatous nodules (GNs) with lobulation and spiculation signs, which are easily misdiagnosed as malignant tumors. Therefore, in this study, we utilised a CT deep learning (DL) model to distinguish GNs with lobulation and spiculation signs from solid lung adenocarcinomas (LADCs), to improve the diagnostic accuracy of preoperative diagnosis.Methods420 patients with pathologically confirmed GNs and LADCs from three medical institutions were retrospectively enrolled. The regions of interest in non-enhanced CT (NECT) and venous contrast-enhanced CT (VECT) were identified and labeled, and self-supervised labels were constructed. Cases from institution 1 were randomly divided into a training set (TS) and an internal validation set (IVS), and cases from institutions 2 and 3 were treated as an external validation set (EVS). Training and validation were performed using self-supervised transfer learning, and the results were compared with the radiologists’ diagnoses.ResultsThe DL model achieved good performance in distinguishing GNs and LADCs, with area under curve (AUC) values of 0.917, 0.876, and 0.896 in the IVS and 0.889, 0.879, and 0.881 in the EVS for NECT, VECT, and non-enhanced with venous contrast-enhanced CT (NEVECT) images, respectively. The AUCs of radiologists 1, 2, 3, and 4 were, respectively, 0.739, 0.783, 0.883, and 0.901 in the (IVS) and 0.760, 0.760, 0.841, and 0.844 in the EVS.ConclusionsA CT DL model showed great value for preoperative differentiation of GNs with lobulation and spiculation signs from solid LADCs, and its predictive performance was higher than that of radiologists.

Amount of ascites impacts survival in patients with hepatocellular carcinoma undergoing transarterial chemoembolization advocating for volumetric assessment

Preliminary work has shown that portal hypertension plays a key role for the prognosis in patients with hepatocellular carcinoma (HCC) undergoing transarterial chemoembolization (TACE). Specifically, the presence of ascites appears to be a strong negative predictor for these patients. However, it remains unclear whether different ascites volumes influence prognosis. Therefore, the aim of this work was to investigate the influence of different ascites volumes on survival for patients with HCC undergoing TACE. A total of 327 treatment-naïve patients with HCC undergoing initial TACE at our tertiary care center between 2010 and 2020 were included. In patients with ascites, the fluid was segmented, and the volume quantified by slice-wise addition using contrast-enhanced CT imaging. Median overall survival (OS) was calculated and univariate and multivariate Cox regression analysis has been performed. Ascites was present in 102 (31.9%) patients. Ascites volume as continuous variable was significantly associated with an increased hazard ratio in univariate analysis (p < 0.001) and remained an independent predictor of impaired median OS in multivariate analysis (p < 0.001). Median OS without ascites was 17.1 months, and therefore significantly longer than in patients with ascites (6.4 months, p < 0.001). When subdivided into groups of low and high ascites volume in relation to the median ascites volume, patients with low ascites volume had a significantly longer median OS (8.6 vs 3.6 months, p < 0.001). Ascites in patients with HCC undergoing TACE is strongly associated with a poor prognosis. Our results show that not only the presence but also the amount of ascites is highly relevant. Therefore, true ascites volume as opportunistic quantitative biomarker is likely to impact clinical decision-making once automated solutions become available.

Imaging Features of Uncommon Entities That Manifest with Torsion.

Torsion is the twisting of an object along the axis, and various structures (organs and tumors) in the body can twist. Torsion causes initial lymphatic and venous outflow obstruction, leading to congestive edema, enlargement, venous hemorrhagic infarction, and surrounding edema. It can also cause subsequent arterial obstruction depending on the degree of torsion, leading to ischemia, infarction, necrosis, gangrene, and surrounding inflammation. Therefore, in several cases of torsion, immediate surgical intervention is required to improve blood flow and prevent serious complications. Clinical manifestations of torsion are often nonspecific and can affect individuals of varying ages and sex. Imaging plays an important role in the early diagnosis and management of torsion. Multiple imaging modalities, including US, radiography, CT, and MRI, are used to evaluate torsion, and each modality has its specific characteristics. The imaging findings reflect the pathophysiologic mechanism: a twisted pedicle (whirlpool sign), enlargement of the torsed structures, reduced blood flow, internal heterogeneity, and surrounding reactive changes. The whirlpool sign is a definitive characteristic of torsion. In some cases, despite poor internal enhancement, capsular enhancement is observed on contrast-enhanced CT and MR images and is considered to be associated with preserved capsular arterial flow or capsular neovascularization due to inflammation. Radiologists should be familiar with the pathophysiologic mechanisms, clinical characteristics, and imaging characteristics of torsion in various structures in the body. Since other articles about common organ torsions already exist, the authors of this article focus on the uncommon entities that manifest with torsion. ©RSNA, 2024.

