Unenhanced CT Research Articles

Atherosclerosis of the iliac arteries for the prediction of erectile dysfunction and epistaxis in men undergoing abdominal CT scan

BackgroundTo investigate the association between erectile dysfunction (ED) as well as epistaxis (ES) in relation to the extent of iliac atherosclerosis.MethodsIn this retrospective cross-sectional study, all consecutive male patients treated at our institution from 01/2016 to 12/2020 undergoing abdominal CT scan were evaluated. Patients (n = 1272) were invited by mail to participate in the study in returning two questionnaires for the evaluation of ED (IIEF-5) and ES. Patients who returned filled-in questionnaires within a 3-month deadline were included in the study. The extent of atherosclerosis in the common iliac artery (CIA) and the internal iliac artery (IIA) was assessed by calcium scoring on unenhanced CT. Stratification of results was performed according to reported IIEF-5 scores and consequential ED groups.ResultsIn total, 437 patients (34.4% of contacted) met the inclusion criteria. Forty-two patients did not fulfill predefined age requirements (< 75 years) and 120 patients had to be excluded as calcium scoring on nonenhanced CT was not feasible. Finally, 275 patients were included in the analysis and stratified into groups of “no-mild” (n = 146) and “moderate-severe” (n = 129) ED. The calcium score (r=-0.28, p < 0.001) and the number of atherosclerotic lesions (r=-0.32, p < 0.001) in the CIA + IIA showed a significant negative correlation to the IIEF-5 score, respectively. Patients differed significantly in CIA + IIA calcium score (difference: 167.4, p < 0.001) and number of atherosclerotic lesions (difference: 5.00, p < 0.001) when belonging to the “no-mild” vs. “moderate-severe” ED group, respectively. A multivariable regression model, after adjusting for relevant baseline characteristics, showed that the number of atherosclerotic CIA + IIA lesions was an independent predictor of ED (OR = 1.05, p = 0.036), whereas CIA + IIA calcium score was not (OR = 1.00031, p = 0.20). No relevant correlation was found between ES episodes and IIEF-5 scores (r=-0.069, p = 0.25), CIA + IIA calcium score (r=-0.10, p = 0.87) or number of atherosclerotic CIA + IIA lesions (r=-0.032, p = 0.60), respectively.ConclusionsThe number of atherosclerotic lesions in the iliac arteries on nonenhanced abdominal CT scans is associated with the severity of ED. This may be used to identify subclinical cardiovascular disease and to quantify the risk for cardiovascular hazards in the future.Trial registrationBASEC-Nr. 2020 − 01637.

4D-DSA for Assessment of the Angioarchitecture and Grading of Cranial Dural AVF.

Time-resolved 3D rotational angiography (4D-DSA) has been used to demonstrate details of the angioarchitecture of AVM, whereas it has rarely been used to describe features of dural AVF. In this exploratory study, we analyzed dural AVFs with a novel 4D software prototype, developed and provided by Siemens, to determine whether identification of the location of the fistulous point, grading, and treatment planning were feasible. 4D-DSA volumes were calculated from existing 3D rotational angiography data sets of patients with dural AVFs. The 4D-DSA volumes were displayed in a virtual DSA mode and MPR or MIP in 3 orthogonal planes and compared with 2D-DSA by 2 experienced neuroradiologists. Fusions with unenhanced CT or MR images were used to improve visualization of adjacent anatomic structures. Comparison with 2D-DSA showed that evaluation of the fistulous point and grading according to the classification of Borden, Cognard, or Barrow was feasible in 26 of 27 cases. In 8 of 27 cases, 4D-DSA was considered advantageous for determining the fistulous point and the course of the draining vein in the dural AVF with cortical venous drainage, especially in the frontoethmoidal and frontoparietal regions. In 6 cases, the display of angioarchitecture was considered inferior to that of 2D-DSA due to motion artifacts, suboptimal selection of the injected vessel, and lack of temporal resolution. Detailed analysis of dural AVFs according to the standardized display of 4D-DSA volumes was feasible and helpful in understanding the angioarchitecture in selected cases. Further improvement and validation of the 4D software should solidify the complementary role of 4D-DSA to conventional 2D-DSA series.

Comparative Study of Artificial Intelligence Detection Technology from Exception Ischemic Stroke Requiring Medical Help Imaging

Artificial intelligence is revolutionizing the interpretation of medical images, helping healthcare professionals save time on magnetic scans, CT scans and X-rays. Stroke is a global health problem, and ischemic stroke is one of the leading causes of death and disability in humans. Symptoms of an ischemic stroke appear suddenly and worsen within minutes, as most ischemic strokes occur suddenly, progress rapidly, and lead to the death of brain tissue within minutes or hours. This is why early detection of stroke is essential and remains a challenge for neurophysicists. Neurophysicists routinely use a variety of detection techniques to detect, assess and evaluate the ex-tent of premature ischemic changes in acute stroke brain imaging. Although several effective techniques exist, these methods have limitations due to unenhanced CT scans and invasive techniques. This study aims to demonstrate the limitations of certain methods, to determine detection methods by comparing the detection performance of automated and human brains. Stroke was evaluated through a literature review of recent studies.This article highlights comparative studies of different artificial intelligence (AI) techniques using medical imaging and allows the authors to orient themselves within these comparative studies, thus projecting themselves into the challenges facing artificial intelligence.

