Non-enhanced Computed Tomography Images Research Articles

Quantitative analysis of lung lesions using unenhanced chest computed tomography images.

Chest radiograph and computed tomography (CT) scans can accidentally reveal pulmonary nodules. Malignant and benign pulmonary nodules can be difficult to distinguish without specific imaging features, such as calcification, necrosis, and contrast enhancement. However, these lesions may exhibit different image texture characteristics which cannot be assessed visually. Thus, a computer-assisted quantitative method like histogram analysis (HA) of Hounsfield unit (HU) values can improve diagnostic accuracy, reducing the need for invasive biopsy. In this exploratory control study, nonenhanced chest CT images of 20 patients with benign (10) and cancerous (10) lesion were selected retrospectively. The appearances of benign and malignant lesions were very similar in chest CT images, and only pathology report was used to discriminate them. Free hand region of interest (ROI) was inserted inside the lesion for all slices of each lesion. Mean, minimum, maximum, and standard deviations of HU values were recorded and used to make HA. HA showed that the most malignant lesions have a mean HU value between 30 and 50, a maximum HU less than 150, and a minimum HU between -30 and 20. Lesions outside these ranges were mostly benign. Quantitative CT analysis may differentiate malignant from benign lesions without specific malignancy patterns on unenhanced chest CT image.

Radiomics-based prediction of nonalcoholic fatty liver disease following pancreatoduodenectomy.

Predicting nonalcoholic fatty liver disease (NAFLD) following pancreaticoduodenectomy (PD) is challenging, which delays therapeutic intervention and makes its prevention difficult. We conducted this study to assess the potential application of preoperative computed tomography (CT) radiomics for predicting NAFLD. The subjects of this retrospective study were 186 patients with PD from a single institution. We extracted the predictors of NAFLD after PD statistically from conventional clinical and radiomic features of the estimated remnant pancreas and whole liver region on preoperative nonenhanced CT images. Based on these predictors, we developed a machine-learning predictive model, which integrated clinical and radiomic features. A comparative model used only clinical features as predictors. The incidence of NAFLD after PD was 43.5%. The variables of the clinicoradiomic model included one shape feature of the pancreas, two texture features of the liver, and sex; the variables of the clinical model were age, sex, and chemoradiotherapy. The accuracy%, precision%, recall%, F1 score, and area under the curve of the two models were 75.0, 72.7, 66.7, 69.6, and 0.80; and 69.6, 68.4, 54.2, 60.5, and 0.69, respectively. Preoperative CT-derived radiomic features from the pancreatic and liver regions are promising for the prediction of NAFLD post-PD. Using these features enhances the predictive model, enabling earlier intervention for high-risk patients.

Computed Tomography-Based Radiomics Analysis of Different Machine Learning Approaches for Differentiating Pulmonary Sarcomatoid Carcinoma and Pulmonary Inflammatory Pseudotumor

Background: Differentiating between pulmonary sarcomatoid carcinoma (PSC) and pulmonary inflammatory pseudotumor (PIP) is challenging using current conventional diagnostic methods. This lack of distinction significantly impacts subsequent clinical treatment decisions. Objectives: This study was conducted to construct an effective method to distinguish between PSC and PIP based on commonly used computed tomography (CT) images. Patients and Methods: A total of 14 patients with PSC and 76 patients with PIP were retrospectively included in the study for CT imaging. Radiomics features were extracted from non-enhanced CT images, and canonical correlation analysis was performed to reduce redundancy. The final radiomics signature was then identified using the least absolute shrinkage and selection operator (LASSO). Logistic regression (LR), classification and regression trees (CART), support vector machine (SVM), k-nearest neighbors (KNN), and gradient boosting machine (GBM) were used to construct the radiomics models. The performance of these different radiomics models was evaluated using the receiver operating characteristic curve. Results: A total of 1186 radiomics features were extracted from non-enhanced CT images. After dimensionality reduction and selection, 7 valuable features were identified. The performance of 5 machine learning models was evaluated to differentiate between PSC and PIP, and the GBM-based radiomics model demonstrated the best performance. The GBM-based radiomics model achieved an accuracy of 0.922, area under the curve (AUC) of 0.98, F1 score of 0.967, and log loss of 0.161. Compared to conventional clinical-radiological diagnosis, the GBM-based radiomics model showed a significant association (odds ratio [OR] = 8.119; P = 0.006). Conclusion: The implementation of the GBM-based radiomics model has the potential to improve the ability to differentiate between PSC and PIP, thereby influencing the timeliness of subsequent surgical interventions and even the prognosis of patients.

