Non-contrast CTs Research Articles

Targeting ventricular tachycardia originating from the anterolateral papillary muscle with stereotactic body radioablation therapy

Abstract Background Ventricular Tachycardia (VT) is prevalent in the setting of non-ischaemic cardiomyopathy (NICM), a heterogeneous group of conditions characterized by myocardial dysfunction in the absence of coronary artery disease. Challenges of catheter ablation to treat VT in NICM patients are well known, and recurrence rates can be as high as 58%. Stereotactic body radioablation therapy (SBRT) involves precise delivery of radiation to arrhythmogenic substrates, and represents a novel treatment to reduce recurrent implantable cardioverter defibrillator (ICD) therapies in patients with refractory VT. Though SBRT has shown preliminary effectiveness in reducing VT episodes, its effectiveness in NICM patients has not been widely evaluated. Purpose To evaluate SBRT effectiveness in NICM patients with scars at the basal anterolateral papillary muscle at a tertiary care center between November-December 2022. Methods Eligible patients had their VT substrate mapped using a single-use ECGI vest and a 4D-CT simulation scan. In addition, MRI scans were used to aid anatomical identification of the anterolateral papillary muscle which could not be seen on the non-contrast diagnostic CTs. SBRT was delivered in a single fraction of 25 Gy under image guidance. Effectiveness of the treatment was evaluated as changes in frequencies of total antitachycardia pacing (ATP) and ICD shock episodes from 6 months pre-, and within to 3, 6, and 12 months follow up post SBRT. Results Participants were 3 males with NICM, aged 56, 60, and 73 with previous dual chamber ICD implantation, 2 of whom had a history of VT ablation/ VT storm and paroxysmal atrial fibrillation. All participants had basal lateral scar, and were receiving beta blockers, mexiletine, and amiodarone with refractory VT before SBRT treatment. Prior to SBRT, ICD device interrogation showed a total of 119 (77, 7, and 35 in patients 1-3 respectively) ATPs, and 27 (10, 1, and 16 in patients 1-3 respectively) ICD shocks delivered. At 12 month follow-up, none of the participants experienced any episodes of ATP or ICD shock. Ejection fraction remained consistent pre-post SBRT. Conclusion SBRT showed remarkable success in reducing ATP and ICD shock episodes, facilitating uncomplicated 1 year follow-up in 3 male patients with NICM and VT from the anterolateral papillary muscle.

Increased risk for thoracic and abdominal aortic aneurysm in the relatives of abdominal aorta aneurysm patients

Abstract Introduction Genetic susceptibilities in aneurysm genes are risk factors for developing aneurysms of the aorta. This translates to a high aneurysm prevalence in relatives of aortic aneurysm patients. For relatives of abdominal aorta aneurysm (AAA) patients, guidelines recommend abdominal ultrasound screening for early detection and management, while for relatives of thoracic aorta aneurysm patients (TAA), transthoracic ultrasound of the ascending aorta and aortic root is recommended. Neither of these guidelines are designed to detect aneurysms in the whole aorta. The incidence of TAA in relatives of AAA index-patients is unknown and may determine whether abdominal ultrasound screening is sufficient for relatives of AAA patients or whether whole aorta screening should be recommended. Furthermore, focused screening of high risk relatives of aneurysm patients may improve the yield of aneurysm detection now that population frequencies are declining. Methods Whole aorta non-contrast CTs of relatives of AAA index-patients were collected retrospectively in a tertiary referral (2016 to 2021). Family history and DNA testing of AAA index-patients were used to classify relatives into high genetic risk (familial disease and/or pathogenic/likely pathogenic variant (P/LP) in an aneurysm gene) or unknown genetic risk (no familial disease or no P/LP detected). TAA were defined as ascending diameters ≥ 45mm or descending diameters ≥ 40mm. Thoracic aortic dilatations (TAD) were defined as ascending diameters of 40-45 mm or descending diameters of 35-40 mm. Abdominal diameters ≥30mm were classified as AAA and diameters between 25-29 mm as abdominal dilatations (AAD). References 1-3. Results CT scans (n=301 relatives, 47% male, age 58(50-65)) showed 34% aortic aneurysms and/or dilatations versus 14% in the general population (age adjusted, p<0.001). TAA was 28-times more frequent (2%, p<0001) and AAA 3 times more frequent (8%, p<0.001) than in the general population. Thoracic and abdominal dilatations occurred in 18% and 17% of relatives. Overall, males were 4 times more likely to have an aneurysm than females (p<0.05). Aneurysms and dilatations occurred at the same rate in relatives with or without a family history or a P/LP in an aneurysm gene. These results endorse total aorta screening for relatives of all AAA patients rather than restricting screening to the abdominal aorta. Conclusions Relatives of AAA patients had a high risk for thoracic and abdominal aneurysms as well as dilatations. This indicates that imaging of the whole aorta will improve early detection of aortic aneurysms in relatives both with and without a family history of aneurysms.

