Second-Order Training of Adaptive Critics for Online Process Control

A new algorithm for training of nonlinear optimal neuro-controllers (in the form of the model-free, action-dependent, adaptive critic paradigm). Overcomes problems with existing stochastic backpropagation training: need for data storage, parameter shadowing and poor convergence, offering significant benefits for online applications.

The effect of type of hip protector and resident characteristics on adherence to use of hip protectors in nursing and residential homes - an exploratory study

Objectives: To investigate the factors influencing the acceptability of hip protectors to residents of nursing and residential homes, especially the effect of hip protector type, and resident characteristics. Design: A randomised controlled trial with 12 weeks follow-up. Participants were randomised to receive either Safehips or HipSaverTM hip protectors. Setting/Participants: 109 residents aged 61 to 98 years from seven residential homes and two nursing homes in Northern Ireland. Main outcome measures: Percentage day-time use of the hip protectors over 12 weeks and ongoing use at 12 weeks. Results: 42% (119/285) of residents invited to enter the studyagreed to take part, and 109 started to wear the hip protectors. 43.1% (47/109) were still using them at 12 weeks. Mean percentage day-time use for all residents during 12 weeks was 48.6%. There was no significant difference in percentage day-time use (p=0.40), or use at 12 weeks (p=0.56) between the residents wearing Safehips and HipSaverTM protectors. Greater percentage daytime use of hip protectors was associated with being resident in a home for the elderlymentallyinfirm (75.1%, pp0.0005), having a low (12 or less) Barthel score (61.1%, pp0.0005), and having been injured in a fall in the last 12 months (57.3%, p=0.012). Conclusions: The type of hip protector appeared to make no difference to their continued use by residents. Residents with a historyof a fall and those who are physicallyand mentallyincapacitated appear to be more likelyto wear hip protectors. These residents, who are at high risk of falling, are also highlydependent on nursing staff. Efforts to increase hip protector use in residential and nursing home should focus on staff, who are in the best position to advise and influence residents and their relatives.