Non Contrast Research Articles

8234 Composite Pheochromocytoma-Ganglioneuroma: Two Clinical Cases Presenting To St. Elizabeth’s Medical Center

Abstract Disclosure: M. El Najjar: None. E. Naous: None. C. Blake: None. A.T. Sweeney: None. Introduction: Pheochromocytomas (PCCS) are rare catecholamine producing tumors that originate from the chromaffin cells of the adrenal medulla. In 3-9% of cases, they may be accompanied by other tumors. Ganglioneuromas are the most encountered tumor type that occur with PCC and are termed composite PCC-GN. The entity is very rare with < 35 cases reported. Herein, we present 2 cases of composite PCC-GN. Case Presentation: Patient 1 (Pt1) & Patient 2 (Pt2): were 35- & 54-year-old obese females. Pt1 had controlled hypertension and symptoms of anxiety along with palpitations, diaphoresis and flushing. Pt2 had type 2 diabetes and complained of abdominal pain. Metabolic studies showed elevated total plasma fractionated metanephrines(MNs). MNs in Pt1 and Pt2 were: 76 and 61pg/ml respectively (normal <57 pg/ml) and normetaneprines levels were 161 and 116pg/ml respectively (normal <148 pg/ml). Abdominal computed tomography (CT) scans demonstrated right adrenal masses in both cases with Pt1 having a 2.3 x 2.3 x 2.6 cm mass (Figure1) with smooth margins measuring 36 Hounsfield units (HU) on non-contrast(NC) CT imaging with 0% washout on delayed imaging after IV contrast. Pt2 on NCCT had a 4.5 x 3.5 cm right adrenal mass measuring 73 HU and a subsequent abdominal MRI(Figure2) demonstrated the mass to have heterogenous signal intensity. Both patients were prepared with pre-operative a and b blockade and underwent laparoscopic robotic adrenalectomies without complications. Pathological analyses in both cases revealed composite PCC-GN. Pt2 underwent genetic testing which did not reveal any known mutations. Surveillance in both patients demonstrated no recurrence. Clinical lesson: The clinical presentation of PCC-GNs is indistinguishable from isolated PCCs. The definitive diagnosis is through histopathological analysis. Histologically the composite tumor is composed of two distinct patterns with PCC showing angular cells with granular, basophilic cytoplasm, arranged in nests, while GNs show ganglion cells in fibrous stroma. Immunohistochemistry show the PCCs staining for chromogranin A and synaptophysin while GNs stain for S-100 protein and neurofilament antibody. Most cases occur in patients between 40 - 60 years old. Most PCC-GNs are functional, with increased MNs. PCC-GN composite tumors may occur sporadically but have also been described as part of syndromes such as MEN2A and NF1. The definitive treatment for PCC-GN is complete surgical excision which is associated with an excellent prognosis. The probability of recurrence is low though distant metastases have rarely been reported. Therefore, lifetime surveillance is critical. Presentation: 6/3/2024

Comparison of non-contrast abbreviated MRI and ultrasound as surveillance modalities for HCC

Ultrasound (US) is recommended for HCC surveillance in high-risk patients but has limited performance in detecting early-stage HCC. We aimed to compare the diagnostic performance of biannual US and annual non-contrast abbreviated magnetic resonance imaging (NC-AMRI) as HCC surveillance modalities in high-risk patients. This prospective, multicenter cohort study enrolled participants with an estimated annual risk of HCC greater than 5% between October 2015 and April 2017. Participants underwent six rounds of HCC surveillance at 6-month intervals, with both US and NC-AMRI at rounds 1, 3, and 5, and only US at rounds 2, 4, and 6. The sensitivity, diagnostic yield (DY), and false referral rate (FRR) for HCC detection by US and NC-AMRI were compared. In total, 208 participants underwent 980 US and 516 NC-AMRI examinations during 30 months of follow-up. Among them, 34 HCCs were diagnosed in 31 participants, with 20 (64.5%) classified as very early-stage and 11 (35.5%) as early-stage HCC. The sensitivity of annual NC-AMRI (71.0%, 22/31) was marginally higher than that of biannual US (45.2%, 14/31; p= 0.077). NC-AMRI showed a significantly higher DY than US (4.26% vs. 1.43%, p <0.001), with a similar FRR (2.91% vs. 3.06%, p= 0.885). A simulation of alternating US and NC-AMRI at 6-month intervals yielded a sensitivity of 83.9% (26/31), significantly exceeding that of biannual US (p= 0.006). Annual NC-AMRI showed a marginally higher sensitivity than biannual US for HCC detection in high-risk patients. The DY of annual NC-AMRI was significantly higher than that of biannual US, without increasing the FRR. Thus, alternating US and NC-AMRI at 6-month intervals could be an optimal surveillance strategy for high-risk patients. Current guidelines permit the use of magnetic resonance imaging (MRI) as a surveillance tool for hepatocellular carcinoma in patients in whom ultrasonography (US) is inadequate. However, the specific indications, imaging sequences, and intervals for MRI surveillance remain unclear. In our study, we found that annual non-contrast abbreviated MRI exhibited marginally higher sensitivity and significantly better diagnostic yield than biannual US in patients at high risk of hepatocellular carcinoma. Alternating US and non-contrast abbreviated MRI at 6-month intervals led to significantly improved sensitivity compared to biannual US, making it a potentially optimal surveillance strategy for high-risk patients. NCT02551250.