A CT-based radiomics nomogram involving the cystic fluid area for differentiating appendiceal mucinous neoplasms from appendicitis with intraluminal fluid

ObjectiveTo develop and validate a radiomics nomogram based on computed tomography (CT) to distinguish appendiceal mucinous neoplasms (AMNs) from appendicitis with intraluminal fluid (AWIF).MethodA total of 211 patients from two medical institutions were retrospectively analysed, of which 109 were pathologically confirmed as having appendicitis with concomitant CT signs of intraluminal fluid and 102 as having AMN. All patients were randomly assigned to a training (147 patients) or validation cohort (64 patients) at a 7:3 ratio. Radiomics features of the cystic fluid area of the appendiceal lesions were extracted from nonenhanced CT images using 3D Slicer software. Minimum redundancy maximum relevance and least absolute shrinkage and selection operator regression methods were employed to screen the radiomics features and develop a radiomics model. Combined radiomics nomogram and clinical-CT models were further developed based on the corresponding features selected after multivariate analysis. Lastly, receiver operating characteristic curves, and decision curve analysis (DCA) were used to assess the models’ performances in the training and validation cohorts.ResultsA total of 851 radiomics features were acquired from the nonenhanced CT images. Subsequently, a radiomics model consisting of eight selected features was developed. The combined radiomics nomogram model comprised rad-score, age, and mural calcification, while the clinical-CT model contained age and mural calcification. The combined model achieved area under the curves (AUCs) of 0.945 (95% confidence interval [CI]: 0.895, 0.976) and 0.933 (95% CI: 0.841, 0.980) in the training and validation cohorts, respectively, which were larger than those obtained by the radiomics (training cohort: AUC, 0.915 [95% CI: 0.865, 0.964]; validation cohort: AUC, 0.912 [95% CI: 0.843, 0.981]) and clinical-CT models (training cohort: AUC, 0.884 [95% CI: 0.820, 0.931]; validation cohort: AUC, 0.767 [95% CI: 0.644, 0.863]). Finally, DCA showed that the clinical utility of the combined model was superior to that of the clinical CT and radiomics models.ConclusionOur combined radiomics nomogram model constituting radiomics, clinical, and CT features exhibited good performance for differentiating AMN from AWIF, indicating its potential application in clinical decision-making.

CT-based nomogram for early identification of T790M resistance in metastatic non-small cell lung cancer before first-line epidermal growth factor receptor-tyrosine kinase inhibitors therapy

BackgroundTo evaluate the value of computed tomography (CT) radiomics in predicting the risk of developing epidermal growth factor receptor (EGFR) T790M resistance mutation for metastatic non-small lung cancer (NSCLC) patients before first-line EGFR-tyrosine kinase inhibitors (EGFR-TKIs) therapy.MethodsA total of 162 metastatic NSCLC patients were recruited and split into training and testing cohort. Radiomics features were extracted from tumor lesions on nonenhanced CT (NECT) and contrast-enhanced CT (CECT). Radiomics score (rad-score) of two CT scans was calculated respectively. A nomogram combining two CT scans was developed to evaluate T790M resistance within up to 14 months. Patients were followed up to calculate the time of T790M occurrence. Models were evaluated by area under the curve at receiver operating characteristic analysis (ROC-AUC), calibration curve, and decision curve analysis (DCA). The association of the nomogram with the time of T790M occurrence was evaluated by Kaplan–Meier survival analysis.ResultsThe nomogram constructed with the rad-score of NECT and CECT for predicting T790M resistance within 14 months achieved the highest ROC-AUCs of 0.828 and 0.853 in training and testing cohorts, respectively. The DCA showed that the nomogram was clinically useful. The Kaplan–Meier analysis showed that the occurrence time of T790M difference between the high- and low-risk groups distinguished by the rad-score was significant (p < 0.001).ConclusionsThe CT-based radiomics signature may provide prognostic information and improve pretreatment risk stratification in EGFR NSCLC patients before EGFR-TKIs therapy. The multimodal radiomics nomogram further improved the capability.Relevance statementRadiomics based on NECT and CECT images can effectively identify and stratify the risk of T790M resistance before the first-line TKIs treatment in metastatic non-small cell lung cancer patients.Key points• Early identification of the risk of T790M resistance before TKIs treatment is clinically relevant.• Multimodel radiomics nomogram holds potential to be a diagnostic tool.• It provided an imaging surrogate for identifying the pretreatment risk of T790M.Graphical