FRI239 Oncocytic Adrenocortical Neoplasm Of Uncertain Malignant Potential

Abstract Disclosure: K. Barnett: None. A. Sandhu: None. F. Hasan: None. Introduction: Oncocytic adrenocortical neoplasms are a rare histological subtype of adrenal mass, characterized by their abundant eosinophilic cytoplasm. Like other adrenal masses, they can be functional, or hormone-producing, and present on a spectrum from benign to malignant. Case: Our patient is a 31 year old female with a past medical history of anxiety who presented to the ER for right upper quadrant abdominal pain. She described it as a colicky, aching pain, lasting for several minutes at a time. She could identify no known trigger. A CT abdomen with contrast showed cholelithiasis and a hypodense lesion on the right adrenal gland. Her primary care physician (PCP) ordered a follow-up MRI abdomen with contrast that showed an indeterminate, 2.9 cm x 2 cm mass on the right adrenal gland without evidence of microscopic fat. After consultation with an oncologist, her PCP then recommended a PET scan, which showed that the mass was mildly hypermetabolic. She presented to the endocrinology office for further evaluation shortly thereafter. Though her abdominal pain had subsided, she did endorse a 7 lb weight loss, attributed to decreased appetite from anxiety. Additional review of systems was grossly negative. Hormonal work-up, including DHEA, renin, AM cortisol, ACTH, aldosterone, and plasma metanephrines was normal. CT-guided biopsy was deferred in favor of a repeat CT abdomen with adrenal protocol, which showed the adrenal nodule measured 40 HU on unenhanced images, with an absolute contrast washout of 50%. As this was less than the 60% threshold typically utilized to indicate a lipid-poor adenoma, the nodule was deemed indeterminate. Given the concerning features of the mass: its lipid-poor nature on unenhanced CT, metabolic activity on PET scan, and washout that was below the threshold for an adenoma, adrenalectomy was recommended. She underwent a laparoscopic right adrenalectomy, with final pathology showing oncocytic adrenocortical neoplasm of uncertain malignant potential, based on Lin-Weiss-Bisceglia criteria. She was seen by oncology, and they recommended no adjuvant treatment at this time. A repeat CTAP abdomen with adrenal protocol is planned for 1 year after this procedure. Discussion: Given the rarity of oncocytic adrenocortical neoplasms, their classification into benign, uncertain malignant potential, or malignant using Lin-Weiss-Bisceglia criteria is essential to determining appropriate treatment and surveillance. A recent retrospective cohort study showed oncocytic neoplasms with uncertain malignant potential have favorable outcomes, however, patients require continued radiographic surveillance due to the risk of late recurrence. Reference: Shirali AS, et al. Differences in Clinicopathologic Behavior of Oncocytic Adrenocortical Neoplasms and Conventional Adrenocortical Carcinomas. Ann Surg Oncol. 2022;29(9):5555-5563. doi:10.1245/s10434-022-11626-w Presentation: Friday, June 16, 2023

Influence of Contrast Materials on Dose Accuracy of MR-Linac in Patients with SBRT Liver Cancer

Objective: Intravenous injection of contrast agent during CT scanning can improve the accuracy of target area contouring, however the contrast agent will cause dose bias due to the high relative electron density. This study aims to explore the influence of contrast agent on the accuracy of dose calculation of the planning system during SBRT based on MRI-Linac for liver cancer treatment. Methods: In this study, 20 patients undergoing stereotactic body radiation therapy (SBRT) for liver cancer were selected, and their complete unenhanced CT, contrast-enhanced CT, and corresponding structures were imported into Monaco V.5.4. The target and organs at risk (OARs) in the unenhanced CT and contrast-enhanced CT were additionally contoured according to the target contouring guidelines and OARs were ranked. The average relative electron densities of OARs (lung, spinal cord, heart, rib, etc.) were calculated with Monaco TPS. The reference plan is based on unenhanced CT for plan calculation (plan1). To compare the dosimetry errors caused by the synthetic CT, the average relative electron density of all structures in unenhanced CT was forced and the plans were recalculated (plan2). To investigate dosimetric differences caused by the changes of relative electron density due to the contrast agent, the average relative electron density of all structures in contrast-enhanced CT was forced and the plans were recalculated(plan3). The dosimetric differences in groups A (plan 1 and plan2), B (plan 1 and plan3), and C (plan2 and plan3) were compared, respectively. There were not significant difference between three groups in the affected lung, heart, liver, blood, all within 3%. However, differences were significantly different in the group B. The maximum deviation of spinal cord Dmax reached 4.78%. In addition, the deviation of the dose parameters in the target area was small, except that the maximum deviation of the CI value in group B was 3.23%. For SBRT planning of liver cancer based on magnetic resonance accelerator, synthetic CT has little influence on the calculation of planned dose. The dose difference caused by contrast materials is also relatively small, although the deviation of the CI value of the target area exceeds 3%, which is also within the clinical acceptance range. However, the deviation of the maximum value of the spinal cord is relatively large, exceeding the clinically acceptable range. Therefore, when optimizing the SBRT plan for liver cancer, attention should be paid to important organs such as the spinal cord, and should be avoided as far as possible when setting the fields.