Artificial Intelligence Efficacy as a Function of Trainee Interpreter Proficiency: Lessons from a Randomized Controlled Trial.

Recently, artificial intelligence tools have been deployed with increasing speed in educational and clinical settings. However, the use of artificial intelligence by trainees across different levels of experience has not been well-studied. This study investigates the impact of artificial intelligence assistance on the diagnostic accuracy for intracranial hemorrhage and large-vessel occlusion by medical students and resident trainees. This prospective study was conducted between March 2023 and October 2023. Medical students and resident trainees were asked to identify intracranial hemorrhage and large-vessel occlusion in 100 noncontrast head CTs and 100 head CTAs, respectively. One group received diagnostic aid simulating artificial intelligence for intracranial hemorrhage only (n = 26); the other, for large-vessel occlusion only (n = 28). Primary outcomes included accuracy, sensitivity, and specificity for intracranial hemorrhage/large-vessel occlusion detection without and with aid. Study interpretation time was a secondary outcome. Individual responses were pooled and analyzed with the t test; differences in continuous variables were assessed with ANOVA. Forty-eight participants completed the study, generating 10,779 intracranial hemorrhage or large-vessel occlusion interpretations. With diagnostic aid, medical student accuracy improved 11.0 points (P < .001) and resident trainee accuracy showed no significant change. Intracranial hemorrhage interpretation time increased with diagnostic aid for both groups (P < .001), while large-vessel occlusion interpretation time decreased for medical students (P < .001). Despite worse performance in the detection of the smallest-versus-largest hemorrhages at baseline, medical students were not more likely to accept a true-positive artificial intelligence result for these more difficult tasks. Both groups were considerably less accurate when disagreeing with the artificial intelligence or when supplied with an incorrect artificial intelligence result. This study demonstrated greater improvement in diagnostic accuracy with artificial intelligence for medical students compared with resident trainees. However, medical students were less likely than resident trainees to overrule incorrect artificial intelligence interpretations and were less accurate, even with diagnostic aid, than the artificial intelligence was by itself.

APIS: a paired CT-MRI dataset for ischemic stroke segmentation - methods and challenges

Stroke, the second leading cause of mortality globally, predominantly results from ischemic conditions. Immediate attention and diagnosis, related to the characterization of brain lesions, play a crucial role in patient prognosis. Standard stroke protocols include an initial evaluation from a non-contrast CT to discriminate between hemorrhage and ischemia. However, non-contrast CTs lack sensitivity in detecting subtle ischemic changes in this phase. Alternatively, diffusion-weighted MRI studies provide enhanced capabilities, yet are constrained by limited availability and higher costs. Hence, we idealize new approaches that integrate ADC stroke lesion findings into CT, to enhance the analysis and accelerate stroke patient management. This study details a public challenge where scientists applied top computational strategies to delineate stroke lesions on CT scans, utilizing paired ADC information. Also, it constitutes the first effort to build a paired dataset with NCCT and ADC studies of acute ischemic stroke patients. Submitted algorithms were validated with respect to the references of two expert radiologists. The best achieved Dice score was 0.2 over a test study with 36 patient studies. Despite all the teams employing specialized deep learning tools, results reveal limitations of computational approaches to support the segmentation of small lesions with heterogeneous density.