A contrast-enhanced computed tomography-based radiomics nomogram for preoperative differentiation between benign and malignant cystic renal lesions.

There is lack of discrimination as to traditional imaging diagnostic methods of cystic renal lesions (CRLs). This study aimed to evaluate the value of machine learning models based on clinical data and contrast-enhanced computed tomography (CECT) radiomics features in the differential diagnosis of benign and malignant CRL. There were 192 patients with CRL (Bosniak class ≥ II) enrolled through histopathological examination, including 144 benign cystic renal lesions (BCRLs) and 48 malignant cystic renal lesions (MCRLs). Radiomics features were extracted from CECT images taken during the medullary phase. Using the light gradient boosting machine (LightGBM) algorithm, the clinical, radiomics and combined models were constructed. A comprehensive nomogram was developed by integrating the radiomics score (Rad-score) with independent clinical factors. Receiver operating characteristic (ROC) curves were plotted. The corresponding area under the curve (AUC) value was worked out to quantify the discrimination performance of the three models in training and validation cohorts. Calibration curves were worked out to assess the accuracy of the probability values predicted by the models. Decision curve analysis (DCA) was worked out to assess the performance of models at different thresholds. Maximum diameter and Bosniak class were independent risk factors of patients with MCRL in the clinical model. Twenty-one radiomics features were extracted to work out a Rad-score. The performance of the clinical model in the training cohort was AUC =0.948, 95% confidence interval (CI): 0.917-0.980, and the performance in the validation cohort was AUC =0.936, 95% CI: 0.859-1.000 (P<0.05). The performance of the radiomics model in the training cohort was AUC =0.990, 95% CI: 0.979-1.000, and the performance in the validation cohort was AUC =0.959, 95% CI: 0.903-1.000 (P<0.05). Compared with the above models, the combined radiomics nomogram had an AUC of 0.989 (95% CI: 0.977-1.000) in the training cohort and an AUC of 0.962 (95% CI: 0.905-1.000) in the validation cohort (P<0.05), showing the best diagnostic efficacy. The radiomics nomogram integrating clinical independent risk factors and radiomics signature improved the diagnostic accuracy in differentiating between BCRL and MCRL, which can provide a reference for clinical decision-making and help clinicians develop individualized treatment strategies for patients.

Fully automated assessment of the future liver remnant in a blood-free setting via CT before major hepatectomy via deep learning

ObjectivesTo develop and validate a deep learning (DL) model for automated segmentation of hepatic and portal veins, and apply the model in blood-free future liver remnant (FLR) assessments via CT before major hepatectomy.Methods3-dimensional 3D U-Net models were developed for the automatic segmentation of hepatic veins and portal veins on contrast-enhanced CT images. A total of 170 patients treated from January 2018 to March 2019 were included. 3D U-Net models were trained and tested under various liver conditions. The Dice similarity coefficient (DSC) and volumetric similarity (VS) were used to evaluate the segmentation accuracy. The use of quantitative volumetry for evaluating resection was compared between blood-filled and blood-free settings and between manual and automated segmentation.ResultsThe DSC values in the test dataset for hepatic veins and portal veins were 0.66 ± 0.08 (95% CI: (0.65, 0.68)) and 0.67 ± 0.07 (95% CI: (0.66, 0.69)), the VS values were 0.80 ± 0.10 (95% CI: (0.79, 0.84)) and 0.74 ± 0.08 (95% CI: (0.73, 0.76)), respectively No significant differences in FLR, FLR% assessments, or the percentage of major hepatectomy patients were noted between the blood-filled and blood-free settings (p = 0.67, 0.59 and 0.99 for manual methods, p = 0.66, 0.99 and 0.99 for automated methods, respectively) according to the use of manual and automated segmentation methods.ConclusionFully automated segmentation of hepatic veins and portal veins and FLR assessment via blood-free CT before major hepatectomy are accurate and applicable in clinical cases involving the use of DL.Critical relevance statementOur fully automatic models could segment hepatic veins, portal veins, and future liver remnant in blood-free setting on CT images before major hepatectomy with reliable outcomes.Key PointsFully automatic segmentation of hepatic veins and portal veins was feasible in clinical practice.Fully automatic volumetry of future liver remnant (FLR)% in a blood-free setting was robust.No significant differences in FLR% assessments were noted between the blood-filled and blood-free settings.Graphical