Automated Intracranial Clot Detection: A Promising Tool for Vascular Occlusion Detection in Non-Enhanced CT.

(1) Background: to test the diagnostic performance of a fully convolutional neural network-based software prototype for clot detection in intracranial arteries using non-enhanced computed tomography (NECT) imaging data. (2) Methods: we retrospectively identified 85 patients with stroke imaging and one intracranial vessel occlusion. An automated clot detection prototype computed clot location, clot length, and clot volume in NECT scans. Clot detection rates were compared to the visual assessment of the hyperdense artery sign by two neuroradiologists. CT angiography (CTA) was used as the ground truth. Additionally, NIHSS, ASPECTS, type of therapy, and TOAST were registered to assess the relationship between clinical parameters, image results, and chosen therapy. (3) Results: the overall detection rate of the software was 66%, while the human readers had lower rates of 46% and 24%, respectively. Clot detection rates of the automated software were best in the proximal middle cerebral artery (MCA) and the intracranial carotid artery (ICA) with 88-92% followed by the more distal MCA and basilar artery with 67-69%. There was a high correlation between greater clot length and interventional thrombectomy and between smaller clot length and rather conservative treatment. (4) Conclusions: the automated clot detection prototype has the potential to detect intracranial arterial thromboembolism in NECT images, particularly in the ICA and MCA. Thus, it could support radiologists in emergency settings to speed up the diagnosis of acute ischemic stroke, especially in settings where CTA is not available.

Usefulness of three-dimensional pulmonary vein-left atrium image reconstructed from non-enhanced computed tomography for atrial fibrillation ablation.

Enhanced computed tomography (CT) is unsuitable for patients with reduced renal function and/or allergy for contrast medium (CM). CT image registration into an electroanatomic system (EAMS) is essential to perform pulmonary vein isolation (PVI) safely and smoothly in patients with atrial fibrillation (AF). To create three-dimensional pulmonary vein-left atrium (3D PV-LA) images from non-enhanced CT images to register them into EAMS for AF ablation. Using a non-enhanced ECG-gated image, 3D PV-LA images were generated by our developed techniques with an EnSite image analyzing tool for patients unfit for CM use (n = 100). Segmentation between tissues was performed as follows: tissues distal from or close to PV-LA were segmented in transverse slices to clearly show the whole LA. Tissues bordering PV-LA, including the pulmonary artery, left ventricle, and right atrium, were segmented manually with great care. Practical ablation parameters were compared with those obtained from enhanced CT (n = 100). 3D PV-LA image reconstruction from non-enhanced CT imaging required a longer time than that from enhanced CT (42 ± 6 vs 14 ± 3 min). All 100 PV-LA non-enhanced CT images were successfully reconstructed and registered into the EAM system without the need for re-segmentation. Practical ablation parameters, including procedural time and AF recurrence rate, did not differ between imaging methods. This study provides clinically useful information on a detailed methodology for 3D PV-LA image reconstruction using non-enhanced CT. Non-enhanced CT 3D PV-LA images were successfully registered into the EAM system and useful for patients unsuitable for CM use.

CT radiomics model for predicting the Ki-67 proliferation index of pure-solid non-small cell lung cancer: a multicenter study.