Radiomics Model for Predicting Pulmonary Metastasis in Sarcoma Patients

The lung is the most common site of metastasis in localized extremity soft tissue sarcomas (STS). Indeterminant lung nodules leads to challenges for both patients and providers. As such, accuracy in the diagnosis of pulmonary metastasis (PM) is important for clinical decision making. The purpose of this is study is to develop and evaluate a CT-based radiomic model to predict PM in sarcoma patients with indeterminate pulmonary nodules. Fifty-seven stage I-III sarcoma patients with indeterminate lung nodules on unenhanced CT that were approved for this study by our institution's internal review board (IRB). Twenty-two developed and categorized as PM and 35 without PM. A total number of 322 indeterminate nodules were investigated, where 165 & 157 nodules were from patients with and without PM, respectively. A total of 96 radiomic features (RF) (including first order statistics, 2d shape features, gray level co-occurrence, gray level size zone, gray level run length, neighboring gray tone difference & gray level dependence matrices) were taken from unenhanced CT images of each patient. We included 2D shape features since the size and shape of indeterminate lung nodules may have some correlation with PM. Spearman correlation were used to rule out redundant features (r2>0.9). T-test statistics was used to determine feature with significant differences that can differentiate responders with and without PM. Support Vector Machine (SVM) model was fitted to combinations of groups of 2-3 radiomic features. The AUC of the ROC curve was used to evaluate the model performance. Pointwise confidence intervals (CI) for AUC were calculated using 100 bootstrap replicates. Four features were found significant after Spearman correlation and T-test statistical testing: 1. Maximum (MAX); 2. Minor Axis Length (MAL); 3. Gray Level Non-uniformity (GLNU); and 4. Interquartile Range (IR). The best performing 2-feature radiomic SVM model had an AUC of 0.69 (95% CI 0.61,0.74) for the MAX and GLNU features. The best performing 3-feature radiomic SVM model had an AUC of 0.74 (95% CI 0.70,0.80) for the MAX, MAL and IR features. CT-based radiomics model has the potential to be used to predict PM in sarcoma patients presenting with indeterminate lung nodules. Additional research will focus on investigating derived parameters for existing RF, such as the ratio of the max RF to mean RF, to see if they improve the predictive performance of the model, as well as testing the models on external patient cohort. These radiomic models could be useful tools in guiding PM screening and providing a patient specific monitoring of sarcoma patients.

GA-Net: A geographical attention neural network for the segmentation of body torso tissue composition

PurposeBody composition analysis (BCA) of the body torso plays a vital role in the study of physical health and pathology and provides biomarkers that facilitate the diagnosis and treatment of many diseases, such as type 2 diabetes mellitus, cardiovascular disease, obstructive sleep apnea, and osteoarthritis. In this work, we propose a body composition tissue segmentation method that can automatically delineate those key tissues, including subcutaneous adipose tissue, skeleton, skeletal muscle tissue, and visceral adipose tissue, on positron emission tomography/computed tomography scans of the body torso. MethodsTo provide appropriate and precise semantic and spatial information that is strongly related to body composition tissues for the deep neural network, first we introduce a new concept of the body area and integrate it into our proposed segmentation network called Geographical Attention Network (GA-Net). The body areas are defined following anatomical principles such that the whole body torso region is partitioned into three non-overlapping body areas. Each body composition tissue of interest is fully contained in exactly one specific minimal body area. Secondly, the proposed GA-Net has a novel dual-decoder schema that is composed of a tissue decoder and an area decoder. The tissue decoder segments the body composition tissues, while the area decoder segments the body areas as an auxiliary task. The features of body areas and body composition tissues are fused through a soft attention mechanism to gain geographical attention relevant to the body tissues. Thirdly, we propose a body composition tissue annotation approach that takes the body area labels as the region of interest, which significantly improves the reproducibility, precision, and efficiency of delineating body composition tissues. ResultsOur evaluations on 50 low-dose unenhanced CT images indicate that GA-Net outperforms other architectures statistically significantly based on the Dice metric. GA-Net also shows improvements for the 95% Hausdorff Distance metric in most comparisons. Notably, GA-Net exhibits more sensitivity to subtle boundary information and produces more reliable and robust predictions for such structures, which are the most challenging parts to manually mend in practice, with potentially significant time-savings in the post hoc correction of these subtle boundary placement errors. Due to the prior knowledge provided from body areas, GA-Net achieves competitive performance with less training data. Our extension of the dual-decoder schema to TransUNet and 3D U-Net demonstrates that the new schema significantly improves the performance of these classical neural networks as well. Heatmaps obtained from attention gate layers further illustrate the geographical guidance function of body areas for identifying body tissues. Conclusions(i) Prior anatomic knowledge supplied in the form of appropriately designed anatomic container objects significantly improves the segmentation of bodily tissues. (ii) Of particular note are the improvements achieved in the delineation of subtle boundary features which otherwise would take much effort for manual correction. (iii) The method can be easily extended to existing networks to improve their accuracy for this application.

