The association of cognitive function and its changes with all-cause mortality among community-dwelling older adults.

Background: The association of cognitive function, its changes, and all-cause mortality has not reached a consensus, and the independence of the association between changes in cognitive function and mortality remains unclear. The purpose of this study was to evaluate the longitudinal association between baseline cognitive function and cognitive changes over 1 year with subsequent all-cause mortality among the older adults aged 60 and above. Methods: A prospective cohort study utilizing the Community Older Adults Health Survey data. Initiated in 2018, the study annually assessed all individuals aged 60+ in Dalang Town, Dongguan City. Cognitive function was assessed using the Chinese version of the Mini-Mental State Examination (MMSE). A total of 6,042 older adults individuals were included, and multivariate Cox proportional hazard models were used to examine cognitive function's impact on mortality. Results: Participants' median age was 70 years, with 39% men. Over a median 3.08-year follow-up, 525 died. Mortality risk increased by 6% per MMSE score decrease (adjusted HR = 1.06, 95%CI: 1.05-1.08). Compared to those with normal cognitive function at baseline, participants with mild cognitive impairment and moderate to severe cognitive impairment had significantly higher mortality risks (adjusted HR = 1.40, 95%CI: 1.07-1.82; HR = 2.49, 95%CI: 1.91-3.24, respectively). The risk of death was 5% higher for each one-point per year decrease in cognitive function change rate (HR = 1.05, 95%CI: 1.02-1.08). Compared with participants with stable cognitive function, those with rapid cognitive decline had a 79% increased risk of death (adjusted HR = 1.79, 95% CI: 1.11-2.87), with baseline cognitive function influencing this relationship significantly (P for interaction = 0.002). Conclusion: Baseline cognitive impairment and rapid cognitive decline are associated with higher all-cause mortality risks in Chinese older adults. Baseline function influences the mortality impact of cognitive changes.

A new domain adaption residual separable convolutional neural network model for cross-domain remaining useful life prediction.

In order to realize the remaining useful life (RUL) prediction of mechanical equipment under different operating conditions, a domain adaption residual separable convolutional neural network (DRSCN) model is proposed in this paper. In the DRSCN model, instead of the traditional convolutional layer, a residual separable convolutional module is developed to improve the feature extraction ability of the model. Moreover, a multi-kernel maximum mean discrepancy metric function and an adversarial learning mechanism are embedded in the DRSCN model to enhance its ability to resist domain shifts, thus improving the cross-domain RUL prediction accuracy of the model. The effectiveness of the DRSCN model is verified on an aircraft engine dataset. The experimental results show that the proposed model can realize high-accuracy RUL prediction.

Dual Neural Network Method for Solving Multiple Definite Integrals.

This study, which examines a calculation method on the basis of a dual neural network for solving multiple definite integrals, addresses the problems of inefficiency, inaccuracy, and difficulty in finding solutions. First, the method offers a dual neural network method to construct a primitive function of the integral problem; it can approximate the primitive function of any given integrand with any precision. On this basis, a neural network calculation method that can solve multiple definite integrals whose upper and lower bounds are arbitrarily given is obtained with repeated applications of the dual neural network to construction of the primitive function. Example simulations indicate that compared with traditional methods, the proposed algorithm is more efficient and precise in obtaining solutions for multiple integrals with unknown integrand, except for the finite input-output data points. The advantages of the proposed method include the following: (1) integral multiplicity shows no influence and restriction on the employment of the method; (2) only a finite set of known sample points is required without the need to know the exact analytical expression of the integrand; and (3) high calculation accuracy is obtained for multiple definite integrals whose integrand is expressed by sample data points.