Deep Learning to Detect Intracranial Hemorrhage in a National Teleradiology Program and the Impact on Interpretation Time.

The diagnostic performance of an artificial intelligence (AI) clinical decision support solution for acute intracranial hemorrhage (ICH) detection was assessed in a large teleradiology practice. The impact on radiologist read times and system efficiency was also quantified. A total of 61 704 consecutive noncontrast head CT examinations were retrospectively evaluated. System performance was calculated along with mean and median read times for CT studies obtained before (baseline, pre-AI period; August 2021 to May 2022) and after (post-AI period; January 2023 to February 2024) AI implementation. The AI solution had a sensitivity of 75.6%, specificity of 92.1%, accuracy of 91.7%, prevalence of 2.70%, and positive predictive value of 21.1%. Of the 56 745 post-AI CT scans with no bleed identified by a radiologist, examinations falsely flagged as suspected ICH by the AI solution (n = 4464) took an average of 9 minutes 40 seconds (median, 8 minutes 7 seconds) to interpret as compared with 8 minutes 25 seconds (median, 6 minutes 48 seconds) for unremarkable CT scans before AI (n = 49 007) (P < .001) and 8 minutes 38 seconds (median, 6 minutes 53 seconds) after AI when ICH was not suspected by the AI solution (n = 52 281) (P < .001). CT scans with no bleed identified by the AI but reported as positive for ICH by the radiologist (n = 384) took an average of 14 minutes 23 seconds (median, 13 minutes 35 seconds) to interpret as compared with 13 minutes 34 seconds (median, 12 minutes 30 seconds) for CT scans correctly reported as a bleed by the AI (n = 1192) (P = .04). With lengthened read times for falsely flagged examinations, system inefficiencies may outweigh the potential benefits of using the tool in a high volume, low prevalence environment. Keywords: Artificial Intelligence, Intracranial Hemorrhage, Read Time, Report Turnaround Time, System Efficiency Supplemental material is available for this article. © RSNA, 2024.

Aneurysmal subarachnoid haemorrhage: Volumetric quantification of the blood distribution pattern to accurately predict the ruptured aneurysm location

BackgroundIn spontaneous subarachnoid haemorrhage (SAH) accurate determination of the bleeding source is paramount to guide treatment. Traditionally, the bleeding pattern has been used to predict the aneurysm location. Here, we have tested a software-based tool, which quantifies the volume of intracranial blood and stratifies it according to the regional distribution, to predict the location of the ruptured aneurysm. MethodsA consecutive series of SAH patients admitted to a single tertiary centre between 2012–2018, within 72 h of onset, harbouring a single intracranial aneurysm. A semi-automatized method of blood quantification, based on the relative density increase, was applied to initial non-contrast CTs. Five regions were used to define the bleeding patterns and to correlate them with aneurysm location: perimesencephalic, interhemispheric, right/left hemisphere and intraventricular. Results68 patients were included for analysis. There was a strong association between the distribution of blood and the aneurysm location (p < 0.001). In particular: ACom and interhemispheric fissure (p < 0.001), MCA and ipsilateral hemisphere (p < 0.001), ICA and ipsilateral hemisphere and perimesencephalic cisterns (p < 0.001), PCom and hemispheric, perimesencephalic and intraventricular (p = 0.019), and PICA and perimesencephalic and intraventricular (p < 0.001). The internal diagnostic value was high (AUROC ≥ 0.900) for these locations. ConclusionRegional automatised volumetry seems a reliable and objective tool to quantify and describe the distribution of blood within the subarachnoid spaces. This tool accurately predicts the location of the ruptured aneurysm; its use may be prospectively considered in the emergency setting when speed and simplicity are attained.