This study aimed to explore the efficacy of the computed tomography (CT) radiomics model for predicting the Ki-67 proliferation index (PI) of pure-solid non-small cell lung cancer (NSCLC). This retrospective study included pure-solid NSCLC patients from five centers. The radiomics features were extracted from thin-slice, non-enhanced CT images of the chest. The minimum redundancy maximum relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) were used to reduce and select radiomics features. Logistic regression analysis was employed to build predictive models to determine Ki-67-high and Ki-67-low expression levels. Three prediction models were established: the clinical model, the radiomics model, and the nomogram model combining the radiomics signature and clinical features. The prediction efficiency of different models was evaluated using the area under the curve (AUC). A total of 211 NSCLC patients with pure-solid nodules or masses were included in the study (N=117 for the training cohort, N=49 for the internal validation cohort, and N=45 for the external validation cohort). The AUC values for the clinical models in the training, internal validation, and external validation cohorts were 0.73 (95% CI: 0.64-0.82), 0.75 (95% CI:0.62-0.89), and 0.72 (95% CI: 0.57-0.86), respectively. The radiomics models showed good predictive ability in diagnosing Ki-67 expression levels in the training cohort (AUC, 0.81 [95% CI: 0.73-0.89]), internal validation cohort (AUC, 0.81 [95% CI: 0.69-0.93]) and external validation cohort (AUC, 0.78 [95% CI: 0.64-0.91]). Compared to the clinical and radiomics models, the nomogram combining both radiomics signatures and clinical features had relatively better diagnostic performance in all three cohorts, with the AUC of 0.83 (95% CI: 0.76-0.90), 0.83 (95% CI: 0.71-0.94), and 0.81 (95% CI: 0.68-0.93), respectively. The nomogram combining the radiomics signature and clinical features may be a potential non-invasive method for predicting Ki-67 expression levels in patients with pure-solid NSCLC.

Dual-phase computed tomography for localization of parathyroid lesions in children and adolescents with primary hyperparathyroidism

BackgroundChildhood and adolescent primary hyperparathyroidism (PHPT) is a rare disease caused by single adenomas in 65–94% of patients. In this patient group, there is no data on computed tomography (CT) for pre-operative parathyroid localization that may facilitate focused parathyroidectomy. MethodsTwo radiologists reviewed dual-phase (nonenhanced and arterial) CT images of twenty-three operated children and adolescents [20:single-gland disease(SGD), 3:multi-glandular disease(MGD)] with proven histopathological PHPT. Percentage arterial enhancement (PAE) was calculated as [100*{arterial-phase Hounsfield unit (HU)-nonenhanced phase HU}/nonenhanced HU] of the parathyroid lesion(s), thyroid, and lymph node. ResultsDual-phase CT lateralized 100%, localized to the correct quadrant/site 85% SGD (including 3/3 ectopic), and identified 1/3 MGD. PAE (cutoff ≥ 112.3%) was sensitive (91.3%) and specific (99.5%) in distinguishing parathyroid lesions from local mimics (P<0.001). The average effective dose was 3.16±1.01mSv, comparable to the planar/single photon emission CT (SPECT) Technetium 99m(Tc)-sestamibi and choline positron emission tomography (PET)/CT scans. Solid-cystic morphology identified in 4 patients harboring pathogenic germline variants (3:CDC73, 1:CASR) may serve as a radiological clue to molecular diagnosis. Nineteen out of 20 (95%) patients with SGD who had undergone single gland resection based on pre-operative CT findings were in remission over a median follow-up of 18 months. ConclusionAs most children/adolescents with PHPT have SGD, dual-phase CT protocols which reduce the effective radiation dose with high localization sensitivity for single parathyroid lesions may be a sustainable pre-operative imaging modality in this patient group.

A combined non-enhanced CT radiomics and clinical variable machine learning model for differentiating benign and malignant sub-centimeter pulmonary solid nodules.