Enhanced and unenhanced: Radiomics models for discriminating between benign and malignant cystic renal masses on CT images: A multi-center study.

Machine learning algorithms used to classify cystic renal masses (CRMs) nave not been applied to unenhanced CT images, and their diagnostic accuracy had not been compared against radiologists. This retrospective study aimed to develop radiomics models that discriminate between benign and malignant CRMs in a triple phase computed tomography (CT) protocol and compare the diagnostic accuracy of the radiomics approach with experienced radiologists. Predictive models were established using a training set and validation set of unenhanced and enhanced (arterial phase [AP] and venous phase [VP]) CT images of benign and malignant CRMs. The diagnostic capabilities of the models and experienced radiologists were compared using Receiver Operating Characteristic (ROC) curves. On unenhanced, AP and VP CT images in the validation set, the AUC, specificity, sensitivity and accuracy for discriminating between benign and malignant CRMs were 90.0 (95%CI: 81-98%), 90.0%, 90.5% and 90.2%; 93.0% (95%CI: 86-99%), 86.7%, 95.2% and 88.3%; and 95.0% (95%CI: 90%-100%), 93.3%, 90.5% and 92.1%, respectively, for the radiomics models. Diagnostic accuracy of the radiomics models differed significantly on unenhanced images in the training set vs. each radiologist (p = 0.001 and 0.003) but not in the validation set (p = 0.230 and 0.590); differed significantly on AP images in the validation set vs. each radiologist (p = 0.007 and 0.007) but not in the training set (p = 0.663 and 0.663); and there were no differences on VP images in the training or validation sets vs. each radiologist (training set: p = 0.453 and 0.051, validation set: p = 0.236 and 0.786). Radiomics models may have clinical utility for discriminating between benign and malignant CRMs on unenhanced and enhanced CT images. The performance of the radiomics model on unenhanced CT images was similar to experienced radiologists, implying it has potential as a screening and diagnostic tool for CRMs.

18F-FDG-PET/CT-based machine learning model evaluates indeterminate adrenal nodules in patients with extra-adrenal malignancies

BackgroundTo assess the value of an 18F-FDG-positron emission tomography/computed tomography (PET/CT)-based machine learning model for distinguishing between adrenal benign nodules (ABNs) and adrenal metastases (AMs) in patients with indeterminate adrenal nodules and extra-adrenal malignancies.MethodsA total of 303 patients who underwent 18F-FDG-PET/CT with indeterminate adrenal nodules and extra-adrenal malignancies from March 2015 to June 2021 were included in this retrospective study (training dataset (n = 182): AMs (n = 97), ABNs (n = 85); testing dataset (n = 121): AMs (n = 68), ABNs (n = 55)). The clinical and PET/CT imaging features of the two groups were analyzed. The predictive model and simplified scoring system for distinguishing between AMs and ABNs were built based on clinical and PET/CT risk factors using multivariable logistic regression in the training cohort. The performances of the predictive model and simplified scoring system in both the training and testing cohorts were evaluated by the areas under the receiver operating characteristic curves (AUCs) and calibration curves. The comparison of AUCs was evaluated by the DeLong test.ResultsThe predictive model included four risk factors: sex, the ratio of the maximum standardized uptake value (SUVmax) of adrenal lesions to the mean liver standardized uptake value, the value on unenhanced CT (CTU), and the clinical stage of extra-adrenal malignancies. The model achieved an AUC of 0.936 with a specificity, sensitivity and accuracy of 0.918, 0.835, and 0.874 in the training dataset, respectively, while it yielded an AUC of 0.931 with a specificity, sensitivity, and accuracy of 1.00, 0.735, and 0.851 in the testing dataset, respectively. The simplified scoring system had comparable diagnostic value to the predictive model in both the training (AUC 0.938, sensitivity: 0.825, specificity 0.953, accuracy 0.885; P = 0.5733) and testing (AUC 0.931, sensitivity 0.735, specificity 1.000, accuracy 0.851; P = 1.00) datasets.ConclusionsOur study showed the potential ability of a machine learning model and a simplified scoring system based on clinical and 18F-FDG-PET/CT imaging features to predict AMs in patients with indeterminate adrenal nodules and extra-adrenal malignancies. The simplified scoring system is simple, convenient, and easy to popularize.

