Improved Nonlinear PCA for Process Monitoring Using Support Vector Data Description

Nonlinear principal component analysis (PCA) based on neural networks has drawn significant attention as a monitoring tool for complex nonlinear processes, but there remains a difficulty with determining the optimal network topology. This paper exploits the advantages of the Fast Recursive Algorithm, where the number of nodes, the location of centres, and the weights between the hidden layer and the output layer can be identified simultaneously for the radial basis function (RBF) networks. The topology problem for the nonlinear PCA based on neural networks can thus be solved. Another problem with nonlinear PCA is that the derived nonlinear scores may not be statistically independent or follow a simple parametric distribution. This hinders its applications in process monitoring since the simplicity of applying predetermined probability distribution functions is lost. This paper proposes the use of a support vector data description and shows that transforming the nonlinear principal components into a feature space allows a simple statistical inference. Results from both simulated and industrial data confirm the efficacy of the proposed method for solving nonlinear principal component problems, compared with linear PCA and kernel PCA.

Do psychosocial factors moderate the association of fixed-term employment with work interference with family and sleeping problems in registered nurses: A cross-sectional questionnaire survey

Background: The use affixed-term employment has increased lately, particularly in Europe and in the health care sector. Previous studies have shown that especially among the health care sector employee's organizational justice perceptions and job control are important factors that are directly related to the welfare and attitudes of employees and may also help to buffer the negative impacts of many detrimental factors.

The Association of Distress and Sleeping Problems With Physicians' Intentions To Change Profession: The Moderating Effect of Job Control

The present study examined whether job control moderated the association between stress indicators (distress and sleeping problems) and intentions to change profession among 2,650 Finnish physicians. Ordinal logistic regression analysis was applied. The authors found that high levels of distress and sleeping problems were associated with higher levels of intentions to change profession, whereas high job control was associated with lower levels of intentions to change profession even after adjusting for the effects of gender, age, and employment sector. In addition, high job control was able to mitigate the positive association that distress and sleeping problems had with intentions to change profession. Our findings highlight the importance of offering more job control to physicians to prevent unnecessary physician turnover.

Paralytic shellfish poisoning (PSP) toxin binders for optical biosensor technology: problems and possibilities for the future: a review

This review examines the developments in optical biosensor technology, which uses the phenomenon of surface plasmon resonance, for the detection of paralytic shellfish poisoning (PSP) toxins. Optical biosensor technology measures the competitive biomolecular interaction of a specific biological recognition element or binder with a target toxin immobilised onto a sensor chip surface against toxin in a sample. Different binders such as receptors and antibodies previously employed in functional and immunological assays have been assessed. Highlighted are the difficulties in detecting this range of low molecular weight toxins, with analogues differing at four chemical substitution sites, using a single binder. The complications that arise with the toxicity factors of each toxin relative to the parent compound, saxitoxin, for the measurement of total toxicity relative to the mouse bioassay are also considered. For antibodies, the cross-reactivity profile does not always correlate to toxic potency, but rather to the toxin structure to which it was produced. Restrictions and availability of the toxins makes alternative chemical strategies for the synthesis of protein conjugate derivatives for antibody production a difficult task. However, when two antibodies with different cross-reactivity profiles are employed, with a toxin chip surface generic to both antibodies, it was demonstrated that the cross-reactivity profile of each could be combined into a single-assay format. Difficulties with receptors for optical biosensor analysis of low molecular weight compounds are discussed, as are the potential of alternative non-antibody-based binders for future assay development in this area.