Aintree University Hospital is a National Health Service hospital in Fazakerley, Liverpool. It is managed by the Liverpool University Hospitals NHS Foundation Trust.
The prognostic role of high-risk plaque (HRP) features, including high coronary calcium scores detected by CT, beyond traditional cardiovascular risk factors and obstructive coronary artery disease (CAD), remains uncertain. This study evaluated the prognostic value of a combined HRP definition in stable chest pain patients with low-to-intermediate pretest probability of CAD. This prespecified analysis included participants randomized to the CT arm of the pragmatic, prospective 26-center European DISCHARGE trial (NCT02400229). The primary endpoint was major adverse cardiovascular events (MACE: cardiovascular death, nonfatal myocardial infarction, or stroke); the secondary endpoint was expanded MACE (transient ischemic attack and major procedure-related complications). Our combined HRP definition was any coronary plaque with positive remodeling, napkin-ring sign, low attenuation, or total calcium score ≥ 400 Agatston units. Among 1745 participants (age: 60 ± 10 years, 990 female), 35 MACE and 47 expanded MACE occurred at a median follow-up of 3.5 years (IQR: 2.9–4.2). After risk factor adjustment, the combined HRP definition was associated with a higher risk of MACE (HR: 3.81; 95
BACKGROUND:Interictal epileptiform discharges (IEDs) are transient spikes or waves that occur in electroencephalography (EEG) records and can help support the diagnosis and classification of epilepsy. High-throughput machine learning models aim to automate the detection of IEDs. Previous evaluations of machine learning models have reported non-inferiority compared to human experts, but these studies predominantly use small datasets of pre-selected, 'IED rich' records, which are not representative of clinical practice. Therefore, this study aims to analyse the accuracy of machine learning models in a large, routine, clinically representative cohort. METHODS:All routine EEGs performed in a large regional hospital in England were identified between June 2024 and February 2025. EEG records were run through the commercial machine learning model P15 and automated IED reports generated. The sensitivity, specificity, positive and negative predictive value of P15-detected IEDs were evaluated using the final clinical report as a reference standard. RESULTS:Of 484 EEG records, 53 were reported to contain at least one IED in the final clinical report. At P15's default sensitivity setting, sensitivity for IED detection was 81.1% (95% CI:77.6-84.6), specificity 59.9% (95% CI: 55.5-64.2), positive predictive value 19.9% (95% CI:16.3-23.5) and negative predictive value 96.3% (95% CI:94.6-98.0). DISCUSSION:This large-scale study of a machine learning model for identification of IEDs in a representative clinical population found a high negative predictive value suggesting that this may be a useful tool to rule out IEDs. However, the low positive predictive value demonstrates the potential for over-calling IEDs in routine EEGs. Future research should evaluate machine learning models alongside clinical feedback before this approach can have sufficient utility in direct clinical care.
Abstract Background and aims CT perfusion (CTP) may be negative or show a perfusion deficit in the wrong territory despite diffusion-weighted imaging (DWI)–positive MRI. We sought determinants of CTP negativity and territorial mismatch, and whether NIHSS subdomains add value beyond total NIHSS. Methods Retrospective cohort of consecutive clinical CTP (Oct 2024–Dec 2025) from two acute stroke centres. Patients were included if MRI was performed within 7 days of CTP (≤8 calendar days) and was DWI-positive. MRI was reported using a structured proforma and radiological variables were extracted. Multivariable logistic regression with 5-fold cross-validation reported AUC (mean, 95% CI). Outcomes were: (i) negative vs positive CTP (excluding wrong-area), and (ii) wrong-area vs correct-area CTP positivity. Results Of 722 CTP studies, 268 had MRI within 8 days; 159 were DWI-positive and included. For negative vs positive CTP, discrimination improved from clinical-only AUC 0.692 (0.518–0.865) to 0.777 (0.613–0.941) after adding MRI phenotype variables. For wrong-area vs correct-area CTP positivity, AUC improved from 0.624 (0.422–0.826) to 0.711 (0.477–0.946) with MRI variables. After adjustment for age, timing and MRI phenotype, adding NIHSS Visual/Gaze improved AUC from 0.772 (0.652–0.892) to 0.806 (0.712–0.900), exceeding the incremental value of total NIHSS. Conclusions Small, deep and/or single infarcts can yield misleading CTP - a negative or discordant CTP should not reassure when infarction is suspected. Conflict of interest Kausik Chatterjee: nothing to declare, Adam Seed: nothing to declare; Fathalla Elnagi: nothing to declare; NANA GYIMAH-KESSIE: nothing to declare Figure 1 - belongs to Methods Figure 2 - belongs to Results