Risk Stratification for Oropharyngeal Squamous Cell Carcinoma Using Texture Analysis on CT - A Step Beyond HPV Status.

Background and Purpose: Human papillomavirus-associated oropharyngeal squamous cell carcinoma (OPSCC) is increasingly prevalent. Despite the overall more favorable outcome, the observed heterogeneous treatment response within this patient group highlights the need for additional means to prognosticate and guide clinical decision-making. Promising prediction models using radiomics from primary OPSCC have been derived. However, no model/s using metastatic lymphadenopathy exist to allow prognostication in those instances when the primary tumor is not seen. The aim of our study was to evaluate whether radiomics using metastatic lymphadenopathy allows for the development of a useful risk assessment model comparable to the primary tumor and whether additional knowledge of the HPV status further improves its prognostic efficacy. Materials and Methods: 80 consecutive patients diagnosed with stage III-IV OPSCC between February 2009 and October 2015, known human papillomavirus status, and pre-treatment CT images were retrospectively identified. Manual segmentation of primary tumor and metastatic lymphadenopathy was performed and the extracted texture features were used to develop multivariate assessment models to prognosticate treatment response. Results: Texture analysis of either the primary or metastatic lymphadenopathy from pre-treatment enhanced CT images can be used to develop models for the stratification of treatment outcomes in OPSCC patients. AUCs range from .78 to .85 for the various OPSCC groups tested, indicating high predictive capability of the models. Conclusions: This preliminary study can form the basis multi-centre trial that may help optimize treatment and improve quality of life in patients with OPSCC in the era of personalized medicine.

Acute stroke imaging selection for mechanical thrombectomy in the extended time window: is it time to go back to basics? A review of current evidence.

Treatment with endovascular therapy in the extended time window for acute ischaemic stroke with large vessel occlusion involves stringent selection criteria based on the two landmark studies DAWN and DEFUSE3. Current protocols typically include the requirement of advanced perfusion imaging which may exclude a substantial proportion of patients from receiving a potentially effective therapy. Efforts to offer endovascular reperfusion therapies to all appropriate candidates may be facilitated by the use of simplified imaging selection paradigms with widely available basic imaging techniques, such as non-contrast CT and CT angiography. Currently available evidence from our literature review suggests that patients meeting simplified imaging selection criteria may benefit as much as those patients selected using advanced imaging techniques (CT perfusion or MRI) from endovascular therapy in the extended time window. A comprehensive understanding of the role of imaging in patient selection is critical to optimising access to endovascular therapy in the extended time window and improving outcomes in acute stroke. This article provides an overview on current developments and future directions in this emerging area.

Radiogenomic Models Using Machine Learning Techniques to Predict EGFR Mutations in Non-Small Cell Lung Cancer.

BACKGROUND: The purpose of this study was to build radiogenomics models from texture signatures derived from computed tomography (CT) and 18F-FDG PET-CT (FDG PET-CT) images of non-small cell lung cancer (NSCLC) with and without epidermal growth factor receptor (EGFR) mutations. METHODS: Fifty patients diagnosed with NSCLC between 2011 and 2015 and with known EGFR mutation status were retrospectively identified. Texture features extracted from pretreatment CT and FDG PET-CT images by manual contouring of the primary tumor were used to develop multivariate logistic regression (LR) models to predict EGFR mutations in exon 19 and exon 20. RESULTS: An LR model evaluating FDG PET-texture features was able to differentiate EGFR mutant from wild type with an area under the curve (AUC), sensitivity, specificity, and accuracy of 0.87, 0.76, 0.66, and 0.71, respectively. The model derived from CT texture features had an AUC, sensitivity, specificity, and accuracy of 0.83, 0.84, 0.73, and 0.78, respectively. FDG PET-texture features that could discriminate between mutations in EGFR exon 19 and 21 demonstrated AUC, sensitivity, specificity, and accuracy of 0.86, 0.84, 0.73, and 0.78, respectively. Based on CT texture features, the AUC, sensitivity, specificity, and accuracy were 0.75, 0.81, 0.69, and 0.75, respectively. CONCLUSION: Non-small cell lung cancer texture analysis using FGD-PET and CT images can identify tumors with mutations in EGFR. Imaging signatures could be valuable for pretreatment assessment and prognosis in precision therapy.

Use of Noncontrast Computed Tomography and Computed Tomographic Perfusion in Predicting Intracerebral Hemorrhage After Intravenous Alteplase Therapy.

