The rearing of poultry for meat production (broilers) is an agricultural food industry with high relevance to the economy and development of some countries. Periodic episodes of extreme climatic conditions during the summer season can cause high mortality among birds, resulting in economic losses. In this context, ventilation systems within poultry houses play a critical role to ensure appropriate indoor climatic conditions. The objective of this study was to develop a multisensor system to evaluate the design of the ventilation system in broiler houses. A measurement system equipped with three types of sensors: air velocity, temperature and differential pressure was designed and built. The system consisted in a laptop, a data acquisition card, a multiplexor module and a set of 24 air temperature, 24 air velocity and two differential pressure sensors. The system was able to acquire up to a maximum of 128 signals simultaneously at 5 second intervals. The multisensor system was calibrated under laboratory conditions and it was then tested in field tests. Field tests were conducted in a commercial broiler farm under four different pressure and ventilation scenarios in two sections within the building. The calibration curves obtained under laboratory conditions showed similar regression coefficients among temperature, air velocity and pressure sensors and a high goodness fit (R² = 0.99) with the reference. Under field test conditions, the multisensor system showed a high number of input signals from different locations with minimum internal delay in acquiring signals. The variation among air velocity sensors was not significant. The developed multisensor system was able to integrate calibrated sensors of temperature, air velocity and differential pressure and operated succesfully under different conditions in a mechanically-ventilated broiler farm. This system can be used to obtain quasi-instantaneous fields of the air velocity and temperature, as well as differential pressure maps to assess the design and functioning of ventilation system and as a verification and validation (V&V) system of Computational Fluid Dynamics (CFD) simulations in poultry farms.

Hidden Markov Models have shown promising results for identification of spike sources in Parkinson's disease treatment, e.g., for deep brain stimulation. Usual classification criteria consist in maximum likelihood rule for the recognition of the correct class. In this paper, we present a different classification scheme based in proximity analysis. For this approach matrices of Markov process are transformed to another space where similarities and differences to other Markov processes are better revealed. The authors present the proximity analysis approach using hidden Markov models for the identification of spike sources (Thalamo and Subthalamo sources, Gpi and GPe sources). Results show that proximity analysis improves recognition performance for about 5% over traditional approach

The well-known inherent artifact on the rheoencephalogram (REG) caused by the pulsatility of the scalp blood flow left the REG out of the clinical practice. In fact, depending on the selected electrode arrangement, the measurement of the brain impedance changes time-locked with the heartbeat can be completely buried on that of the scalp. In this work, a novel mathematical method based on the physiological differences between the brain and scalp perfusions is proposed to extract the intracranial information from REG. This method is experimentally applied to REG signals recorded at five electrode positions and results are compared with those derived from our previous theoretical works. Intracranial components extracted from the REG signals are consistent with the stated hypothesis and reproduce the unexpected results obtained with our theoretical models. Although further studies would be needed, the evidences found in this work suggest that the method proposed in this work extracts the intracranial information from the REG signal