Search

While the importance of automatic image analysis is increasing at an enormous pace, recent meta-research revealed major flaws with respect to algorithm validation. Specifically, performance metrics are key for objective, transparent and comparative performance assessment, but relatively little attention has been given to the practical pitfalls when using specific metrics for a given image analysis task. A common mission of several international initiatives is therefore to provide researchers with guidelines and tools to choose the performance metrics in a problem-aware manner. This dynamically updated document has the purpose to illustrate important limitations of performance metrics commonly applied in the field of image analysis. The current version is based on a Delphi process on metrics conducted by an international consortium of image analysis experts.

See Herms and Schön (doi:10.1093/brain/awx100) for a scientific commentary on this article. Glaucoma is often diagnosed late when vision loss has already occurred. Cordeiro et al. report a new fluorescent marker for retinal imaging that can safely visualise real-time in vivo neuronal apoptosis in patients. Increased labelling is observed in patients with progressive neurodegenerative disease compared to healthy controls., See Herms and Schön (doi:10.1093/brain/awx100) for a scientific commentary on this article. Retinal cell apoptosis occurs in many ocular neurodegenerative conditions including glaucoma—the major cause of irreversible blindness worldwide. Using a new imaging technique that we have called DARC (detection of apoptosing retinal cells), which until now has only been demonstrated in animal models, we assessed if annexin 5 labelled with fluorescent dye DY-776 (ANX776) could be used safely in humans to identify retinal cell apoptosis. Eight patients with glaucomatous neurodegeneration and evidence of progressive disease, and eight healthy subjects were randomly assigned to intravenous ANX776 doses of 0.1, 0.2, 0.4 and 0.5 mg in an open-label, phase 1 clinical trial. In addition to assessing the safety, tolerability and pharmacokinetics of ANX776, the study aimed to explore whether DARC could successfully visualize individual retinal cell apoptosis in vivo in humans, with the DARC count defined as the total number of unique ANX776-labelled spots. DARC enabled retinal cell apoptosis to be identified in the human retina using ANX776. Single ANX776-labelled cells were visualized in a dose-dependent pattern (P < 0.001) up to 6 h after injection. The DARC count was significantly higher (2.37-fold, 95% confidence interval: 1.4–4.03, P = 0.003) in glaucoma patients compared to healthy controls, and was significantly (P = 0.045) greater in patients who later showed increasing rates of disease progression, based on either optic disc, retinal nerve fibre layer or visual field parameters. Additionally, the DARC count significantly correlated with decreased central corneal thickness (Spearman’s R = −0.68, P = 0.006) and increased cup-disc ratios (Spearman’s R = 0.47, P = 0.038) in glaucoma patients and with increased age (Spearman’s R = 0.77, P = 0.001) in healthy controls. Finally, ANX776 was found to be safe and well-tolerated with no serious adverse events, and a short half-life (10–36 min). This proof-of-concept study demonstrates that retinal cell apoptosis can be identified in the human retina with increased levels of activity in glaucomatous neurodegenerative disease. To our knowledge, this is the first time individual neuronal apoptosis has been visualized in vivo in humans and is the first demonstration of detection of individual apoptotic cells in a neurodegenerative disease. Furthermore, our results suggest the level of apoptosis (‘DARC count’) is predictive of disease activity, indicating the potential of DARC as a surrogate marker. Although further trials are clearly needed, this study validates experimental findings supporting the use of DARC as a method of detection and monitoring of patients with glaucomatous neurodegeneration, where retinal ganglion cell apoptosis is an established process and where there is a real need for tools to non-invasively assess treatment efficacy.

International audience

Genetic frontotemporal dementia is most commonly caused by mutations in the progranulin (GRN), microtubule-associated protein tau (MAPT) and chromosome 9 open reading frame 72 (C9orf72) genes. Previous small studies have reported the presence of cerebral white matter hyperintensities (WMH) in genetic FTD but this has not been systematically studied across the different mutations. In this study WMH were assessed in 180 participants from the Genetic FTD Initiative (GENFI) with 3D T1- and T2-weighed magnetic resonance images: 43 symptomatic (7 GRN, 13 MAPT and 23 C9orf72), 61 presymptomatic mutation carriers (25 GRN, 8 MAPT and 28 C9orf72) and 76 mutation negative non-carrier family members. An automatic detection and quantification algorithm was developed for determining load, location and appearance of WMH. Significant differences were seen only in the symptomatic GRN group compared with the other groups with no differences in the MAPT or C9orf72 groups: increased global load of WMH was seen, with WMH located in the frontal and occipital lobes more so than the parietal lobes, and nearer to the ventricles rather than juxtacortical. Although no differences were seen in the presymptomatic group as a whole, in the GRN cohort only there was an association of increased WMH volume with expected years from symptom onset. The appearance of the WMH was also different in the GRN group compared with the other groups, with the lesions in the GRN group being more similar to each other. The presence of WMH in those with progranulin deficiency may be related to the known role of progranulin in neuroinflammation, although other roles are also proposed including an effect on blood-brain barrier permeability and the cerebral vasculature. Future studies will be useful to investigate the longitudinal evolution of WMH and their potential use as a biomarker as well as post-mortem studies investigating the histopathological nature of the lesions., Highlights • In genetic FTD white matter hyperintensities (WMH) are found most prominently in symptomatic patients with GRN mutations. • Frontal and occipital lobes are the most affected regions. • WMH are more likely to occur close to the ventricles. • WMH have a more homogenous appearance possibly suggestive of inflammatory rather than vascular lesions.

$\textbf{Background:}$ In the 24-month MS-STAT phase 2 trial, we showed that high-dose simvastatin significantly reduced the annualised rate of whole brain atrophy in patients with secondary progressive multiple sclerosis (SPMS). We now describe the results of the MS-STAT cognitive substudy, in which we investigated the treatment effect on cognitive, neuropsychiatric, and health-related quality-of-life (HRQoL) outcome measures. $\textbf{Methods:}$ We did a secondary analysis of MS-STAT, a 24-month, double-blind, controlled trial of patients with SPMS done at three neuroscience centres in the UK between Jan 28, 2008, and Nov 4, 2011. Patients were randomly assigned (1:1) to either 80 mg simvastatin (n=70) or placebo (n=70). The cognitive assessments done were the National Adult Reading Test, Wechsler Abbreviated Scale of Intelligence, Graded Naming Test, Birt Memory and Information Processing Battery (BMIPB), Visual Object and Space Perception battery (cube analysis), Frontal Assessment Battery (FAB), and Paced Auditory Serial Addition Test. Neuropsychiatric status was assessed using the Hamilton Depression Rating Scale and the Neuropsychiatric Inventory Questionnaire. HRQoL was assessed using the self-reported 36-Item Short Form Survey (SF-36) version 2. Assessments were done at study entry, 12 months, and 24 months. Patients, treating physicians, and outcome assessors were masked to treatment allocation. Analyses were by intention to treat. MS-STAT is registered with ClinicalTrials.gov, number NCT00647348. $\textbf{Findings:}$ Baseline assessment revealed impairments in 60 (45%) of 133 patients on the test of frontal lobe function (FAB), and in between 13 (10%) and 43 (33%) of 130 patients in tests of non-verbal and verbal memory (BMIPB). Over the entire trial, we noted significant worsening on tests of verbal memory (T score decline of 5·7 points, 95% CI 3·6-7·8; p