Radiomics has been used to predict pulmonary nodule (PN) malignancy. However, most of the studies focused on pulmonary ground-glass nodules. The use of computed tomography (CT) radiomics in pulmonary solid nodules, particularly sub-centimeter solid nodules, is rare. This study aims to develop a radiomics model based on non-enhanced CT images that can distinguish between benign and malignant sub-centimeter pulmonary solid nodules (SPSNs, <1cm). The clinical and CT data of 180 SPSNs confirmed by pathology were analyzed retrospectively. All SPSNs were divided into two groups: training set (n=144) and testing set (n=36). From non-enhanced chest CT images, over 1000 radiomics features were extracted. Radiomics feature selection was performed using the analysis of variance and principal component analysis. The selected radiomics features were fed into a support vector machine (SVM) to develop a radiomics model. The clinical and CT characteristics were used to develop a clinical model. Associating non-enhanced CT radiomics features with clinical factors were used to develop a combined model using SVM. The performance was evaluated using the area under the receiver-operating characteristic curve (AUC). The radiomics model performed well in distinguishing between benign and malignant SPSNs, with an AUC of 0.913 (95% confidence interval [CI], 0.862-0.954) in the training set and an AUC of 0.877 (95% CI, 0.817-0.924) in the testing set. The combined model outperformed the clinical and radiomics models with an AUC of 0.940 (95% CI, 0.906-0.969) in the training set and an AUC of 0.903 (95% CI, 0.857-0.944) in the testing set. Radiomics features based on non-enhanced CT images can be used to differentiate SPSNs. The combined model, which included radiomics and clinical factors, had the best discrimination power between benign and malignant SPSNs.

Development and validation of a radiomics-based nomogram for the prediction of postoperative malnutrition in stage IB1-IIA2 cervical carcinoma.

In individuals with stage IB1-IIA2 cervical cancer (CC) who received postoperative radiotherapy ± chemotherapy (PORT/CRT), the interaction between sarcopenia and malnutrition remains elusive, let alone employing a nomogram model based on radiomic features of psoas extracted at the level of the third lumbar vertebra (L3). This study was set to develop a radiomics-based nomogram model to predict malnutrition as per the Patient-Generated Subjective Global Assessment (PG-SGA) for individuals with CC. In total, 120 individuals with CC underwent computed tomography (CT) scans before PORT/CRT. The radiomic features of psoas at L3 were obtained from non-enhanced CT images. Identification of the optimal features and construction of the rad-score formula were conducted utilizing the least absolute shrinkage and selection operator (LASSO) logistic regression to predict malnutrition in the training dataset (radiomic model). Identification of the major clinical factors in the clinical model was performed by means of binary logistic regression analysis. The radiomics-based nomogram was further developed by integrating radiomic signatures and clinical risk factors (combined model). The receiver operating characteristic (ROC) curves and decision curves analysis (DCA) were employed for the evaluation and comparison of the three models in terms of their predictive performance. Twelve radiomic features in total were chosen, and the rad-score was determined with the help of the non-zero coefficient from LASSO regression. Multivariate analysis revealed that besides rad-score, age and Eastern Cooperative Oncology Group performance status could independently predict malnutrition. As per the data of this analysis, a nomogram prediction model was constructed. The area under the ROC curves (AUC) values of the radiomic and clinical models were 0.778 and 0.847 for the training and 0.776 and 0.776 for the validation sets, respectively. An increase in the AUC was observed up to 0.972 and 0.805 in the training and validation sets, respectively, in the combined model. DCA also confirmed the clinical benefit of the combined model. This radiomics-based nomogram model depicted potential for use as a marker for predicting malnutrition in stage IB1-IIA2 CC patients who underwent PORT/CRT and required further investigation with a large sample size.

Deep Learning Models for Severity Prediction of Acute Pancreatitis in the Early Phase From Abdominal Nonenhanced Computed Tomography Images.

To develop and validate deep learning (DL) models for predicting the severity of acute pancreatitis (AP) by using abdominal nonenhanced computed tomography (CT) images. The study included 978 AP patients admitted within 72 hours after onset and performed abdominal CT on admission. The image DL model was built by the convolutional neural networks. The combined model was developed by integrating CT images and clinical markers. The performance of the models was evaluated by using the area under the receiver operating characteristic curve. The clinical, Image DL, and the combined DL models were developed in 783 AP patients and validated in 195 AP patients. The combined models possessed the predictive accuracy of 90.0%, 32.4%, and 74.2% for mild, moderately severe, and severe AP. The combined DL model outperformed clinical and image DL models with 0.820 (95% confidence interval, 0.759-0.871), the sensitivity of 84.76% and the specificity of 66.67% for predicting mild AP and the area under the receiver operating characteristic curve of 0.920 (95% confidence interval, 0.873-0.954), the sensitivity of 90.32%, and the specificity of 82.93% for predicting severe AP. The DL technology allows nonenhanced CT images as a novel tool for predicting the severity of AP.

