Search

Objectives To investigate the clinical meaning of brain parenchymal computed-tomography hyperdensities (CTHD) in patients treated of anterior circulation acute stroke with reperfusion therapy.Methods Patients were retrospectively enrolled from three different hospitals. Brain CT scans were assessed at four time points: We recorded ASPECT scores of pre-treatment CTs, assessed ASPECT scores and the presence of CTHD on posttreatment CTs acquired within 24-30 h and 24-72 h, and examined a one-month CTs follow-up to determine the ischemic evolution of CTHD. We correlated the presence of CTHD with clinical and radiological data to define its predictive and prognostic factors.Results In total, 165 patients were evaluated. At post-treatment CTs acquired within 24-30 h, 68 (41%) patients showed the presence of CTHD. On post-treatment CTs acquired within 24-72 h, 43 (63%) of the CTHD showed hemorrhagic transformation. Sixty-five (95%) out of the 68 CTHD evolved in a final ischemic brain area. Multivariate statistical analysis identified puncture to recanalization time to be the only independent factors predicting the presence of CTHD (p = 0.045). The presence of CTHD at the first post-treatment CTs was an independent factor for clinical outcome determined with mRS scores at 3-month follow-up (p = 0.05). Outcomes were worse for hemorrhagic transformation at follow-up CTs compared to the ischemic evolution of the CTHD (p = 0.01).Conclusions The presence of CTHD at CTs imaging acquired within 24-30 h after reperfusion therapy is an independent prognostic factor of a worse clinical outcome, regardless of its ASPECT score at baseline CTs and of its hemorrhagic evolution.

Purpose To evaluate safety and diagnostic accuracy of gadoteridol vs. other macrocyclic gadolinium-based contrast agents (GBCAs) in a large cohort of consecutive and non-selected patients referred for CE-MRI of the CNS. Material and methods Between November 2017 and March 2018, we prospectively enrolled a consecutive cohort of patients referred for neuroradiological CE-MRI (1.5 T MRI). Image quality and adverse events were assessed. Diagnostic performance was determined for a subgroup of patients with truth standard findings available. Comparison was made between patients receiving gadoteridol and patients receiving other macrocyclic GBCAs. Inter-reader agreement (kappa) between two expert neuroradiologists was calculated for the diagnosis of malignancy. Results Overall, 460 patients (220 M/240F; mean age 54 ± 16 years) were enrolled of which 230 received gadoteridol (Group 1) and 230 either gadoteric acid or gadobutrol [n = 83 (36.1%) and n = 147 (63.9%), respectively; Group 2]. Image quality was rated as good or excellent in both groups. The sensitivity, specificity and diagnostic accuracy for determination of malignancy was 88.2%, 96.5% and 95.4%, respectively, for Group 1 and 93.7%, 97.4% and 96.9%, respectively, for Group 2, with no significant differences between groups (P > 0.75) for any determination. Inter-reader agreement for the identification of malignancy was excellent [K = 0.877 (95%CI: 0.758–0.995) and K = 0.818 (95%CI: 0.663–0.972) for groups 1 and 2, respectively; P = 0.0913]. Adverse events occurred in 5 of 460 (1.09%) patients overall, with no significant difference (P = 0.972) between groups. Conclusion Gadoteridol was safe and guaranteed good image quality without significant differences when compared to gadobutrol and gadoteric acid in a wide range of CNS pathologies.

OBJECTIVES The aims of this study were to present a deep learning approach for the automated classification of multiple sclerosis and its mimics and compare model performance with that of 2 expert neuroradiologists. MATERIALS AND METHODS A total of 268 T2-weighted and T1-weighted brain magnetic resonance imagin scans were retrospectively collected from patients with migraine (n = 56), multiple sclerosis (n = 70), neuromyelitis optica spectrum disorders (n = 91), and central nervous system vasculitis (n = 51). The neural network architecture, trained on 178 scans, was based on a cascade of 4 three-dimensional convolutional layers, followed by a fully dense layer after feature extraction. The ability of the final algorithm to correctly classify the diseases in an independent test set of 90 scans was compared with that of the neuroradiologists. RESULTS The interrater agreement was 84.9% (Cohen κ = 0.78, P < 0.001). In the test set, deep learning and expert raters reached the highest diagnostic accuracy in multiple sclerosis (98.8% vs 72.8%, P < 0.001, for rater 1; and 81.8%, P < 0.001, for rater 2) and the lowest in neuromyelitis optica spectrum disorders (88.6% vs 4.4%, P < 0.001, for both raters), whereas they achieved intermediate values for migraine (92.2% vs 53%, P = 0.03, for rater 1; and 64.8%, P = 0.01, for rater 2) and vasculitis (92.1% vs 54.6%, P = 0.3, for rater 1; and 45.5%, P = 0.2, for rater 2). The overall performance of the automated method exceeded that of expert raters, with the worst misdiagnosis when discriminating between neuromyelitis optica spectrum disorders and vasculitis or migraine. CONCLUSIONS A neural network performed better than expert raters in terms of accuracy in classifying white matter disorders from magnetic resonance imaging and may help in their diagnostic work-up.

Parkinson’s disease (PD) is a progressive neurodegenerative disorder, characterized by motor and non-motor symptoms due to the degeneration of the pars compacta of the substantia nigra (SNc) with dopaminergic denervation of the striatum. Although the diagnosis of PD is principally based on a clinical assessment, great efforts have been expended over the past two decades to evaluate reliable biomarkers for PD. Among these biomarkers, magnetic resonance imaging (MRI)-based biomarkers may play a key role. Conventional MRI sequences are considered by many in the field to have low sensitivity, while advanced pulse sequences and ultra-high-field MRI techniques have brought many advantages, particularly regarding the study of brainstem and subcortical structures. Nowadays, nigrosome imaging, neuromelanine-sensitive sequences, iron-sensitive sequences, and advanced diffusion weighted imaging techniques afford new insights to the non-invasive study of the SNc. The use of these imaging methods, alone or in combination, may also help to discriminate PD patients from control patients, in addition to discriminating atypical parkinsonian syndromes (PS). A total of 92 articles were identified from an extensive review of the literature on PubMed in order to ascertain the-state-of-the-art of MRI techniques, as applied to the study of SNc in PD patients, as well as their potential future applications as imaging biomarkers of disease. Whilst none of these MRI-imaging biomarkers could be successfully validated for routine clinical practice, in achieving high levels of accuracy and reproducibility in the diagnosis of PD, a multimodal MRI-PD protocol may assist neuroradiologists and clinicians in the early and differential diagnosis of a wide spectrum of neurodegenerative disorders.