Cryoballoon ablation without use of contrast for the treatment of paroxysmal atrial fibrillation.

Cryoballoon ablation (CBA) for atrial fibrillation (AF) is usually preceded by demonstrating pulmonary vein (PV) occlusion using contrast. The aim of the study was to determine efficacy and safety of a simplified protocol for CBA performed without demonstrating PV occlusion and compare achieved results with conventional CBA. Paroxysmal AF patients undergoing a first-time CBA were prospectively included. In the non-contrast (NC) group CBA was performed using standardized protocol without demonstrating PV occlusion. In the conventional contrast (CC) group ablations were performed after confirmation of PV occlusion. The NC and CC groups comprised 51 and 22 patients, respectively. PVI according to the group assignment was achieved in 34 (67%) and 21 (95.5%) patients from the NC and CC groups, respectively (p < 0.001). In the NC group, 184 (90%) out of 204 veins were isolated without venography. There were no differences between the NC and CC groups in terms of procedure duration (89.7 ± 22.6 vs. 90.0 ± 20.6 min; p = 0.7) and fluoroscopy time (15.3 ± 6.3 vs. 15 ± 4.5 min; p = 0.8). In the NC group, the use of contrast was significantly lower compared to the CC group (4.9 ± 10.1 vs. 19.4 ± 8.6 mL, p < 0.001). There were no serious adverse events in both groups. A 1-year freedom from AF was achieved in 73.5% and 71.5% of patients from the NC and CC groups, respectively (p = 1). Cryoballoon ablation without demonstrating vein occlusion with contrast is safe and feasible. Proposed simplified approach enables isolation of the vast majority of pulmonary veins with a significant reduction in the amount of contrast used.

Correcting Posterior Paraspinal Muscle Computed Tomography Density for Intravenous Contrast Material Independent of Sex and Vascular Phase.

Intravenous contrast poses challenges to computed tomography (CT) muscle density analysis. We developed and tested corrections for contrast-enhanced CT muscle density to improve muscle analysis and the utility of CT scans for the assessment of myosteatosis. Using retrospective images from 240 adults who received routine abdominal CT imaging from March to November 2020 with weight-based iodine contrast, we obtained paraspinal muscle density measurements from noncontrast (NC), arterial, and venous-phase images. We used a calibration sample to develop 9 different mean and regression-based corrections for the effect of contrast. We applied the corrections in a validation sample and conducted equivalence testing. We evaluated 140 patients (mean age 52.0y [SD: 18.3]; 60% female) in the calibration sample and 100 patients (mean age 54.8y [SD: 18.9]; 60% female) in the validation sample. Contrast-enhanced muscle density was higher than NC by 8.6HU (SD: 6.2) for the arterial phase (female, 10.4HU [SD: 5.7]; male, 6.0HU [SD:6.0]) and by 6.4HU [SD:8.1] for the venous phase (female, 8.0HU [SD: 8.6]; male, 4.0HU [SD: 6.6]). Corrected contrast-enhanced and NC muscle density was equivalent within 3HU for all correctionns. The -7.5HU correction, independent of sex and phase, performed well for arterial (95% CI: -0.18, 1.80HU) and venous-phase data (95% CI: -0.88, 1.41HU). Our validated correction factor of -7.5HU renders contrast-enhanced muscle density statistically similar to NC density and is a feasible rule-of-thumb for clinicians to implement.

The combination of non-contrast abbreviated MRI and alpha foetoprotein has high performance for hepatocellular carcinoma screening