Can Machine Learning Models Based on Computed Tomography Radiomics and Clinical Characteristics Provide Diagnostic Value for Epstein-Barr Virus-Associated Gastric Cancer?

The aim of this study was to explore whether machine learning model based on computed tomography (CT) radiomics and clinical characteristics can differentiate Epstein-Barr virus-associated gastric cancer (EBVaGC) from non-EBVaGC. Contrast-enhanced CT images were collected from 158 patients with GC (46 EBV-positive, 112 EBV-negative) between April 2018 and February 2023. Radiomics features were extracted from the volumes of interest. A radiomics signature was built based on radiomics features by the least absolute shrinkage and selection operator logistic regression algorithm. Multivariate analyses were used to identify significant clinicoradiological variables. We developed 6 ML models for EBVaGC, including logistic regression, Extreme Gradient Boosting, random forest (RF), support vector machine, Gaussian Naive Bayes, and K-nearest neighbor algorithm. The area under the receiver operating characteristic curve (AUC), the area under the precision-recall curves (AP), calibration plots, and decision curve analysis were applied to assess the effectiveness of each model. Six ML models achieved AUC of 0.706-0.854 and AP of 0.480-0.793 for predicting EBV status in GC. With an AUC of 0.854 and an AP of 0.793, the RF model performed the best. The forest plot of the AUC score revealed that the RF model had the most stable performance, with a standard deviation of 0.003 for AUC score. RF also performed well in the testing dataset, with an AUC of 0.832 (95% confidence interval: 0.679-0.951), accuracy of 0.833, sensitivity of 0.857, and specificity of 0.824, respectively. The RF model based on clinical variables and Rad_score can serve as a noninvasive tool to evaluate the EBV status of gastric cancer.

Automatic Segmentation of Type A Aortic Dissection on Computed Tomography Images Using Deep Learning Approach.

Type A aortic dissection (TAAD) is a life-threatening aortic disease. The tear involves the ascending aorta and progresses into the separation of the layers of the aortic wall and the occurrence of a false lumen. Accurate segmentation of TAAD could provide assistance for disease assessment and guidance for clinical treatment. This study applied nnU-Net, a state-of-the-art biomedical segmentation network architecture, to segment contrast-enhanced CT images and quantify the morphological features for TAAD. CT datasets were acquired from 24 patients with TAAD. Manual segmentation and annotation of the CT images was used as the ground-truth. Two-dimensional (2D) nnU-Net and three-dimensional (3D) nnU-Net architectures with Dice- and cross entropy-based loss functions were utilized to segment the true lumen (TL), false lumen (FL), and intimal flap on the images. Four-fold cross validation was performed to evaluate the performance of the two nnU-Net architectures. Six metrics, including accuracy, precision, recall, Intersection of Union, Dice similarity coefficient (DSC), and Hausdorff distance, were calculated to evaluate the performance of the 2D and 3D nnU-Net algorithms in TAAD datasets. Aortic morphological features from both 2D and 3D nnU-Net algorithms were quantified based on the segmented results and compared. Overall, 3D nnU-Net architectures had better performance in TAAD CT datasets, with TL and FL segmentation accuracy up to 99.9%. The DSCs of TLs and FLs based on the 3D nnU-Net were 88.42% and 87.10%. For the aortic TL and FL diameters, the FL area calculated from the segmentation results of the 3D nnU-Net architecture had smaller relative errors (3.89-6.80%), compared to the 2D nnU-Net architecture (relative errors: 4.35-9.48%). The nnU-Net architectures may serve as a basis for automatic segmentation and quantification of TAAD, which could aid in rapid diagnosis, surgical planning, and subsequent biomechanical simulation of the aorta.