Adrenal limb thickness is associated with metabolism profiles in patients with diabetes: A cross-sectional study.

The association between adrenal size and metabolic profiles in patients with diabetes mellitus (DM) is unclear. This study was conducted to determine whether the adrenal thickness measured by computed tomography (CT) is correlated with the metabolic profiles of patients with DM. This was a cross-sectional study including 588 Chinese hospitalized patients with DM without comorbidities or medications known to affect adrenal morphology or hormone secretion. Adrenal limb thickness was measured on unenhanced chest CT. Participants were stratified into tertiles according to their total adrenal limb thickness. Linear and logistic regression models were used to estimate the correlations. After adjustment for sex and age, the adrenal thickness was positively associated with body mass index (BMI), waist circumference (WC), urinary albumin/creatinine ratio, and 24-h urinary free cortisol (UFC) and negatively correlated with high-density lipoprotein cholesterol. The sequential equation model (SEM) suggested UFC partially mediated the effect of adrenal limb thickness on WC by 12%. Adrenal thickness, but not UFC, was associated with a higher risk of existing hypertension (odds ratio [OR] = 3.78, 95% confidence interval [CI] 1.58, 9.02) and hyperlipidemia (OR = 2.76, 95% CI 1.03, 7.38), independent of age, gender, BMI, and WC. The adrenal thickness is independently associated with BMI, WC, cortisol levels, urinary albumin/creatinine ratio, hypertension, and dyslipidemia but not glycemic parameters in patients with diabetes. Our study encourages further studies to investigate the role of adrenal physiology in patients with diabetes.

Optimizing the Use of Iodinated Contrast Media: Conservation Strategies Used Across Canada During the 2022 Shortage

&#x0D; A shortage of iodinated contrast media (ICM) used in contrast-enhanced CT exams led to the adoption of necessary conservation strategies across Canada.&#x0D; Conservation strategies included multidispensing from single-use and multiuse ICM bottles, diluting or reducing ICM dose volumes, switching to weight-based dosing from fixed-based dosing, lower tube voltage, performing unenhanced CT scans, using alternative imaging modalities, or prioritizing urgent cases. One of the more common alternative conservation strategies was to prioritize urgent cases for contrast-enhanced CT exams.&#x0D; Most medical imaging staff who responded to a national survey on ICM conservation strategies reported they would return to their regular doses used before the shortage despite little to no perceived effect on the contrast conspicuity of images or on patient adverse events with reduced ICM volumes.&#x0D; The ICM shortage represents an opportunity to reconsider ICM usage practices given environmental sustainability concerns with ICM and potential cost savings in reducing its use.&#x0D;

Retrolaminar migration as a complication of intraocular silicone oil injection detected on unenhanced CT.

To describe the clinical and radiologic features of retrolaminar migration silicone oil (SiO) and observe the dynamic position of ventricular oil accumulation in supine and prone. For this retrospective study, 29 patients who had a history of SiO injection treatment and underwent unenhanced head computed tomography (CT) were included from January 2019 to October 2022. The patients were divided into migration-positive and negative groups. Clinical history and CT features were compared using Whitney U and Fisher's exact tests. The dynamic position of SiO was observed within the ventricular system in supine and prone. CT images were visually assessed for SiO migration along the retrolaminar involving pathways for vision (optic nerve, chiasm, and tract) and ventricular system. Intraocular SiO migration was found in 5 of the 29 patients (17.24%), with SiO at the optic nerve head (n=1), optic nerve (n=4), optic chiasm (n=1), optic tract (n=1), and within lateral ventricles (n=1). The time interval between SiO injection and CT examination of migration-positive cases was significantly higher than that of migration-negative patients (22.8±16.5mo vs 13.1±2.6mo, P<0.001). The hyperdense lesion located in the frontal horns of the right lateral ventricle migrated to the fourth ventricle when changing the position from supine to prone. Although SiO retrolaminar migration is unusual, the clinician and radiologist should be aware of migration routes. The supine combined with prone examination is the first-choice method to confirm the presence of SiO in the ventricular system.

Computed tomography-determined skeletal muscle density predicts 3-year mortality in initial-dialysis patients in China.