ObjectivesThis study aimed to compare two abbreviated MRI (AMRI) protocols to complete MRI for HCC detection: non-contrast (NC)-AMRI without/with alpha foetoprotein (AFP) and dynamic contrast-enhanced (Dyn)-AMRI.MethodsThis retrospective single-center study included 351 patients (M/F: 264/87, mean age: 57y) with chronic liver disease, who underwent MRI for HCC surveillance between 2014 and 2020. Two reconstructed AMRI sets were obtained based on complete MRI: NC-AMRI (T2-weighted imaging (WI) + diffusion-WI) and Dyn-AMRI (T2-WI + dynamic T1-WI) and were assessed by 2 radiologists who reported all suspicious lesions, using LI-RADS/adapted LI-RADS classification. The reference standard was based on all available patient data. Inter-reader agreement was assessed and MRI diagnostic performance was compared to the reference standard.ResultsThe reference standard demonstrated 83/351 HCC-positive patients (prevalence: 23.6%, median size: 22 mm, and positive MRIs: 83/631). Inter-reader agreement was substantial for all sets. Sensitivities of Dyn-AMRI and complete MRI (both 92.8%) were similar, higher than NC-AMRI (72.3%, p < 0.001). Specificities were not different between sets. NC-AMRI + AFP (92.8%) had similar sensitivity to Dyn-AMRI and complete MRI. In patients with small size HCCs (≤ 2 cm), sensitivities of Dyn-AMRI (85.3%) and complete MRI (88.2%) remained similar (p = 0.564), also outperforming NC-AMRI (52.9%, p < 0.05). NC-AMRI + AFP had similar sensitivity (88.2%) to Dyn-AMRI and complete MRI (p = 0.706 and p = 1, respectively).ConclusionsDyn-AMRI has similar diagnostic performance to complete MRI for HCC detection, while both outperform NC-AMRI, especially for small size HCCs. NC-AMRI + AFP demonstrates similar sensitivity to Dyn-AMRI and complete MRI.Clinical relevance statementDue to the low sensitivity of ultrasound for hepatocellular screening, new screening methods are needed. Abbreviated MRI (AMRI) is a candidate, especially non-contrast AMRI with serum alpha foetoprotein as the acquisition time is low, without the need for contrast medium injection.Key Points• Dynamic contrast-enhanced abbreviated MRI using extracellular gadolinium-based contrast agent and complete MRI have similar diagnostic performance for hepatocellular carcinoma detection in an at-risk population.• Non-contrast abbreviated MRI with alpha foetoprotein has similar diagnostic performance to dynamic contrast-enhanced abbreviated MRI and complete MRI, including when considering small size hepatocellular carcinoma ≤ 2 cm.• Non-contrast abbreviated MRI and dynamic contrast-enhanced abbreviated MRI can be performed in 7 and 10 min, excluding patient setup time.

Transformer-based 3D U-Net for pulmonary vessel segmentation and artery-vein separation from CT images.

Transformer-based methods have led to the revolutionizing of multiple computer vision tasks. Inspired by this, we propose a transformer-based network with a channel-enhanced attention module to explore contextual and spatial information in non-contrast (NC) and contrast-enhanced (CE) computed tomography (CT) images for pulmonary vessel segmentation and artery-vein separation. Our proposed network employs a 3D contextual transformer module in the encoder and decoder part and a double attention module in skip connection to effectively finish high-quality vessel and artery-vein segmentation. Extensive experiments are conducted on the in-house dataset and the ISICDM2021 challenge dataset. The in-house dataset includes 56 NC CT scans with vessel annotations and the challenge dataset consists of 14 NC and 14 CE CT scans with vessel and artery-vein annotations. For vessel segmentation, Dice is 0.840 for CE CT and 0.867 for NC CT. For artery-vein separation, the proposed method achieves a Dice of 0.758 of CE images and 0.602 of NC images. Quantitative and qualitative results demonstrated that the proposed method achieved high accuracy for pulmonary vessel segmentation and artery-vein separation. It provides useful support for further research associated with the vascular system in CT images. The code is available at https://github.com/wuyanan513/Pulmonary-Vessel-Segmentation-and-Artery-vein-Separation .

EVALUATION OF RADIATION DOSE IN MULTI-SLICE COMPUTED TOMOGRAPHY PROTOCOLS OF HEAD AND NECK REGIONS.

In head and neck computed tomography (CT) imaging, the optimisation of radiation dose is crucial due to the presence of radio-sensitive organs. This study aimed to evaluate the radiation dose in multi-slice CT for head and neck examinations. Volume CT dose index, dose length product and effective dose (E) were assessed for 10 head and neck CT scans performed on 292 adult patients (mean age 49.2±15.9y). The study resulted in median E values of 0.82, 1.62, 2.43, 0.93, 1.70, 0.83, 3.55, 6.25, 2.19 and 5.26mSv, respectively, for sinuses (non-contrast (NC)), sinuses (NC) and contrast-enhanced (CE), petrous bone (PTB)/internal auditory meatus (IAM) (NC+CE), PTB/IAM (NC), orbit (NC+CE), orbit (NC), brain with the orbit (NC), brain CT angiography (CTA) subtraction, neck (NC) and brain/neck (NC). Furthermore, the overall radiation doses of this institution were found to be below the values suggested by similar studies. However, optimisation of the dose is required for brain CTA.