BACKGROUND AND PURPOSE: Intracerebral hemorrhage is a feared complication of intravenous alteplase therapy in patients with acute ischemic stroke. We explore the use of multimodal computed tomography in predicting this complication. METHODS: All patients were administered intravenous alteplase with/without intra-arterial therapy. An age- and sex-matched case-control design with classic and conditional logistic regression techniques was chosen for analyses. Outcome was parenchymal hemorrhage on 24- to 48-hour imaging. Exposure variables were imaging (noncontrast computed tomography hypoattenuation degree, relative volume of very low cerebral blood volume, relative volume of cerebral blood flow ≤7 mL/min·per 100 g, relative volume of Tmax ≥16 s with all volumes standardized to z axis coverage, mean permeability surface area product values within Tmax ≥8 s volume, and mean permeability surface area product values within ipsilesional hemisphere) and clinical variables (NIHSS [National Institutes of Health Stroke Scale], onset to imaging time, baseline systolic blood pressure, blood glucose, serum creatinine, treatment type, and reperfusion status). RESULTS: One-hundred eighteen subjects (22 patients with parenchymal hemorrhage versus 96 without, median baseline NIHSS score of 15) were included in the final analysis. In multivariable regression, noncontrast computed tomography hypoattenuation grade (P<0.006) and computerized tomography perfusion white matter relative volume of very low cerebral blood volume (P=0.04) were the only significant variables associated with parenchymal hemorrhage on follow-up imaging (area under the curve, 0.73; 95% confidence interval, 0.63-0.83). Interrater reliability for noncontrast computed tomography hypoattenuation grade was moderate (κ=0.6). CONCLUSIONS: Baseline hypoattenuation on noncontrast computed tomography and very low cerebral blood volume on computerized tomography perfusion are associated with development of parenchymal hemorrhage in patients with acute ischemic stroke receiving intravenous alteplase.

Arterial spin labelling reveals multi-regional cerebral hypoperfusion in patients with transient ischemic attack that are unrelated to ischemia location: A proof-of-concept study.

Background and Aims: Patients with transient ischemic attack (TIA) have a substantially increased risk of early dementia. In this exploratory study, we aim to determine whether patients with TIA have 1) measurable regional cerebral hypoperfusion unrelated to the location of ischemia, and 2) determine the relationship of regional cerebral blood flow (rCBF) with their cognitive profiles. Methods: Patients with TIA (N = 49) and seventy-nine (N = 79) age and sex matched controls underwent formal neuropsychological testing and MRI. Quantitative arterial spin labelling rCBF maps (mL/min/100 g) were registered to the corresponding high resolution T1-weighted image. Linear regression was used to determine the association between demographic, clinical and cognitive variables and rCBF. Results: Patients with TIA had significantly (p < 0.05) lower cognitive scores in the MMSE, MOCA, ACE-R, WAIS-IV DS Coding and Trail Making Tests A and B compared to controls. TIA patients had significantly lower rCBF in the left entorhinal cortex (p = 0.03), right posterior cingulate (p = 0.04), and right precuneus (p = 0.05), after adjusting for age and sex, that were unrelated to the regional anatomical volume and DWI positivity. Regional hypoperfusion in the right posterior cingulate and right precuneus was associated with impaired visual memory (BVMT total, p = 0.05 for both regions) and slower processing speed (TMT A, p = 0.04 and p = 0.01), respectively after adjusting for age and sex. Conclusions: TIA patients have patterns of regional hypoperfusion in multiple cortical regions unrelated to the parcellated regional anatomical volume or the presence of a DWI lesion. Regional hypoperfusion in patients with TIA may be an early marker conferring risk of future cognitive decline that needs to be confirmed by future studies.

Identifying Thrombus on Non-Contrast CT in Patients with Acute Ischemic Stroke.

The hyperdense sign is a marker of thrombus in non-contrast computed tomography (NCCT) datasets. The aim of this work was to determine optimal Hounsfield unit (HU) thresholds for thrombus segmentation in thin-slice non-contrast CT (NCCT) and use these thresholds to generate 3D thrombus models. Patients with thin-slice baseline NCCT (≤2.5 mm) and MCA-M1 occlusions were included. CTA was registered to NCCT, and three regions of interest (ROIs) were placed in the NCCT, including: the thrombus, contralateral brain tissue, and contralateral patent MCA-M1 artery. Optimal HU thresholds differentiating the thrombus from non-thrombus tissue voxels were calculated using receiver operating characteristic analysis. Linear regression analysis was used to predict the optimal HU threshold for discriminating the clot only based on the average contralateral vessel HU or contralateral parenchyma HU. Three-dimensional models from 70 participants using standard (45 HU) and patient-specific thresholds were generated and compared to CTA clot characteristics. The optimal HU threshold discriminating thrombus in NCCT from other structures varied with a median of 51 (IQR: 49-55). Experts chose 3D models derived using patient-specific HU models as corresponding better to the thrombus seen in CTA in 83.8% (31/37) of cases. Patient-specific HU thresholds for segmenting the thrombus in NCCT can be derived using normal parenchyma. Thrombus segmentation using patient-specific HU thresholds is superior to conventional 45 HU thresholds.