Skeletal muscle mass and quality assessed by computed tomography (CT) images of the third lumbar vertebra (L3) level have been established as risk factors for poor clinical outcomes in several illnesses, but the relevance for dialysis patients is unclear. A few studies have suggested a correlation between CT-determined skeletal muscle mass and quality at the first lumbar vertebra (L1) level and adverse outcomes. Generally, chest CT does not reach beyond L1. We aimed to determine whether opportunistic CT scan (chest CT)-determined skeletal muscle mass and quality at L1 are associated with mortality in initial-dialysis patients. This 3-year multicentric retrospective study included initial-dialysis patients from four centres between 2014 and 2017 in China. Unenhanced CT images of the L1 and L3 levels were obtained to assess skeletal muscle mass [by skeletal muscle index, (SMI), cm2 /m2 ] and quality [by skeletal muscle density (SMD), HU]. Skeletal muscle measures at L1 were compared with those at L3. The sex-specific optimal cutoff values of L1 SMI and L1 SMD were determined in relation to all-cause mortality. The outcomes were all-cause death and cardiac death. Cox regression models were applied to investigate the risk factors for death. A total of 485 patients were enrolled, of whom 257 had both L1 and L3 images. Pearson's correlation coefficient between L1 and L3 SMI was 0.84 (P<0.001), and that between L1 and L3 SMD was 0.90 (P<0.001). No significant association between L1 SMI and mortality was observed (P>0.05). Low L1 SMD (n=280, 57.73%) was diagnosed based on the optimal cutoff value (<39.56 HU for males and <33.06 HU for females). Multivariate regression analysis revealed that the low L1 SMD group had higher risks of all-cause death (hazard ratio 1.80; 95% confidence interval 1.05-3.11, P=0.034) and cardiac death (hazard ratio 3.74; 95% confidence interval 1.43-9.79, P=0.007). In initial-dialysis patients, there is high agreement between the L1 and L3 measures for SMI and SMD. Low SMD measured at L1, but not low SMI, is an independent predictor of both all-cause death and cardiac death.

Development and external validation of the multichannel deep learning model based on unenhanced CT for differentiating fat-poor angiomyolipoma from renal cell carcinoma: a two-center retrospective study.

There are undetectable levels of fat in fat-poor angiomyolipoma. Thus, it is often misdiagnosed as renal cell carcinoma. We aimed to develop and evaluate a multichannel deep learning model for differentiating fat-poor angiomyolipoma (fp-AML) from renal cell carcinoma (RCC). This two-center retrospective study included 320 patients from the First Affiliated Hospital of Sun Yat-Sen University (FAHSYSU) and 132 patients from the Sun Yat-Sen University Cancer Center (SYSUCC). Data from patients at FAHSYSU were divided into a development dataset (n = 267) and a hold-out dataset (n = 53). The development dataset was used to obtain the optimal combination of CT modality and input channel. The hold-out dataset and SYSUCC dataset were used for independent internal and external validation, respectively. In the development phase, models trained on unenhanced CT images performed significantly better than those trained on enhanced CT images based on the fivefold cross-validation. The best patient-level performance, with an average area under the receiver operating characteristic curve (AUC) of 0.951 ± 0.026 (mean ± SD), was achieved using the "unenhanced CT and 7-channel" model, which was finally selected as the optimal model. In the independent internal and external validation, AUCs of 0.966 (95% CI 0.919-1.000) and 0.898 (95% CI 0.824-0.972), respectively, were obtained using the optimal model. In addition, the performance of this model was better on large tumors (≥ 40mm) in both internal and external validation. The promising results suggest that our multichannel deep learning classifier based on unenhanced whole-tumor CT images is a highly useful tool for differentiating fp-AML from RCC.

Utility of multiparametric renal CT for differentiation of low-grade from high-grade cT1a clear cell renal cell carcinoma.

To determine if CT can differentiate low-grade from high-grade clear cell renal cell carcinoma (ccRCC) in cT1a solid ccRCC. This retrospective cross-sectional study evaluated 78 < 4 cm solid (>25% enhancing) ccRCC in 78 patients with renal CT within 12 months of surgery between January 2016 and December 2019. Two radiologists (R1/R2), blinded to pathology, independently recorded mass:size, calcification, attenuation and heterogeneity (5-point Likert scale) and recorded a 5-point ccRCC CT Score. Multivariate logistic regression (LR) was performed. There were 64.1% (50/78) low-grade (5/50 Grade 1 and 45/50 Grade 2) and 35.9% (28/78) high-grade (27/28 Grade 3 and 1/28 Grade 4) tumors.Unenhanced CT attenuation was higher (35.9±10.3 R1 and 34.9±10.7 R2 high-grade vs 29.7±10.2 R1 and 29.5±9.8 R2 low-grade, p=0.01-0.02), absolute corticomedullary phase attenuation ratio (CMphase-ratio; 0.67±0.16 R1 and 0.66±0.16 R2 vs 0.93±0.83 R1 and 0.80±0.33 R2, p=0.04-0.05) and 3-tiered stratification of CMphase-ratio (p=0.02) lower in high-grade tumors.A two-variable LR-model including unenhanced CT attenuation and CM.phase-ratio achieved area under the receiver operator characteristic curve of: 73% (95% confidence intervals 59-86%) and 72% (59-84%) for R1 and R2.ccRCC CT score differed by ccRCC grade (p<0.01 R1, R2) with high-grade tumors occurring most commonly in moderately enhancing ccRCC score 4 (46.4% [13/28] R1 and 54% [15/28]). Among cT1a ccRCC, high-grade tumors have higher unenhanced CT attenuation and are less avidly enhancing. High-grade ccRCC have higher attenuation (possibly due to less microscopic fat) and lower corticomedullary phase enhancement compared to low-grade tumors. This may result in categorization of high-grade tumors in lower ccRCC diagnostic algorithm categories.