A Nomogram Prediction Model for Internal Hernia Using Clinical Parameters and Non-Enhanced Computed Tomography Imaging

A Nomogram Prediction Model for Internal Hernia Using Clinical Parameters and Non-Enhanced Computed Tomography Imaging

Prediction of Central Lymph Node Metastasis in cN0 Papillary Thyroid Carcinoma by CT Radiomics

To explore the feasibility of the preoperative prediction of pathological central lymph node metastasis (CLNM) status in patients with negative clinical lymph node (cN0) papillary thyroid carcinoma (PTC) using a computed tomography (CT) radiomics signature. A total of 97 PTC cN0 nodules with CLNM pathology data (pN0, with CLNM, n=59; pN1, without CLNM, n=38) in 85 patients were divided into a training set (n=69) and a validation set (n=28). For each lesion, 321 radiomic features were extracted from nonenhanced, arterial and venous phase CT images. Minimum redundancy and maximum relevance and the least absolute shrinkage and selection operator were used to find the most important features with which to develop a radiomics signature in the training set. The performance of the radiomics signature was evaluated by receiver operating characteristic curves, calibration curves and decision curve analysis . Three nonzero the least absolute shrinkage and selection operator coefficient features were selected for radiomics signature construction. The radiomics signature for distinguishing the pN0 and pN1 groups achieved areas under the curve of 0.79 (95% CI 0.67, 0.91) in the training set and 0.77 (95% CI 0.55, 0.99) in the validation set. The calibration curves demonstrated good agreement between the radiomics score-predicted probability and the pathological results in the two sets (p= 0.399, p=0.191). The decision curve analysis curves showed that the model was clinically useful. This radiomic signature could be helpful to predict CLNM status in cN0 PTC patients.

Evaluation of radiological and temporal characteristics of acute appendicitis on the non-enhanced computed tomography images.

To investigate the relationships and interactions between temporal and radiological features of gangrene and perforation of inflamed appendices. A total of 402 patients were included who underwent laparoscopic appendectomies between January 1, 2016 and March 30, 2020 and had pathologically proved acute appendicitis and preoperative non-enhanced CT examinations. The radiological features (appendix diameter, appendicolith, appendiceal intraluminal gas, periappendiceal gas, periappendiceal fat stranding/fluid, and short axial diameter of the mesenteric lymph nodes) were obtained from the preoperative CT images of 382 patients with visible appendices. Clinical parameters and temporal variables (pre-CT delay, preoperative delay, estimated complication delay, symptom delay, and system delay) were recorded. Among simple/suppurative, gangrenous, and perforated appendicitis, the radiological characteristics except for short axial diameters of lymph nodes, and the temporal variables other than system delay were significantly different. The Cox regression analysis identified the appendicolith as the independent risk factor for both gangrene and perforation of inflamed appendices by using the preoperative delay or estimated complication delay. By the preoperative delay, the median time for gangrene and perforation was 76.23 (95%CI 73.89-78.58) h and 77.55 (95%CI 74.12-80.98) h, respectively, if appendicolith was present. If estimated complication delay was used as the elapsed time and the appendicolith was perceptible, the median time for gangrene and perforation and was 72.33 (95%CI 62.93-81.74) h and 75.07 (95%CI 69.48-80.65) h, respectively. There were interactions between the time evolution and radiological features of acute appendicitis. The evaluation of gangrene and perforation rate of acute appendicitis could be benefitted from combining the preoperative delay/estimated complication delay with CT characteristics in the preoperative urgent radiological analysis.