Delta Radiomics Model Predicts Lesion-Level Responses to Tyrosine Kinase Inhibitors in Patients with Advanced Renal Cell Carcinoma: A Preliminary Result

This study aimed to develop and internally validate computed tomography (CT)-based radiomic models to predict the lesion-level short-term response to tyrosine kinase inhibitors (TKIs) in patients with advanced renal cell carcinoma (RCC). This retrospective study included consecutive patients with RCC that were treated using TKIs as the first-line treatment. Radiomic features were extracted from noncontrast (NC) and arterial-phase (AP) CT images. The model performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA). A total of 36 patients with 131 measurable lesions were enrolled (training: validation = 91: 40). The model with five delta features achieved the best discrimination capability with AUC values of 0.940 (95% CI, 0.890‒0.990) in the training cohort and 0.916 (95% CI, 0.828‒1.000) in the validation cohort. Only the delta model was well calibrated. The DCA showed that the net benefit of the delta model was greater than that of the other radiomic models, as well as that of the treat-all and treat-none criteria. Models based on CT delta radiomic features may help predict the short-term response to TKIs in patients with advanced RCC and aid in lesion stratification for potential treatments.

Optimising CT-guided biopsies of sclerotic bone lesions in cancer patients

ObjectivesInvestigate the laboratory, imaging and procedural factors that are associated with a tumour-positive and/or NGS-feasible CT-guided sclerotic bone lesion biopsy result in cancer patients.MethodsIn total, 113 CT-guided bone biopsies performed in cancer patients by an interventional radiologist in one institution were retrospectively reviewed. Sixty-five sclerotic bone biopsies were eventually included and routine blood parameters and tumour marker levels were recorded. Non-contrast (NC) biopsy CTs (65), contrast-enhanced CTs (24), and PET/CTs (22) performed within four weeks of biopsy were reviewed; lesion location, diameter, lesion-to-cortex distance, and NC-CT appearance (dense-sclerosis versus mild-sclerosis) were noted. Mean NC-CT, CE-CT HU, and PET SUVmax were derived from biopsy tract and lesion segmentations. Needle diameter, tract length, and number of samples were noted. Comparisons between tumour-positive/negative and next-generation sequencing (NGS)-feasible/non-feasible biopsies determined significant (p < 0.05) laboratory, imaging, and procedural parameter differences.ResultsSeventy-four percent of biopsies were tumour-positive. NGS was feasible in 22/30 prostate cancer patients (73%). Neither laboratory blood parameters, PET/CT availability, size, nor lesion-to-cortex distance affected diagnostic yield or NGS feasibility (p > 0.298). Eighty-seven percent of mildly sclerotic bone (mean 244 HU) biopsies were positive compared with 56% in dense sclerosis (622 HU, p = 0.005) and NC-CT lesion HU was significantly lower in positive biopsies (p = 0.003). A 610 HU threshold yielded 89% PPV for tumour-positive biopsies and a 370 HU threshold 94% PPV for NGS-feasible biopsies. FDG-PET and procedural parameters were non-significant factors (each p > 0.055).ConclusionIn cancer patients with sclerotic bone disease, targeting areas of predominantly mild sclerosis in lower CT-attenuation lesions can improve tumour tissue yield and NGS feasibility.Key Points• Areas of predominantly mild sclerosis should be preferred to areas of predominantly dense sclerosis for CT-guided bone biopsies in cancer patients.• Among sclerotic bone lesions in prostate cancer patients, lesions with a mean HU below 370 should be preferred as biopsy targets to improve NGS feasibility.• Laboratory parameters and procedure related factors may have little implications for CT-guided sclerotic bone biopsy success.

Radiomics Analysis of Computed Tomography for Prediction of Thyroid Capsule Invasion in Papillary Thyroid Carcinoma: A Multi-Classifier and Two-Center Study.

ObjectiveTo investigate the application of computed tomography (CT)-based radiomics model for prediction of thyroid capsule invasion (TCI) in papillary thyroid carcinoma (PTC).MethodsThis retrospective study recruited 412 consecutive PTC patients from two independent institutions and randomly assigned to training (n=265), internal test (n=114) and external test (n=33) cohorts. Radiomics features were extracted from non-contrast (NC) and artery phase (AP) CT scans. We also calculated delta radiomics features, which are defined as the absolute differences between the extracted radiomics features. One-way analysis of variance and least absolute shrinkage and selection operator were used to select optimal radiomics features. Then, six supervised machine learning radiomics models (k-nearest neighbor, logistic regression, decision tree, linear support vector machine [L-SVM], Gaussian-SVM, and polynomial-SVM) were constructed. Univariate was used to select clinicoradiological risk factors. Combined models including optimal radiomics features and clinicoradiological risk factors were constructed by these six classifiers. The prediction performance was evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).ResultsIn the internal test cohort, the best combined model (L-SVM, AUC=0.820 [95% CI 0.758–0.888]) performed better than the best radiomics model (L-SVM, AUC = 0.733 [95% CI 0.654–0.812]) and the clinical model (AUC = 0.709 [95% CI 0.649–0.783]). Combined-L-SVM model combines 23 radiomics features and 1 clinicoradiological risk factor (CT-reported TCI). In the external test cohort, the AUC was 0.776 (0.625–0.904) in the combined-L-SVM model, showing that the model is stable. DCA demonstrated that the combined model was clinically useful.ConclusionsOur combined model based on machine learning incorporated with CT radiomics features and the clinicoradiological risk factor shows good predictive ability for TCI in PTC.