Abstract TP204: Performance of Rapidai NCCT Stroke Software in Colombia's Early Stroke System: Preliminary Results

Background: Automated stroke solutions in developing countries have the potential to reduce healthcare gaps. This study assesses the performance of the RAPID NCCT Stroke module compared to local neuroradiologists’ evaluation of acute stroke. Methods: Rapid NCCT stroke is a new multi-module AI tool that simultaneously evaluates non-contrast CT scans (NCCT) for ICH (any subtype, volume &gt;0.4 ml), the ASPECT score, and suspicion of LVO (i.e., intracranial ICA or horizontal segment of the MCA). The software determines suspicion of LVO based on an algorithm that evaluates the hyperdense MCA sign and early ischemic changes in the MCA territory. Consensus readings were defined by three blinded neuroradiologists based on a subset of Non-contrast CTs obtained between January 2013 to December 2018 at Fundacion Santa Fe De Bogota University Hospital. The overall diagnostic performance of the RAPID NCCT was compared to the assessments made by the neuroradiologists. Results: 179 total stroke cases with technically adequate NCCT scans were identified. Based on the consensus of the neuroradiologists, these cases included 151 ischemic strokes, 27 ICHs &gt;0.4 ml, 1 ICH &lt;0.4 ml, and 28 LVO cases (verified by CT angiography). The Rapid NCCT Stroke module demonstrated high specificity (96%), NPV (98%), and accuracy (95%) for the detection of ICH. For determining the ASPECT score, the software had a specificity of 93% and an NPV of 76%. Seventy percent of the CTA-verified LVOs were detected from NCCT alone with a specificity and NPV of 82%. Conclusions: The RAPID NCCT Stroke software displayed promising results for the early identification of ICH and characterization of ischemic strokes. The software identified 70% of the LVOs from NCCT alone. This tool can potentially enhance patient outcomes by bridging healthcare gaps in evolving stroke systems. Multicenter studies in additional Latin American countries are needed for further validation.

Aneurysmal subarachnoid haemorrhage: Volumetric quantification of the blood distribution pattern to accurately predict the ruptured aneurysm location

BackgroundIn spontaneous subarachnoid haemorrhage (SAH) accurate determination of the bleeding source is paramount to guide treatment. Traditionally, the bleeding pattern has been used to predict the aneurysm location. Here, we have tested a software-based tool, which quantifies the volume of intracranial blood and stratifies it according to the regional distribution, to predict the location of the ruptured aneurysm. MethodsA consecutive series of SAH patients admitted to a single tertiary centre between 2012–2018, within 72 h of onset, harbouring a single intracranial aneurysm. A semi-automatized method of blood quantification, based on the relative density increase, was applied to initial non-contrast CTs. Five regions were used to define the bleeding patterns and to correlate them with aneurysm location: perimesencephalic, interhemispheric, right/left hemisphere and intraventricular. Results68 patients were included for analysis. There was a strong association between the distribution of blood and the aneurysm location (p < 0.001). In particular: ACom and interhemispheric fissure (p < 0.001), MCA and ipsilateral hemisphere (p < 0.001), ICA and ipsilateral hemisphere and perimesencephalic cisterns (p < 0.001), PCom and hemispheric, perimesencephalic and intraventricular (p = 0.019), and PICA and perimesencephalic and intraventricular (p < 0.001). The internal diagnostic value was high (AUROC ≥ 0.900) for these locations. ConclusionRegional automatised volumetry seems a reliable and objective tool to quantify and describe the distribution of blood within the subarachnoid spaces. This tool accurately predicts the location of the ruptured aneurysm; its use may be prospectively considered in the emergency setting when speed and simplicity are attained.

Enhancing hospital course and outcome prediction in patients with traumatic brain injury: A machine learning study.