Unenhanced computed tomography radiomics help detect endoleaks after endovascular repair of abdominal aortic aneurysm

ObjectivesTo explore the feasibility of unenhanced CT images for endoleak detection of abdominal aortic aneurysm (AAA) after endovascular repair (EVAR).MethodsPatients who visited our hospital after EVAR from July 2014 to September 2021 were retrospectively collected. Two radiologists evaluated the presence or absence of endoleaks using the combination of contrast-enhanced and unenhanced CT as the referenced standard. After segmenting the aneurysm sac of the unenhanced CT, the radiomic features were automatically extracted from the region of interest. Histogram features of patients with and without endoleak were statistically analyzed to explore the differences between the two groups. Twelve common machine learning (ML) models based on radiomic features were constructed to evaluate the performance of endoleak detection with unenhanced CT images.ResultsThe study included 216 patients (69 ± 8 years; 191 men) with AAA, including 64 patients with endoleaks. A total of 1955 radiomic features of unenhanced CT were extracted. Compared with patients without endoleak, the aneurysm sac outside the stent of patients with endoleak had higher CT attenuation (41.7 vs. 33.6, p < 0.001) with smaller dispersion (51.5 vs. 58.8, p < 0.001). The average area under the curve (AUC) of the ML models constructed with unenhanced CT radiomics was 0.86 ± 0.05, the accuracy was 81% ± 4, the sensitivity was 88% ± 10, and the specificity was 78% ± 5. When fixing the sensitivity to > 90% (92% ± 2), the models retained specificity at 72% ± 10.ConclusionsUnenhanced CT features exhibit significant differences between patients with and without endoleak and can help detect endoleaks in AAA after EVAR with high sensitivity.Clinical relevance statementUnenhanced CT radiomics can help provide an alternative method of endoleak detection in patients who have adverse reactions to contrast media. This study further exploits the value of unenhanced CT examinations in the clinical management and surveillance of postoperative abdominal aortic aneurysm.Key Points• Unenhanced CT features of the aneurysm sac outside the stent exhibit significant differences between patients with and without endoleak. The endoleak group showed higher unenhanced CT attenuation (41.7 vs 33.6, p < .001) with smaller dispersion (51.5 vs 58.8, p < .001) than the nonendoleak group.• Unenhanced CT radiomics can help detect endoleaks after intervention. The average area under the curve (AUC) of twelve common machine learning models constructed with unenhanced CT radiomics was 0.86 ± 0.05, the average accuracy was 81% ± 4.• When fixing the sensitivity to > 90% (92% ± 2), the machine learning models retained average specificity at 72% ± 10.

Radiological predictors of final infarct volume in patients with proximal vascular occlusion

IntroductionVarious clinical and radiologic variables impact the neurologic prognosis of patients with ischemic cerebrovascular accidents. About 30% of ischemic cerebrovascular accidents are caused by proximal obstruction of the anterior circulation; in these cases, systemic thrombolysis is of limited usefulness. CT angiography is indicated in candidates for endovascular treatment. Various radiologic factors, including the grade of leptomeningeal collateral circulation, as well as the length, density, and extension of the thrombus, have been identified as predictors of neurologic prognosis after anterior ischemic cerebrovascular accidents due to proximal vascular obstruction. Final infarct volume correlations with mortality and long-term functional outcome in these patients. This study aimed to determine the best predictors of final infarct volume on CT angiography in patients with ischemic cerebral accidents due to proximal occlusion. Materials and methodsThis retrospective observational study included adults with ischemic cerebrovascular accidents due to obstruction of the anterior circulation diagnosed by CT angiography in the period comprising June 2009 through December 2019. We measured the length and density of the thrombus in unenhanced CT images, and we used the clot burden score to record the grade of leptomeningeal collateral circulation and the extension of the thrombus. Then we measured the final infarct volume on follow-up CT and analyzed the correlations among these radiologic factors in the infarct volume. ResultsWe included 54 patients [mean age, 82 y; 41 (75%) women] with ischemic cerebrovascular accidents due to proximal occlusion. About 60% of the cerebrovascular accidents affected the right cerebral hemisphere, and the most commonly affected vessel was the M1 segment of the medial cerebral artery (40.7%). Final infarct volume correlated with the grade of leptomeningeal collateral circulation (p=0.03) and with the clot burden score (p=0.01). Neither the length nor the density of the thrombus correlated with final infarct volume. ConclusionThe final infarct volume can be estimated on the initial CT angiogram. Nevertheless, we found no useful predictive factors in unenhanced CT images. The best independent radiologic predictors of the final infarct volume are the grade of collateral circulation and the clot burden score, especially in patients who did not undergo mechanical thrombectomy, because mechanical thrombectomy improves outcomes. These factors are important for decision making in the management of patients with ischemic cerebrovascular accidents due to proximal occlusion.