Meta-Analysis of the Accuracy of Abbreviated Magnetic Resonance Imaging for Hepatocellular Carcinoma Surveillance: Non-Contrast versus Hepatobiliary Phase-Abbreviated Magnetic Resonance Imaging.

Simple SummaryUltrasonography is recommended as a standard surveillance modality, but the performance of surveillance ultrasound for detecting early-stage hepatocellular carcinoma (HCC) is limited. Motivated to provide a more sensitive method, abbreviated magnetic resonance imaging (AMRI) protocols have been introduced for HCC surveillance. We aimed to systematically determine the diagnostic performance of surveillance AMRI for detecting HCC. This meta-analysis of 10 studies comprising 1547 patients found that the pooled sensitivity and specificity of surveillance AMRI for detecting HCC were 86% and 96%, respectively. Hepatobiliary phase contrast-enhanced AMRI showed significantly higher sensitivities for detecting HCC than non-contrast AMRI (87% vs. 82%), but significantly lower specificities (93% vs. 98%). Therefore, surveillance AMRI had overall good diagnostic performance for detecting HCC and might be clinically useful for HCC surveillance. In addition, AMRI protocol should be selected with consideration of the advantages and disadvantages of each protocol.We aimed to determine the performance of surveillance abbreviated magnetic resonance imaging (AMRI) for detecting hepatocellular carcinoma (HCC), and to compare the performance of surveillance AMRI according to different protocols. Original research studies reporting the performance of surveillance AMRI for the detection of HCC were identified in MEDLINE, EMBASE, and Cochrane databases. The pooled sensitivity and specificity of surveillance AMRI were calculated using a hierarchical model. The pooled sensitivity and specificity of contrast-enhanced hepatobiliary phase (HBP)-AMRI and non-contrast (NC)-AMRI were calculated and compared using bivariate meta-regression. Ten studies, including 1547 patients, reported the accuracy of surveillance AMRI. The pooled sensitivity and specificity of surveillance AMRI for detecting any-stage HCC were 86% (95% confidence interval (CI), 80–90%; I2 = 0%) and 96% (95% CI, 93–98%; I2 = 80.5%), respectively. HBP-AMRI showed a significantly higher sensitivity for detecting HCC than NC-AMRI (87% vs. 82%), but significantly lower specificity (93% vs. 98%) (p = 0.03). Study quality and MRI magnet field strength were factors significantly associated with study heterogeneity (p ≤ 0.01). In conclusion, surveillance AMRI showed good overall diagnostic performance for detecting HCC. HBP-AMRI had significantly higher sensitivity for detecting HCC than NC-AMRI, but lower specificity.

Evaluation of radiomics and machine learning in identification of aggressive tumor features in renal cell carcinoma (RCC).

The purpose of this study was to evaluate the use of CT radiomics features and machine learning analysis to identify aggressive tumor features, including high nuclear grade (NG) and sarcomatoid (sarc) features, in large renal cell carcinomas (RCCs). CT-based volumetric radiomics analysis was performed on non-contrast (NC) and portal venous (PV) phase multidetector computed tomography images of large (> 7cm) untreated RCCs in 141 patients (46W/95M, mean age 60years). Machine learning analysis was applied to the extracted radiomics data to evaluate for association with high NG (grade 3-4), with multichannel analysis for NG performed in a subset of patients (n = 80). A similar analysis was performed in a sarcomatoid rich cohort (n = 43, 31M/12F, mean age 63.7 years) using size-matched non-sarcomatoid controls (n = 49) for identification of sarcomatoid change. The XG Boost Model performed best on the tested data. After manual and machine feature extraction, models consisted of 3, 7, 5, 10 radiomics features for NC sarc, PV sarc, NC NG and PV NG, respectively. The area under the receiver operating characteristic curve (AUC) for these models was 0.59, 0.65, 0.69 and 0.58 respectively. The multichannel NG model extracted 6 radiomic features using the feature selection strategy and showed an AUC of 0.67. Statistically significant but weak associations between aggressive tumor features (high nuclear grade, sarcomatoid features) in large RCC were identified using 3D radiomics and machine learning analysis.

Prediction of histologic grade and type of small (< 4 cm) papillary renal cell carcinomas using texture and neural network analysis: a feasibility study.