We aimed to use machine learning (ML) algorithms with clinical, lab, and imaging data as input to predict various outcomes in traumatic brain injury (TBI) patients. In this retrospective study, blood samples were analyzed for glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase L1 (UCH-L1). The non-contrast head CTs were reviewed by two neuroradiologists for TBI common data elements (CDE). Three outcomes were designed to predict: discharged or admitted for further management (prediction 1), deceased or not deceased (prediction 2), and admission only, prolonged stay, or neurosurgery performed (prediction 3). Five ML models were trained. SHapley Additive exPlanations (SHAP) analyses were used to assess the relative significance of variables. Four hundred forty patients were used to predict predictions 1 and 2, while 271 patients were used in prediction 3. Due to Prediction 3's hospitalization requirement, deceased and discharged patients could not be utilized. The Random Forest model achieved an average accuracy of 1.00 for prediction 1 and an accuracy of 0.99 for prediction 2. The Random Forest model achieved a mean accuracy of 0.93 for prediction 3. Key features were extracranial injury, hemorrhage, UCH-L1 for prediction 1; TheGlasgow Coma Scale, age, GFAP for prediction 2; and GFAP, subdural hemorrhage volume, and pneumocephalus for prediction 3, per SHAP analysis. Combining clinical and laboratory parameters with non-contrast CT CDEs allowed our ML models to accurately predict the designed outcomes of TBI patients. GFAP and UCH-L1 were among the significant predictor variables, demonstrating the importance of these biomarkers.

Establishment of Age- and Sex-Specific Reference Cerebral Ventricle Volumes

The diagnosis of hydrocephalus is dependent on clinical symptoms and radiographic findings including ventriculomegaly. Our goal was to generate a data set of ventricular volume utilizing non-pathologic computed tomography (CT) scans for adults to help define reference ventricle size. We performed a retrospective analysis of non-contrast head CTs for adults at a single institution to identify patients who had undergone imaging and did not have a diagnosis of hydrocephalus, history of ventriculoperitoneal shunting, or treatments for hydrocephalus. A convolutional neural network was trained on hand-segmented scans from a variety of age ranges and then utilized to automate the segmentation of the entire data set. Ventricles on 866 CT scans were segmented to generate a reference range of volumes for both male and female individuals ranging in age from 18-99 years. The generated data were binned by age ranges. We have developed a convolutional neural network that can segment the ventricles on CT scans of adult patients over a range of ages. This network was used to measure the ventricular volume of non-pathologic head CTs to produce reference ranges for several age bins. This data set could be utilized to aid in the diagnosis of hydrocephalus by comparing potentially pathologic scans to reference ventricular volumes.

Incorporating algorithmic uncertainty into a clinical machine deep learning algorithm for urgent head CTs.

Machine learning (ML) algorithms to detect critical findings on head CTs may expedite patient management. Most ML algorithms for diagnostic imaging analysis utilize dichotomous classifications to determine whether a specific abnormality is present. However, imaging findings may be indeterminate, and algorithmic inferences may have substantial uncertainty. We incorporated awareness of uncertainty into an ML algorithm that detects intracranial hemorrhage or other urgent intracranial abnormalities and evaluated prospectively identified, 1000 consecutive noncontrast head CTs assigned to Emergency Department Neuroradiology for interpretation. The algorithm classified the scans into high (IC+) and low (IC-) probabilities for intracranial hemorrhage or other urgent abnormalities. All other cases were designated as No Prediction (NP) by the algorithm. The positive predictive value for IC+ cases (N = 103) was 0.91 (CI: 0.84-0.96), and the negative predictive value for IC- cases (N = 729) was 0.94 (0.91-0.96). Admission, neurosurgical intervention, and 30-day mortality rates for IC+ was 75% (63-84), 35% (24-47), and 10% (4-20), compared to 43% (40-47), 4% (3-6), and 3% (2-5) for IC-. There were 168 NP cases, of which 32% had intracranial hemorrhage or other urgent abnormalities, 31% had artifacts and postoperative changes, and 29% had no abnormalities. An ML algorithm incorporating uncertainty classified most head CTs into clinically relevant groups with high predictive values and may help accelerate the management of patients with intracranial hemorrhage or other urgent intracranial abnormalities.