Value of nonenhanced CT combined with laboratory examinations in the diagnosis of acute suppurative cholecystitis treated with percutaneous cholecystostomy: a retrospective study

PurposesIn this study, we aimed to identify the distribution of presenting laboratory and nonenhanced computed tomography (CT) imaging features within 48 h before percutaneous cholecystostomy (PC) and create a model to appropriately guide the diagnosis of acute suppurative cholecystitis (ASC).MethodsThe study population included 204 acute cholecystitis patients who underwent PC. Based on the timing of the last laboratory and CT examinations before PC, the patients were divided into two groups: within 48 h before PC (Group 1, n = 138) and over 48 h before PC (Group 2, n = 63). The clinical features of the ASC patients in the two groups were compared. A multivariable model for the diagnosis of ASC in the patients in Group 1 was developed.ResultsThirty-nine patients in Group 1 had ASC (28.3%). Gallbladder stones, common bile duct stones, gallbladder wall thickness > 2.85 mm, and neutrophil granulocytes > 82.55% were confirmed to be independent risk factors for ASC. The receiver operating characteristic curve of the recurrence prediction model verified its accuracy (area under the curve: 0.803). Compared with the ASC patients in Group 2, the ASC patients in Group 1 had a higher proportion of pericholecystic exudation or fluid (P = 0.013) and thicker gallbladder walls (P = 0.033).ConclusionsUsing nonenhanced CT imaging features and cutoffs for neutrophil granulocytes, we were able to identify a simple algorithm to discriminate ASC. The degree of local inflammation of the gallbladder in ASC patients progressively increases over time, and these changes can be observed on nonenhanced CT images. However, the symptoms of abdominal pain are of little help in estimating the disease duration in elderly patients.

Is infarct core growth linear? Infarct volume estimation by computed tomography perfusion imaging.

Current guidelines for recanalization treatment are based on the time elapsed between symptom onset and treatment and visualization of existing penumbra in computed tomography perfusion (CTP) imaging. The time window for treatment options relies on linear growth of infarction although individual infarct growth rate may vary. We aimed to test how accurately the estimated follow-up infarct volume (eFIV) can be approximated by using a linear growth model based on CTP baseline imaging. If eFIV did not fall within the margins of +/- 19% of the follow-up infarct volume (FIV) measured at 24h from non-enhanced computed tomography images, the results would imply that the infarct growth is not linear. All consecutive endovascularly treated (EVT) patients from 11/2015 to 9/2019 at the Helsinki University Hospital with large vessel occlusion (LVO), CTP imaging, and known time of symptom onset were included. Infarct growth rate was assumed to be linear and calculated by dividing the ischemic core volume (CTPcore ) by the time from symptom onset to baseline imaging. eFIV was calculated by multiplying the infarct growth rate with the time from baseline imaging to recanalization or in case of futile recanalization to follow-up imaging at 24h, limited to the penumbra. Collateral flow was estimated by calculating hypoperfusion intensity ratio (HIR). Of 5234 patients, 48had LVO, EVT, CTP imaging, and known time of symptom onset. In 40/48 patients (87%), infarct growth was not linear. HIR did not differ between patients with linear and nonlinear growth (p>.05). As expected, in over half of the patients with successful recanalization eFIV exceeded FIV. Infarct growth was not linear in most patients and thus time elapsed from symptom onset and CTPcore appear to be insufficient parameters for clinical decision-making in EVT candidates.

Abstract WP108: Is Infarct Core Growth Truly Linear? Follow-up Infarct Volume Estimation By Rapid Baseline Infarct Growth Rate And Linear Model

