The historical development, current practice, and the future of interventional neuroradiology are intricately linked to the advancements in the imaging and devices used for neuroendovascular treatments. This thesis explores the advanced imaging potential of the C-arm imaging systems used in the neurointerventional suite and investigates the initial clinical experience with a new flow diverter device to treat the intracranial aneurysms. A cohort of aneurysmal SAH patients who developed delayed cerebral ischaemia (DCI) were prospectively studied with a new parenchymal blood volume (PBV) research protocol C-arm CT examination concurrent with a magnetic resonance (MR) imaging examination that included perfusion and diffusion weighted sequences. Using a robust quantitative volume-of-interest analysis, it was demonstrated that C-arm CT PBV measurements are in agreement with MR-PWI CBV and CBF, and the PBV represents a composite perfusion parameter with both blood-flow (≈60%) and blood-volume (≈40%) weightings. Subsequently, using a voxel-wise ROC curve analysis and MR-DWI, it was shown that using optimal thresholds, C-arm CT PBV measurements allow reliable demarcation of the irreversibly infarcted parenchyma. For evaluation of ischaemic parenchyma, the PBV measurements were reliable for moderate-to-severe ischaemia but were prone to underestimate the mild-to-moderate ischaemia. A catalogue of reference mean PBV measurements was then created for various anatomical regions encompassing the whole brain after excluding any locations with ongoing ischaemia or infarction. Next, using an ROI-based analysis of the C-arm CT projection data, steady-state contrast concentration assumption underlying the PBV calculations was investigated. It was demonstrated that for clinical scans, the ideal steady-state assumption is not fully met, however, for a large majority of C-arm CT examinations the temporal characteristics of TDCs closely approximate the expected ideal steady-state. The degree to which the TDC of a C-arm CT scan approximates the ideal steady-state was found to influence the resulting PBV measurements and their agreement to MR-CBV. Moreover, the temporal characteristics of TDCs showed inter-subject variation. Finally, the C-arm CT cross-sectional soft tissue images were demonstrated to be of adequate quality for the assessment of ventricles and for the detection of procedural vessel rupture. These findings advance the understanding of the nature of PBV parameter, establish the optimal PBV thresholds for infarction, provide reference PBV measurements, and highlight the limitations of C-arm CT PBV imaging. The work is of considerable clinical significance and has implications for implementation of C-arm CT PBV imaging in the interventional suite for management of patients with acute brain ischaemia. In regards to the initial clinical experience with the flow diversion treatment of intracranial aneurysms, the procedural, angiographic, and clinical outcomes were studied. Several pertinent technical and clinical issues were highlighted for this new treatment approach. Based on the observations made during this work, a new grading schema was then developed to monitor the angiographic outcomes after flow diversion treatment. Using the angiographic data for patients treated with FD, the new grading schema was demonstrated to be sufficiently sensitive to register gradual aneurysm occlusion and evaluate parent artery patency, with an excellent inter-rater reliability and applicability to various aneurysm morphologies. This work (largest multi-centre series at the time of its publication) informed the interventional neuroradiology community about the safety, efficacy, and outcomes of flow diversion treatment. Additionally, it provided a sensitive and reliable scale to evaluate the angiographic outcomes after flow diversion treatment, in both research and clinical practice.

The objective of this study is to determine the diagnostic performance of computed tomography perfusion (CTP) with and without computed tomography angiography (CTA) in assessment of hemodynamically significant coronary artery lesions in comparison to invasive fractional flow reserve (FFR). Materials and Methods PubMed and Cochrane Center Register of Controlled Trials from January 2010 searched through December 2014. Nine original studies were selected evaluating the diagnostic performance of CTP with and without CTA to invasive coronary angiography in evaluation of hemodynamic significance of coronary lesions (n = 951). Results The sensitivity, specificity, LR+ and LR- and DOR of CTA+CTP were 0.85 [95% confidence interval (CI: 0.79-0.89)] 0.94 (CI: 0.91-0.97), 15.8 (CI: 7.99-31.39), 0.146 (CI: 0.08-0.26), and 147.2 (CI: 69.77-310.66). Summary Receiver Operating Characteristics (SROC) results showed area under the curve (AUC) of 0.97 indicating that CTA+CTP may detect hemodynamically significant coronary artery lesions with high accuracy. The sensitivity, specificity, LR+ and LR- and DOR of CTP were 0.83 (CI: 0.78-0.87), 0.84 (CI: 0.80-0.87) 5.26 (CI: 2.93-9.43), 0.209 (CI: 0.12-0.36), and 31.97 (CI: 11.59-88.20). Conclusions This result suggests that CTP with CTA significantly improves diagnostic performance of coronary artery lesions compared to CTA alone and closely comparable with invasive FFR.