Clinical impact of pancreatic steatosis measured by CT on the risk of post-ERCP pancreatitis: a multicenter prospective trial

Pancreatic steatosis (PS) may be a risk factor for acute pancreatitis. Whether it is also a risk factor for post-ERCP pancreatitis (PEP) has not been evaluated. This study aimed to determine the impact of PS on PEP development. This multicenter prospective trial enrolled 786 consecutive patients who underwent contrast-enhanced abdominal CT and subsequent first-time ERCP. PS was evaluated based on pancreatic attenuation on unenhanced CT images. The risk of PS for the development of PEP was evaluated using a logistic regression model. Of 527 patients included in the study, 157 (29.8%) had PS and 370 (70.2%) did not. At 24 hours after ERCP, there was a significant difference in the PEP identified in 22 patients (14.0%) in the PS group and 23 patients (6.2%) in the "no PS" (NPS) group (P= .017). Diabetes and hypertension were more common in the PS group than in the NPS group; no differences in dyslipidemia were found. Patients with PS had a higher risk for the development of PEP than those with NPS (odds ratio, 2.09; 95% confidence interval, 1.08-4.03). No other variables were identified as risk factors for PEP. PS is a significant risk factor for PEP for which preventive measures should be considered. Standardized measurement protocols to assess PS by CT are needed. (Clinical trial registration number: KCT0006068.).

Radiomics in the characterization of lipid-poor adrenal adenomas at unenhanced CT: time to look beyond usual density metrics

ObjectivesIn this study, we developed a radiomic signature for the classification of benign lipid-poor adenomas, which may potentially help clinicians limit the number of unnecessary investigations in clinical practice. Indeterminate adrenal lesions of benign and malignant nature may exhibit different values of key radiomics features.MethodsPatients who had available histopathology reports and a non-contrast-enhanced CT scan were included in the study. Radiomics feature extraction was done after the adrenal lesions were contoured. The primary feature selection and prediction performance scores were calculated using the least absolute shrinkage and selection operator (LASSO). To eliminate redundancy, the best-performing features were further examined using the Pearson correlation coefficient, and new predictive models were created.ResultsThis investigation covered 50 lesions in 48 patients. After LASSO-based radiomics feature selection, the test dataset’s 30 iterations of logistic regression models produced an average performance of 0.72. The model with the best performance, made up of 13 radiomics features, had an AUC of 0.99 in the training phase and 1.00 in the test phase. The number of features was lowered to 5 after performing Pearson’s correlation to prevent overfitting. The final radiomic signature trained a number of machine learning classifiers, with an average AUC of 0.93.ConclusionsIncluding more radiomics features in the identification of adenomas may improve the accuracy of NECT and reduce the need for additional imaging procedures and clinical workup, according to this and other recent radiomics studies that have clear points of contact with current clinical practice.Clinical relevance statementThe study developed a radiomic signature using unenhanced CT scans for classifying lipid-poor adenomas, potentially reducing unnecessary investigations that scored a final accuracy of 93%.Key Points• Radiomics has potential for differentiating lipid-poor adenomas and avoiding unnecessary further investigations.• Quadratic mean, strength, maximum 3D diameter, volume density, and area density are promising predictors for adenomas.• Radiomics models reach high performance with average AUC of 0.95 in the training phase and 0.72 in the test phase.

Deep Learning Radiomics Nomogram Based on Multiphase Computed Tomography for Predicting Axillary Lymph Node Metastasis in Breast Cancer.

This study aims to develop and validate a deep learning radiomics nomogram (DLRN) for prediction of axillary lymph node metastasis (ALNM) in breast cancer patients. We retrospectively enrolled 196 patients with non-specific invasive breast cancer confirmed by pathology, radiomics and deep learning features were extracted from unenhanced and biphasic (arterial and venous phase) contrast-enhanced CT, and the non-linear support vector machine was used to construct the radiomics signature and the deep learning signature, respectively. Next, a DLRN was developed with independent predictors and evaluated the performance of models in terms of discrimination and clinical utility. Multivariate logistic regression analysis showed that the radiomics signature, deep learning signature, and clinical n stage were independent predictors. The DLRN accurately predicted ALNM and yielded an area under the receiver operator characteristic curve of 0.893 (95% confidence interval, 0.814-0.972) in the validation set, with good calibration. Decision curve analysis confirmed that the DLRN had higher clinical utility than other predictors. The DLRN had good predictive value for ALNM in breast cancer patients and provide valuable information for individual treatment.