Generalizability of electroencephalographic interpretation using artificial intelligence: An external validation study.

OBJECTIVE: The automated interpretation of clinical electroencephalograms (EEGs) using artificial intelligence (AI) holds the potential to bridge the treatment gap in resource-limited settings and reduce the workload at specialized centers. However, to facilitate broad clinical implementation, it is essential to establish generalizability across diverse patient populations and equipment. We assessed whether SCORE-AI demonstrates diagnostic accuracy comparable to that of experts when applied to a geographically different patient population, recorded with distinct EEG equipment and technical settings. METHODS: We assessed the diagnostic accuracy of a "fixed-and-frozen" AI model, using an independent dataset and external gold standard, and benchmarked it against three experts blinded to all other data. The dataset comprised 50% normal and 50% abnormal routine EEGs, equally distributed among the four major classes of EEG abnormalities (focal epileptiform, generalized epileptiform, focal nonepileptiform, and diffuse nonepileptiform). To assess diagnostic accuracy, we computed sensitivity, specificity, and accuracy of the AI model and the experts against the external gold standard. RESULTS: We analyzed EEGs from 104 patients (64 females, median age = 38.6 [range = 16-91] years). SCORE-AI performed equally well compared to the experts, with an overall accuracy of 92% (95% confidence interval [CI] = 90%-94%) versus 94% (95% CI = 92%-96%). There was no significant difference between SCORE-AI and the experts for any metric or category. SCORE-AI performed well independently of the vigilance state (false classification during awake: 5/41 [12.2%], false classification during sleep: 2/11 [18.2%]; p = .63) and normal variants (false classification in presence of normal variants: 4/14 [28.6%], false classification in absence of normal variants: 3/38 [7.9%]; p = .07). SIGNIFICANCE: SCORE-AI achieved diagnostic performance equal to human experts in an EEG dataset independent of the development dataset, in a geographically distinct patient population, recorded with different equipment and technical settings than the development dataset.

Identification of Genetic Loci Associated With Intracerebral Hemorrhage Using a Multitrait Analysis Approach.

BACKGROUND AND OBJECTIVES: Genome-wide association studies (GWASs) have only 2 loci associated with spontaneous intracerebral hemorrhage (ICH): APOE for lobar and 1q22 for nonlobar ICH. We aimed to discover new loci through an analysis that combines correlated traits (multi-trait analysis of GWAS [MTAG]) and explore a gene-based analysis, transcriptome-wide association study (TWAS), and proteome-wide association study (PWAS) to understand the biological mechanisms of spontaneous ICH providing potential therapeutic targets. METHODS: We use the published MTAG of ICH (patients with spontaneous intraparenchymal bleeding) and small-vessel ischemic stroke. For all ICH, lobar ICH, and nonlobar ICH, a pairwise MTAG combined ICH with traits related to cardiovascular risk factors, cerebrovascular diseases, or Alzheimer disease (AD). For the analysis, we assembled those traits with a genetic correlation ≥0.3. A new MTAG combining multiple traits was performed with those traits whose pairwise MTAG yielded new GWAS-significant single nucleotide polymorphisms (SNPs), with a posterior-probability of model 3 (GWAS-pairwise) ≥0.6. We perform TWAS and PWAS that correlate the genetic component of expression or protein levels with the genetic component of a trait. We use the ICH cohort from UK Biobank as replication. RESULTS: For all ICH (1,543 ICH, 1,711 controls), the mean age was 72 ± 2 in cases and 70 ± 2 in controls, and half of them were women. Replication cohort: 700 ICH and 399,717 controls. Novel loci were found only for all ICH (the trait containing lobar and nonlobar ICH), combining data of ICH and small vessel stroke, white matter hyperintensities volume, fractional anisotropy, mean diffusivity, and AD. We replicated 6 SNPs belonging to 2q33.2 (ICA1L, ß = 0.20, SE = 0.03, p value = 8.91 × 10-12), 10q24.33 (OBFC1, ß = -0.12, SE = 0.02, p value = 1.67 × 10-8), 13q34 (COL4A2, ß = 0.02, SE = 0.02, p value = 2.34 × 10-11), and 19q13.32 (APOC1, ß = -0.19, SE = 0.03, p value = 1.38 × 10-12; APOE, ß = 0.21, SE = 0.03, p value = 2.70 × 10-11; PVRL2:CTB-129P6.4, ß = 0.15, SE = 0.03, p value = 1.38 × 10-8); 2 genes (SH3PXD2A, Z-score = 4.83, p value = 6.67 × 10-7; and APOC1, Z-score: = 5.11, p value = 1.60 × 10-7); and ICA1L transcript (Z-score = 6.8, p value = 9.1 × 10-12) and protein levels (Z-score = -5.8, p value = 6.7 × 10-9). DISCUSSION: Our results reinforce the role of APOE in ICH risk, replicate previous ICH-associated loci (2q33 and 13q34), and point to new ICH associations with OBFC1, PVRL2:CTB-129P6.4, APOC1, and SH3PXD2A. Our study used data from European subjects, our main limitation. These molecules could be potential targets for future studies for modulating ICH risk.