A Deep Learning Approach to Visualize Aortic Aneurysm Morphology Without the Use of Intravenous Contrast Agents.

Intravenous contrast agents are routinely used in CT imaging to enable the visualization of intravascular pathology, such as with abdominal aortic aneurysms. However, the injection is contraindicated in patients with iodine allergy and is associated with renal complications. In this study, we investigate if the raw data acquired from a noncontrast CT image contains sufficient information to differentiate blood and other soft tissue components. A deep learning pipeline underpinned by generative adversarial networks was developed to simulate contrast enhanced CTA images using noncontrast CTs. Two generative models (cycle- and conditional) are trained with paired noncontrast and contrast enhanced CTs from seventy-five patients (total of 11,243 pairs of images) with abdominal aortic aneurysms in a 3-fold cross-validation approach with a training/testing split of 50:25 patients. Subsequently, models were evaluated on an independent validation cohort of 200 patients (total of 29,468 pairs of images). Both deep learning generative models are able to perform this image transformation task with the Cycle-generative adversarial network (GAN) model outperforming the Conditional-GAN model as measured by aneurysm lumen segmentation accuracy (Cycle-GAN: 86.1% ± 12.2% vs Con-GAN: 85.7% ± 10.4%) and thrombus spatial morphology classification accuracy (Cycle-GAN: 93.5% vs Con-GAN: 85.7%). This pipeline implements deep learning methods to generate CTAs from noncontrast images, without the need of contrast injection, that bear strong concordance to the ground truth and enable the assessment ofimportant clinical metrics. Our pipeline is poised to disrupt clinical pathways requiring intravenous contrast.

The Effect of an Iodinated Contrast Material Shortage on Ordering Patterns in the Emergency Department.

Introduction Iodinated contrast media (ICM) is essential to emergency department care in differentiating and delineating life-threatening pathologies. In May 2022, due to the COVID-19 outbreak, there was an unprecedented disruption in the manufacturing of iodinated contrast. The primary goal of this study was to describe the effects of an ICM shortage on the ordering patterns of emergency medicine physicians. Methods This was a retrospective, observational study with a pre-/post-test design. The study included two 28-day periods. All subjects who underwent a CT were included in the study. The subgroup of patients who underwent a repeat CT with ICM contrast within 1-24 hours was identified. Results During the pre- and post-implementation study periods, 4,574 and 3,973 CT studies were performed. The median length of stay (p=0.013) and time to first CT (p<0.001) both decreased during the post-implementation period. During the post-implementation period, more non-contrast CTs were ordered (p<0.001). During the post-implementation period, there was an increase in non-contrast studies followed by a repeat study with contrast (p=0.003). Conclusions A global ICM shortage resulted in a shift in the ordering patterns of Emergency Medicine (EM) physicians. More non-contrast CT scans were ordered. However, there was also an increase in repeat imaging with ICM material.

The Accuracy 0f CT Chest in Determining Between Ground Glass Opacities And Those Caused by Covid-19 And Other Medications