Background: Current guidelines for recanalization treatment are based on the time window between symptom onset and treatment in addition to ischemic core and perfusion lesion volumes by computed tomography perfusion imaging (CTP). Linear growth of infarction is commonly assumed.The aim was to test, whether measured follow-up infract volume (FIV) could be approximated from the linear growth model (eFIV) based on CTP baseline infarct growth rate.We assumed the infarct growth to stop, when recanalization was achieved or when the eFIV reached the volume of the perfusion lesion (T max &gt;6s volume). Methods: All consecutive stroke code patients from 11/2015-9/2019 transferred to Helsinki University Hospital as candidates for endovascular treatment (EVT) were screened; patients with large vessel occlusion (LVO), EVT, CTP and known time of symptom onset were included to study.The infarct growth rate was calculated by dividing the CTP core by the time from symptom onset to baseline imaging.eFIV was calculated by infarct growth rate multiplied with the time from baseline imaging to recanalization or follow-up imaging. We assumed a performance of +/- 19% for the accuracy of the CTP core assessment. FIV was measured from the 24h non-enhanced computed tomography images. Recanalization was defined as modified Treatment in Cerebral Infarction (mTICI) scale as successful (TICI 2b or 3) or futile (TICI 0,1,2a). Results: Out of 5234 patients, 48 had LVO and EVT, CTP imaging and known time of symptom onset (Figure 1). In 40/48 (83%) patients, infarct growth was not within the 19% margins of linear growth. eFIV exceeded FIV in 25/42 patients with successful recanalization (median absolute difference 25 mL,7-73). Conclusions: eFIV from linearly approximated growth model did not support linear growth of the infarct.

Machine Learning Based Non-Enhanced CT Radiomics for the Identification of Orbital Cavernous Venous Malformations: An Innovative Tool.

To evaluate the capability of non-enhanced computed tomography (CT) images for distinguishing between orbital cavernous venous malformations (OCVM) and non-OCVM, and to identify the optimal model from radiomics-based machine learning (ML) algorithms. A total of 215 cases of OCVM and 120 cases of non- OCVM were retrospectively analyzed in this study. A stratified random sample of 268 patients (80%) was used as the training set (172 OCVM and 96 non-OCVM); the remaining data were used as the testing set. Six feature selection techniques and thirteen ML models were evaluated to construct an optimal classification model. There were statistically significant differences between the OCVM and non-OCVM groups in the density and tumor location (P < 0.05), whereas other indicators were comparable (age, gender, sharp, P > 0.05). Linear regression (area under the curve [AUC] = 0.9351; accuracy = 0.8657) and Stochastic Gradient Descent (AUC = 0.9448; accuracy = 0.8806) classifiers, both of which coupled with the f test and L1-based feature selection method, achieved optimal performance. The support vector machine (AUC = 0.9186; accuracy = 0.8806), Random Forest (AUC = 0.9288; accuracy = 0.8507) and eXtreme Gradient Boosting (AUC = 0.9147; accuracy = 0.8507) classifier combined with f test method showed excellent average performance among our study, respectively. The effect of non-enhanced CT images in OCVM not only can help ophthalmologist to find and locate lesion, but also bring great help for the qualitative diagnosis value using radiomic- based ML algorithms.

Deep-learning single-shot detector for automatic detection of brain metastases with the combined use of contrast-enhanced and non-enhanced computed tomography images

PurposeTo develop a deep-learning object detection model for automatic detection of brain metastases that simultaneously uses contrast-enhanced and non-enhanced images as inputs, and to compare its performance with that of a model that uses only contrast-enhanced images. MethodA total of 116 computed tomography (CT) scans of 116 patients with brain metastases were included in this study. They showed a total of 659 metastases, 428 of which were used for training and validation (mean size, 11.3 ± 9.9 mm) and 231 were used for testing (mean size, 9.0 ± 7.0 mm). Single-shot detector (SSD) models were constructed with a feature fusion module, and their results were compared per lesion at a confidence threshold of 50%. ResultsThe sensitivity was 88.7% for the model that used both contrast-enhanced and non-enhanced CT images (the CE + NECT model) and 87.6% for the model that used only contrast-enhanced CT images (the CECT model). The positive predictive value (PPV) was 44.0% for the CE + NECT model and 37.2% for the CECT model. The number of false positives per patient was 9.9 for the CE + NECT model and 13.6 for the CECT model. The CE + NECT model had a significantly higher PPV (t test, p < 0.001), significantly fewer false positives (t test, p < 0.001), and a tendency to be more sensitive (t test, p = 0.14). ConclusionsThe results indicate that the information on true contrast enhancement obtained by comparing the contrast-enhanced and non-enhanced images may prevent the detection of pseudolesions, suppress false positives, and improve the performance of deep-learning object detection models.