To predict the histologic grade and type of small papillary renal cell carcinomas (pRCCs) using texture analysis and machine learning algorithms. This was a retrospective HIPAA-compliant study. 24 noncontrast (NC), 22 corticomedullary (CM) phase, and 24 nephrographic (NG) phase CTs of small (< 4 cm) surgically resected pRCCs were identified. Surgical pathology classified the tumors as low- or high-Fuhrman histologic grade and type 1 or 2. The axial image with the largest cross-sectional tumor area was exported and segmented. Six histogram and 31 texture (20 gray-level co-occurrences and 11 gray-level run-lengths) features were calculated for each tumor in each phase. Feature values in low- versus high-grade and type 1 versus 2 pRCCs were compared. Area under the receiver operating curve (AUC) was calculated for each feature to assess prediction of histologic grade and type of pRCCs in each phase. Histogram, texture, and combined histogram and texture feature sets were used to train and test three classification algorithms (support vector machine (SVM), random forest, and histogram-based gradient boosting decision tree (HGBDT)) with stratified shuffle splits and threefold cross-validation; AUCs were calculated for each algorithm in each phase to assess prediction of histologic grade and type of pRCCs. Individual histogram and texture features did not have statistically significant differences between low- and high-grade or type 1 and type 2 pRCCs across all phases. Individual features had low predictive power for tumor grade or type in all phases (AUC < 0.70). HGBDT was highly accurate at predicting pRCC histologic grade and type using histogram, texture or combined histogram and texture feature data from the CM phase (AUCs = 0.97-1.0). All algorithms had highest AUCs using CM phase feature data sets; AUCs decreased using feature sets from NC or NG phases. The histologic grade and type of small pRCCs can be predicted with classification algorithms using CM histogram and texture features, which outperform NC and NG phase image data. The accurate prediction of pRCC histologic grade and type may be able to further guide management of patients with small (< 4 cm) pRCCs being considered for active surveillance.

Qualitative versus automatic evaluation of CT perfusion parameters in acute posterior circulation ischaemic stroke

PurposeTo compare the diagnostic accuracy (ACC) in the detection of acute posterior circulation strokes between qualitative evaluation of software-generated colour maps and automatic assessment of CT perfusion (CTP) parameters.MethodsWere retrospectively collected 50 patients suspected of acute posterior circulation stroke who underwent to CTP (GE “Lightspeed”, 64 slices) within 24 h after symptom onset between January 2016 and December 2018. The Posterior circulation-Acute Stroke Prognosis Early CT Score (pc-ASPECTS) was used for quantifying the extent of ischaemic areas on non-contrast (NC)CT and colour-coded maps generated by CTP4 (GE) and RAPID (iSchemia View) software. Final pc-ASPECTS was calculated on follow-up NCCT and/or MRI (Philips Intera 3.0 T or Philips Achieva Ingenia 1.5 T). RAPID software also elaborated automatic quantitative mismatch maps.ResultsBy qualitative evaluation of colour-coded maps, MTT-CTP4D and Tmax-RAPID showed the highest sensitivity (SE) (88.6% and 90.9%, respectively) and ACC (84% and 88%, respectively) compared with the other perfusion parameters (CBV, CBF). Baseline NCCT and CBF provided by RAPID quantitative perfusion mismatch maps had the lowest SE (29.6% and 6.8%, respectively) and ACC (38% and 18%, respectively). CBF and Tmax assessment provided by quantitative RAPID perfusion mismatch maps showed significant lower SE and ACC than qualitative evaluation. No significant differences were found between the pc-ASPECTSs assessed on colour-coded MTT and Tmax maps neither between the scores assessed on colour-coded CBV-CTP4D and CBF-RAPID maps.ConclusionQualitative analysis of colour-coded maps resulted more sensitive and accurate in the detection of ischaemic changes than automatic quantitative analysis.

Texture analysis and machine learning algorithms accurately predict histologic grade in small (&lt; 4 cm) clear cell renal cell carcinomas: a pilot study

To predict the histologic grade of small clear cell renal cell carcinomas (ccRCCs) using texture analysis and machine learning algorithms. Fifty-two noncontrast (NC), 26 corticomedullary (CM) phase, and 35 nephrographic (NG) phase CTs of small (< 4cm) surgically resected ccRCCs were retrospectively identified. Surgical pathology classified the tumors as low- or high-Fuhrman histologic grade. The axial image with the largest cross-sectional tumor area was exported and segmented. Six histogram and 31 texture (gray-level co-occurrences (GLC) and gray-level run-lengths (GLRL)) features were calculated for each tumor in each phase. T testing compared feature values in low- and high-grade ccRCCs, with a (Benjamini-Hochberg) false discovery rate of 10%. Area under the receiver operating curve (AUC) was calculated for each feature to assess prediction of low- and high-grade ccRCCs in each phase. Histogram, texture, and combined histogram and texture data sets were used to train and test four algorithms (k-nearest neighbor (KNN), support vector machine (SVM), random forests, and decision tree) with tenfold cross-validation; AUCs were calculated for each algorithm in each phase to assess prediction of low- and high-grade ccRCCs. Zero, 23, and 0 features in the NC, CM, and NG phases had statistically significant differences between low and high-grade ccRCCs. CM histogram skewness and GLRL short run emphasis had the highest AUCs (0.82) in predicting histologic grade. All four algorithms had the highest AUCs (0.97) predicting histologic grade using CM histogram features. The algorithms' AUCs decreased using histogram or texture features from NC or NG phases. The histologic grade of small ccRCCs can be accurately predicted with machine learning algorithms using CM histogram features, which outperform NC and NG phase image data.