Introduction: Diagnostic ct scanning of the chest is currently being investigated for its ability to distinguish between ground glass opacities (GGO) caused by coronavirus 2019 (COVID-19) and GGO produced by other causes. Place and Duration: From January 2022 until June 2022, I will work as a Radiologist at Qazi Hussain Ahmad Hospital in Nowshera. Methods: This study was cross sectional study carried out at the Qazi Hussain Ahmad Hospital, Nowshera for a period of six months. The overall sample size in the current study was100 non-contrast chest CTs. Eexperienced radiologists analyzed the CT images of the chest after redacting any personal information. laboratory results and the patient's medical history were noted. Results: The participants comprised 46 people with COVID19 and 100 without COVID19 who also had ground glass opacities on chest CT. There was no statistically significant difference in age between the groups (p-value = 0.212). Out of the non-COVID-19 GGO cases, three patients have hypersensitivity pneumonia in 3, eosinophilic pneumonia in 3, interstitial pneumonia in 7, pulmonary pneumonia in 3, pulmonary fibrosis in 7, drug-induced lung damage in 7, pulmonary alveolar hemorrhage in 3, and pulmonary emphysema in 11. Practical implication: This study will provide physician with the data to compare the likelihood that COVID-19 causes ground glass opacities on a chest CT scan versus the likelihood that they are caused by other probable causes Conclusion: Moreover, the specificity of chest CT in differentiating COVID-19 from viral pneumonia is only intermediate, and the specificity of chest CT in distinguishing COVID-19 from other reasons of ground glass opacities is poor. Keywords: CT Chest, Determining, Ground-Glass Opacities, Covid-19, Medications

Deep-learning-based hepatic fat assessment (DeHFt) on non-contrast chest CT and its association with disease severity in COVID-19 infections: A multi-site retrospective study.

BackgroundHepatic steatosis (HS) identified on CT may provide an integrated cardiometabolic and COVID-19 risk assessment. This study presents a deep-learning-based hepatic fat assessment (DeHFt) pipeline for (a) more standardised measurements and (b) investigating the association between HS (liver-to-spleen attenuation ratio <1 in CT) and COVID-19 infections severity, wherein severity is defined as requiring invasive mechanical ventilation, extracorporeal membrane oxygenation, death.MethodsDeHFt comprises two steps. First, a deep-learning-based segmentation model (3D residual-UNet) is trained (N = 80) to segment the liver and spleen. Second, CT attenuation is estimated using slice-based and volumetric-based methods. DeHFt-based mean liver and liver-to-spleen attenuation are compared with an expert's ROI-based measurements. We further obtained the liver-to-spleen attenuation ratio in a large multi-site cohort of patients with COVID-19 infections (D1, N = 805; D2, N = 1917; D3, N = 169) using the DeHFt pipeline and investigated the association between HS and COVID-19 infections severity.FindingsThe DeHFt pipeline achieved a dice coefficient of 0.95, 95% CI [0.93–0.96] on the independent validation cohort (N = 49). The automated slice-based and volumetric-based liver and liver-to-spleen attenuation estimations strongly correlated with expert's measurement. In the COVID-19 cohorts, severe infections had a higher proportion of patients with HS than non-severe infections (pooled OR = 1.50, 95% CI [1.20–1.88], P < .001).InterpretationThe DeHFt pipeline enabled accurate segmentation of liver and spleen on non-contrast CTs and automated estimation of liver and liver-to-spleen attenuation ratio. In three cohorts of patients with COVID-19 infections (N = 2891), HS was associated with disease severity. Pending validation, DeHFt provides an automated CT-based metabolic risk assessment.FundingFor a full list of funding bodies, please see the Acknowledgements.

Trends of Acute Renal Colic During COVID-19 Lockdown: An Experience From Saudi Arabia.

IntroductionThe COVID-19 pandemic represents an unprecedented challenge for healthcare systems around the world. Saudi Arabia was one of the first countries to experience a lockdown and postponement of elective surgical procedures. The objective of this study was to assess the trends of acute renal colic presenting to our emergency department.MethodsThis retrospective study targeted all patients who presented with acute renal colic during the lockdown period (March 23, 2019 to June 20, 2019). Patients’ and stone data were collected. The patient’s data included age, gender, BMI, and comorbidities. Stones' data included stone size, location, side, evidence of obstruction and UTI, and planned and conducted management.ResultsA total of 137 patients were identified; 92 (67.2%) patients were males with a mean age of 44 ± 16 years. Positive history of urolithiasis was reported in 47 (34.3%). The most common initial investigation was non-contrast CTs (93.4%). The majority of patients had a stone size of < 10 mm (93%) and ureteric stones (81.2%). A total of 32 patients (32.4%) had evidence of UTI and 63.4% had evidence of obstruction. Most of the patients (73.7%) were offered medical expulsive therapy (MET). Only 2.2% did not receive the planned management.ConclusionThe observed pattern shows that the management during the lockdown did not differ from the original recommendations. This could be due to the fact that most patients had stone sizes between 5 and 10 mm and consequently were managed by METs. Larger data need to be conducted to provide concrete evidence. Such data are relevant to provide a clear guide for management and to establish protocols for emergency lockdown situations.