Le diagnostic de mort cérébrale est une étape complexe et chronophage lors de l'évaluation des patients en soins intensifs soupçonnés d'être en décès neurologique. Bien que les critères neurologiques cliniques qui déterminent la mort cérébrale soient largement acceptés dans le monde, le diagnostic reste imparfait et l'utilisation de tests auxiliaires tels que la perfusion tomographique cérébrale (CTP) est souvent nécessaire pour le confirmer. L'objectif principal de ce travail était d'explorer la faisabilité de classer la mort cérébrale à partir de scans CTP par le traitement automatique de l’image. Les scans CTP de l'étude prospective canadienne multicentrique de validation du CTP pour le diagnostic de décès neurologique ont été regroupées à partir de 11 sites participants (INDex-CTP, ClinicalTrials.gov, NCT03098511). Des caractéristiques spatiales et temporelles ont été extraites en utilisant une combinaison de deux modules de convolution et utilisées pour prédire la mort neurologique. Les performances du modèle ont également été évaluées sur différentes catégories de blessures cérébrales. Les études de 217 patients ont été utilisées pour entraîner le modèle. Nous rapportons une AUC de 0,79 (IC95 % 0,76-0,82), un score F1 de 0,82 (IC95 % 0,80-0,83), une précision de 0,92 (IC95 % 0,91-0,93), un rappel de 0,76 (CI95 % 0,72-0,79) ainsi qu'une valeur prédictive négative de 0,49 (CI95 % 0,45-0,53). En raison de la petite taille d'échantillon, nous n'avons pas effectué de tests statistiques sur des sous-ensembles de lésions cérébrales, mais avons signalé une valeur prédictive négative du modèle présumé plus élevée sur des blessures cérébrales anoxiques avec 0,82 (CI95 % 0,77-0,87). Ce modèle montre des preuves préliminaires soutenant la faisabilité de développer un réseau neuronal profond pour classer les patients comateux comme étant neurologiquement décédés ou non. Des recherches supplémentaires sont nécessaires pour valider et améliorer le modèle en utilisant des ensembles de données plus vastes et diversifiés. / The diagnostic of brain death is a complex and chronophage step when evaluating patients in critical care suspected of being neurologically deceased. Although the clinical neurological criteria that determine brain death are mostly accepted worldwide, the diagnosis remains imperfect and often the use of ancillary tests such as brain computed tomography perfusion (CTP) are required to confirm it. The main objective of this work was to explore the feasibility of classifying brain death from CTP scans using deep learning. CTP studies from a multicenter prospective diagnostic cohort study with the primary objective of evaluating the diagnostic accuracy of neurological death using CTP were pooled from 11 participating sites (INDex-CTP, ClinicalTrials.gov, NCT03098511). Spatial and temporal features were extracted using a combination of two convolution modules and used to predict neurological death. The performance of the model was also evaluated on subsets of cerebral injuries. 217 patients' studies were used to train the model. We report an AUC of 0.79 (IC95% 0.76-0.82), a F1 score of 0.82 (IC95% 0.80-0.83), a precision of 0.92 (IC95% 0.91-0.93), a recall of 0.76 (CI95% 0.72-0.79) as well as a negative predictive value of 0.49 (CI95% 0.45-0.53). Due to a lack of sample size, we did not perform statistical tests on subsets of cerebral injury, but report suspected higher model negative predictive value on anoxic cerebral injury with 0.82 (CI95% 0.77-0.87). This model shows preliminary evidence supporting the feasibility of developing a deep neural network to classify comatose patients as neurologically deceased or not. Additional research is needed to validate and refine the model by employing larger and more diverse datasets.