03:54 PM Abstract No. 341 Imaging triage of lower gastrointestinal bleed: assessing utility of selected contrast phases of multiphasic CTA

Many institutions advocate for obtaining arterial (Art), non-contrast (NC), and delayed (Del) phases when assessing for active lower gastrointestinal bleed (LGIB) on multiphasic computed tomography arteriograms (CTA). We aimed to evaluate the necessity of all three phases of CTA for accurate diagnosis of active LGIB. This study is IRB approved and HIPAA compliant. We identified all CTA exams for LGIB performed at our institution from Oct 2012 to Oct 2017. We selected a random sample of 50 exams positive for active LGIB as well as an age/sex/protocol-matched cohort of 50 exams with no evidence of LGIB based on initial interpretation and throughout workup. A randomized experimental block design was used where three interventional radiologists blinded to patient data each assessed all exams using (1) Art only, (2) Art/NC, and (3) Art/NC/Del phase configurations accounting for presence of active LGIB, location, source vessel, and time required for interpretation. Evaluations were spaced one month apart. This experimental design allowed for estimation of incremental benefit of additional phases while controlling for carryover effects. Sensitivity and specificity were powered for equally (50/50). Generalized estimating equations (GEE) with sandwich estimation were used to estimate the relative difference of each phase using SAS 9.4. Mean patient age was 73.8 years. Specificity increased with added contrast phases (Art only 0.72; Art/NC 0.86; Art/NC/Del 0.95, p<0.001) without sacrifice in sensitivity (Art only 0.72; Art/NC 0.70; Art/NC/Del 0.72, p=0.3047) or increased mean time required per study (sec, Art only 31.7; Art/NC 30.6; Art/NC/Del 32.9, p=0.1813). Overall agreement among readers (Kappa) similarly increased (Art only 0.47; Art/NC 0.65; Art/NC/Del 0.79). When assessing for LGIB on multiphasic CTA, additional non-contrast and delayed phases incrementally increase specificity without decreased sensitivity or increased time required per study. In addition, the overall agreement between readers for diagnosis of LGIB as well as identifying bleed source and location increases with additional phases per exam, altogether supporting use of triphasic CTA in diagnosing LGIB.

Dose optimization in computed tomography of brain using CARE kV and CARE Dose 4D

Background: Computed Tomography (CT) scan of brain is the most widely used CT examination. Latest CT scanners have the potential to deliver very low radiation dose by utilizing tube potential and tube current modulation techniques. We aim to determine the application of CARE kV (tube potential modulation) and CARE Dose4D (tube current modulation) in CT scan of brain. Both CARE kV and CARE Dose4D are well-established innovative technology of Siemens Medical Solutions.&#x0D; Methodology: A prospective hospital-based study was conducted during four months at Tribhuvan University Teaching Hospital (TUTH). The data were collected on a Siemens Somatom Definition Edge 128 slices CT scanner. Non-random purposive sampling technique was employed. Ethical approval and consent to participate were taken for every participant. Non-contrast (NC) CT images were acquired without using CARE kV and CARE Dose4D, whereas during contrast-enhanced (CE) investigation, both were turned on keeping other scanning parameters constant for each individual.&#x0D; Results: A total of 72 patients, 42 males and 28 females - mean age 41y (range 16-87y) participated in this study. The Body Mass Index (BMI) was 22.0, range 20.1-25.0. The mean value of Computed Tomography Dose Index (CTDI), Dose Length Product (DLP) and Effective Dose (ED) before and after switching on both CARE kV and CARE Dose4D were 58.19±0.35 and 39.67±3.59 milli-Gray (mGy), 946.67 and 652.58 mGy-cm, and 1.98 and 1.36 milli-Sievert (mSv) respectively.&#x0D; Conclusion: CARE kV and CARE Dose4D can reduce radiation dose in CT scan of brain without loss of image quality.

Rate and risk factors for AKI after CT scans in a cancer cohort .

