Motor signs in Alzheimer's disease and vascular dementia: Detection through natural language processing, co-morbid features and relationship to adverse outcomes

Predictors for survival in patients with Alzheimer's disease: a large comprehensive meta-analysis

The prevalence of Alzheimer's disease (AD) is increasing as the population ages, and patients with AD have a poor prognosis. However, knowledge on factors for predicting the survival of AD remains sparse. Here, we aimed to systematically explore predictors of AD survival. We searched the PubMed, Embase and Cochrane databases for relevant literature from inception to December 2022. Cohort and case-control studies were selected, and multivariable adjusted relative risks (RRs) were pooled by random-effects models. A total of 40,784 reports were identified, among which 64 studies involving 297,279 AD patients were included in the meta-analysis after filtering based on predetermined criteria. Four aspects, including demographic features (n = 7), clinical features or comorbidities (n = 13), rating scales (n = 3) and biomarkers (n = 3), were explored and 26 probable prognostic factors were finally investigated for AD survival. We observed that AD patients who had hyperlipidaemia (RR: 0.69) were at a lower risk of death. In contrast, male sex (RR: 1.53), movement disorders (including extrapyramidal signs) (RR: 1.60) and cancer (RR: 2.07) were detrimental to AD patient survival. However, our results did not support the involvement of education, hypertension, APOE genotype, Aβ42 and t-tau in AD survival. Our study comprehensively summarized risk factors affecting survival in patients with AD, provided a better understanding on the role of different factors in the survival of AD from four dimensions, and paved the way for further research.

Identifying Military Service Status in Electronic Healthcare Records from Psychiatric Secondary Healthcare Services: A Validation Exercise Using the Military Service Identification Tool

Electronic healthcare records (EHRs) are a rich source of information with a range of uses in secondary research. In the United Kingdom, there is no pan-national or nationally accepted marker indicating veteran status across all healthcare services. This presents significant obstacles to determining the healthcare needs of veterans using EHRs. To address this issue, we developed the Military Service Identification Tool (MSIT), using an iterative two-staged approach. In the first stage, a Structured Query Language approach was developed to identify veterans using a keyword rule-based approach. This informed the second stage, which was the development of the MSIT using machine learning, which, when tested, obtained an accuracy of 0.97, a positive predictive value of 0.90, a sensitivity of 0.91, and a negative predictive value of 0.98. To further validate the performance of the MSIT, the present study sought to verify the accuracy of the EHRs that trained the MSIT models. To achieve this, we surveyed 902 patients of a local specialist mental healthcare service, with 146 (16.2%) being asked if they had or had not served in the Armed Forces. In total 112 (76.7%) reported that they had not served, and 34 (23.3%) reported that they had served in the Armed Forces (accuracy: 0.84, sensitivity: 0.82, specificity: 0.91). The MSIT has the potential to be used for identifying veterans in the UK from free-text clinical documents and future use should be explored.

Association between Motor Signs and Cognitive Performance in Cognitively Unimpaired Older Adults: A Cross-Sectional Study Using the NACC Database

Aiming to examine whether specific motor signs are associated with worse performance in specific cognitive domains among cognitively unimpaired (CU) individuals, we performed a cross-sectional analysis of data from the baseline evaluations of older, CU participants from the National Alzheimer's Coordinating Center (NACC) Uniform Data Set. In total, 8149 CU (≥60 years) participants were included. Of these, 905 individuals scored ≥ 2 on at least one of the motor domains of the Unified Parkinson's Disease Rating Scale part III (UPDRSIII). Cognitively impaired individuals, participants with psychiatric disorders and/or under treatment with antipsychotic, anxiolytic, sedative or hypnotic agents were excluded. Nine motor signs were examined: hypophonia, masked facies, resting tremor, action/postural tremor, rigidity, bradykinesia, impaired chair rise, impaired posture/gait and postural instability. Their association with performance on episodic memory, semantic memory, language, attention, processing speed or executive function was assessed using crude and adjusted linear regression models. Individuals with impaired chair rise had worse episodic memory, semantic memory, processing speed and executive function, while those with bradykinesia had worse language, processing speed and executive function. Sensitivity analyses, by excluding participants with cerebrovascular disease or PD, or other Parkinsonism, produced similar results with the exception of the relationship between bradykinesia and language performance.

English Text Readability Measurement Based on Convolutional Neural Network: A Hybrid Network Model

Text readability is very important in meeting people's information needs. With the explosive growth of modern information, the measurement demand of text readability is increasing. In view of the text structure of words, sentences, and texts, a hybrid network model based on convolutional neural network is proposed to measure the readability of English texts. The traditional method of English text readability measurement relies too much on the experience of artificial experts to extract features, which limits its practicability. With the increasing variety and quantity of text readability measurement features to be extracted, it is more and more difficult to extract deep features manually, and it is easy to introduce irrelevant features or redundant features, resulting in the decline of model performance. This paper introduces the concept of hybrid network model in deep learning; constructs a hybrid network model suitable for English text readability measurement by combining convolutional neural network, bidirectional long short-term memory network, and attention mechanism network; and replaces manual automatic feature extraction by machine learning, which greatly improves the measurement efficiency and performance of text readability.