COVID19-associated new-onset movement disorders: a follow-up study.

BACKGROUND: Neurological symptoms are common manifestation in acute COVID-19. This includes hyper- and hypokinetic movement disorders. Data on their outcome, however, is limited. METHODS: Cases with new-onset COVID-19-associated movement disorders were identified by searching the literature. Authors were contacted for outcome data which were reviewed and analyzed. RESULTS: Movement disorders began 12.6 days on average after the initial onset of COVID-19. 92% of patients required hospital admission (mean duration 23 days). In a fraction of patients (6 of 27; 22%; 4 males/2 females, mean age 66.8 years) the movement disorder (ataxia, myoclonus, tremor, parkinsonism) was still present after a follow-up period of 7.5 ± 3 weeks. Severe COVID-19 in general and development of encephalopathy were risk factors, albeit not strong predictors, for the persistence. CONCLUSIONS: The prognosis of new-onset COVID-19-associated movement disorder appears to be generally good. The majority recovered without residual symptoms within several weeks or months. Permanent cases may be due to unmasking of a previous subclinical movement disorder or due to vascular/demyelinating damage. Given the relatively low response rate of one third only and the heterogeneity of mechanisms firm conclusions on the (long-term) outome cannot, however, be drawn.

SURINPARK: Safinamide for Urinary Symptoms in Parkinson's Disease.

BACKGROUND: Urinary symptoms are common, disabling and generally unresponsive to treatment in Parkinson´s disease (PD). Safinamide is approved as an add-on therapy to levodopa to improve fluctuations. METHODS: Retrospective analysis of electronic records of nondemented PD patients seen consecutively in a Movement Disorders Unit (November 2018-February 2019). All were assessed with Scale for Outcomes in Parkinson's disease for Autonomic Symptoms-Urinary subscale (SCOPA-AUT-U) by the attending neurologist, and a month afterwards by an independent researcher blinded to treatment and clinical records in a routine clinical practice setting. Clinical variables were compared among patients who were prescribed safinamide (SA+) for the treatment of motor fluctuations and those with different treatment regimes (SA-). RESULTS: From 169 patients screened initially, 54 were excluded due to severe incontinence, absence of urinary symptoms or previous safinamide treatment. Thirty-five patients were included in SA+ and 79 in SA-. Both groups were comparable in terms of clinical variables, except in basal urinary symptoms, with more severity in the SA+ group. In the follow-up assessment, total SCOPA-AUT-U, as well as urgency, incontinence, frequency and nocturia subscales improved significantly in the SA+ group, while the SA- group remained unchanged. CONCLUSIONS: Safinamide could be helpful in the improvement of urinary symptoms in PD.

Use of the Cardiovascular Polypill in Secondary Prevention of Cerebrovascular Disease: A Real-Life Tertiary Hospital Cohort Study of 104 Patients.

BACKGROUND: The use of the cardiovascular polypill, a fixed-dose combination treatment, is conceived to improve adherence. However, randomized controlled trials (RCTs) may overestimate it. Studies focusing on cerebrovascular disease and real-life efficacy compared with conventional treatment are lacking. METHODS: This is a retrospective, hospital-based cohort study of acute ischaemic stroke patients who were prescribed a polypill (aspirin 100 mg, atorvastatin 20/40 mg, ramipril 2.5/5/10 mg) versus conventional treatment (aspirin 100 mg and other blood pressure/lipid-lowering agents) in secondary prevention (2017-2018). Clinical records were reviewed 90 days after discharge for stroke recurrence, vascular risk factor control, and safety. Adherence was assessed using the adapted Morisky-Green scale. RESULTS: A total of 104 patients were included (61% male; mean age 69.7 ± 13.9 years); 54 were treated with the polypill and 50 with conventional treatment. No baseline differences in clinical or demographic variables were detected. No recurrences were registered in the polypill group, compared to 1 recurrence in the conventional treatment group. A significant reduction of systolic blood pressure (SBP) was achieved in the polypill group (12.1 mm Hg) compared to the conventional treatment group (6.8 mm Hg) (p = 0.002). No significant differences were detected regarding the goal of LDL cholesterol ≤70 mg/dL (41 vs. 44%). The adverse events were mild and their frequency was similar in the two groups (9 vs. 2%, ns). Adherence was similarly good in the two groups (93 vs. 88%, ns). Polypill group adherence was similar to that reported in a previous meta-analysis of RCTs (93 vs. 84%, ns). CONCLUSION: In our experience, the cardiovascular polypill achieved a higher reduction in SBP levels and was well tolerated. Adherence was similar to that found in the previous literature, which is remarkable given the real-life setting of our study.