Aims: The proinflammatory milieu in cancer patients may expose them to increased risk for acute kidney injury (AKI) after IV contrast (CON). The aims of this study were to determine: (1) the rates of AKI after CON and noncontrast (NC) CT scans in cancer inpatients, (2) if rates differed among cancer subtypes, and (3) whether recent chemotherapy, comorbid conditions, or nephrotoxins increase AKI after CON. Materials and methods: Retrospective data was collected on adults who had received a CON or NC CT from January 1, 2012 to December 30, 2014. AKI was defined as a > 1.5× increased baseline creatinine. Data was analyzed using Rao-Scott χ2-test, propensity score matching, and logistic regressions. Results: A total of 7,512 CT scans were performed in 4,456 patients (4,958 NC, 2,554 CON). The rate for AKI with CON was 7.3% and 11.4% (p < 0.001) with NC imaging. The risk of AKI increased with lower baseline eGFR: for eGFR ≤ 29 mL/min/1.73m2, OR = 1.83 (p = 0.0002); for eGFR 30 – 59 mL/min/1.73m2, OR = 1.5 compared to eGFR ≥ 60 mL/min/1.73m2 (p < 0.0001). AKI rates were higher when any chemotherapy was given within 60 days of CT (OR = 1.22, p < 0.02), with congestive heart failure (OR 1.51, p = 0.0006), and history of AKI (OR 3.89, p < 0.0001). In 1:1 propensity score matched samples, the OR for AKI after CON was 0.87 (p = 0.23) compared to NC. Conclusion: In cancer patients, eGFR below 59 mL/min/1.73m2 were associated with increased rate of AKI, independent of contrast exposure. Congestive heart failure and prior AKI were also associated with increased rates of AKI.

Density and enhancement of the pancreatic tail on computer tomography predicts acinar score and pancreatic fistula after pancreatoduodenectomy

BackgroundAcinar score calculated at the pancreatic resection margin is associated with postoperative pancreatic fistula (POPF) after pancreatoduodenectomy (PD). The present study evaluates the association between (i) computed tomography (CT) density of the pancreas and the acinar score of the pancreatic resection margin, and (ii) CT density of the pancreas and POPF after PD. MethodsConsecutive patients who underwent PD were included for analysis. CT densities of the pancreatic head, neck, body and tail were measured in non-contrast (NC), arterial (ART) and portal venous (PV) phases. Histologic slides of the pancreatic resection margin were scored for acinar cell density. ResultsNinety patients were included for analysis. Non-contrast density of the pancreatic tail was a good predictor of POPF (AUROC 0.704, p = 0.036), and a cut-off value of >40 Hounsfield units predicted POPF with 70.0% sensitivity and 73.4% specificity. The ratio of densities between PV and NC phases in the pancreatic tail was also a good predictor of POPF (AUROC 0.712, p = 0.030), and a cut-off value of <2.29 predicted POPF with 70.9% sensitivity and 80% specificity. ConclusionNon-contrast CT density of the pancreatic tail correlates with acinar cell density of the pancreatic resection margin and predicts the development of POPF after PD.

Can multiphase CT scan distinguish between papillary renal cell carcinoma type 1 and type 2?

To investigate the utility of multiphase computed tomography (CT) and percutaneous renal mass biopsy (PRMB) in differentiating between papillary renal cell carcinoma (pRCC)-Type 1 and -Type 2, as emerging data have suggested differential enhancement patterns in different renal tumor histologies. Retrospective analysis of 51 patients (23 pRCC-Type 1/28 pRCC-Type 2) who underwent multiphase CT followed by surgery from July 2011 to April 2016 was performed. Data were analyzed between subgroups based on histology. Multiphase CT was analyzed for tumor size, and attenuation [Hounsfield Units (HU)]. Change in HU (ΔHU) was calculated between noncontrast (NC), corticomedullary (CM), nephrographic (N), and delayed (D) phases. Subset analysis was carried out on patients who underwent PRMB prior to surgery. There was no difference in median tumor size (pRCC-Type 1 2.8 vs. pRCC-Type 2 2.6 cm, p=0.832). In addition to tumor size being similar between groups, distribution of tumor stages between groups was also similar (p=0.651). Greater proportion of high-grade tumors (III/IV) was noted in pRCC-Type 2 (42.9% vs. 8.7%) (p=0.011). There was no difference in HU values for NC (p=0.961), CM (p=0.118), N (p=0.277), and D (p=0.256) phases, and in ΔHU between CM-NC (p=0.278), N-NC (p=0.316), and D-NC (p=0.103). Thirteen patients underwent percutaneous biopsy, 11 of whom had diagnostic samples. Examination of 10/11 (90.9%) samples accurately predicted correct histology, and of 6/11 (54.5%) samples correctly identified high-vs. low-grade histology. Our findings suggest substantial overlap of CT findings, despite pRCC-Type 2 having greater proportion of high-grade tumors. Utility of CT is limited in the differentiation between pRCC subtypes. Patients with suggested pRCC on CT imaging being considered for a non-extirpative strategy should undergo PRMB for risk stratification.