The analysis of the prevalence, importance and reporting rates of incidental lung lesions in non-contrast abdominal CTs performing for the evaluation of the urinary system

The analysis of the prevalence, importance and reporting rates of incidental lung lesions in non-contrast abdominal CTs performing for the evaluation of the urinary system

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.

Ordering Characteristics Predictive of Noncontrast CT Head Positivity in the Emergency Department

Recent decades have seen a steady increase in noncontrast head CT utilization in the emergency department with a concurrent rise in the practice of physician assistants (PAs) and nurse practitioners (NPs). The goal of this study was to identify ordering and patient characteristics predictive of positive noncontrast head CTs in the ED. We hypothesized NP/PAs would have lower positivity rates compared to physicians, suggestive of relative overutilization. We retrospectively identified ED patients who underwent noncontrast head CTs at a single institution: a nonlevel 1 trauma center, during a 7-year period, recording examination positivity, ordering provider training/experience, and multiple additional ordering/patient attributes. Exam positivity was defined as any intracranial abnormality necessitating a change in acute management, such as acute hemorrhage, hydrocephalus, herniation, or worsening prior findings. 6624 patients met inclusion criteria. 4.6% (280/6107) of physician exams were positive while 3.7% (19/517) of NP/PA exams were positive; however, differences were not significant. Increasing provider experience was not associated with positivity. Attributes with increased positivity were patient age (p < 0.001), daytime exam (p < 0.05), and indications regarding malignancy (p < 0.001) or focal neurologic deficit (p=0.001). Attributes with decreased positivity were indications of trauma (p < 0.001) or vertigo/dizziness (p < 0.05). We found no significant difference in rates of exam positivity between physicians and NP/PAs, even accounting for years of experience. This suggests increasing utilization of head CTs in the ED is not due to the increasing presence of NP/PAs, and may be reflective of general practice trends and clear diagnostic algorithms leading to head CT.

Labeling Noncontrast Head CT Reports for Common Findings Using Natural Language Processing.

Prioritizing reading of noncontrast head CT examinations through an automated triage system may improve time to care for patients with acute neuroradiologic findings. We present a natural language-processing approach for labeling findings in noncontrast head CT reports, which permits creation of a large, labeled dataset of head CT images for development of emergent-finding detection and reading-prioritization algorithms. In this retrospective study, 1002 clinical radiology reports from noncontrast head CTs collected between 2008 and 2013 were manually labeled across 12 common neuroradiologic finding categories. Each report was then encoded using an n-gram model of unigrams, bigrams, and trigrams. A logistic regression model was then trained to label each report for every common finding. Models were trained and assessed using a combination of L2 regularization and 5-fold cross-validation. Model performance was strongest for the fracture, hemorrhage, herniation, mass effect, pneumocephalus, postoperative status, and volume loss models in which the area under the receiver operating characteristic curve exceeded 0.95. Performance was relatively weaker for the edema, hydrocephalus, infarct, tumor, and white-matter disease models (area under the receiver operating characteristic curve > 0.85). Analysis of coefficients revealed finding-specific words among the top coefficients in each model. Class output probabilities were found to be a useful indicator of predictive error on individual report examples in higher-performing models. Combining logistic regression with n-gram encoding is a robust approach to labeling common findings in noncontrast head CT reports.