Segmental and regional bioimpedance measurements have become of great clinical interest in the monitoring of many pathologies. However, the use of commercial bioimpedance analyzers to carry out measurements on small segments of the body, with their associated low inter-electrode tissue impedances, leads to some unexpected problems. Low interelectrode tissue impedances are very prone to adverse effects resulting from the relatively large electrode contact impedances and, especially, from electrode contact impedance mismatches. The authors highlight these problems, first with measurements on electrical models of the tissue and electrode/skin contact impedances and then with measurements carried out on human calves. It is concluded that commercial equipment must be used carefully, especially when carrying out novel, localised bioimpedance measurements for which the devices were not originally designed.

The characterization of linear electrical components for frequencies up to a few MHz can be performed through impedance spectroscopy techniques that include the determination of the impedance frequency response and the algorithms that identify the equivalent circuit components. However, at higher frequencies some extra challenges are present since it is not practical to measure the impedance frequency response and s-parameters must be used. Although the conversion between s-parameters and equivalent impedance is well known, the application of the circuit identification algorithms is not straightforward. This paper presents a new procedure to perform circuit identification, through impedance spectroscopy, for components that operate at UHF. The s-parameters are measured and the equivalent circuit is obtained by using evolutionary algorithms, namely genetic algorithms and gene expression programming. The procedure is validated through experimental characterization of an RFID device.

The estimation of gait temporal parameters with wearable inertial measurement units (IMU) has been a research topic of interest in clinical gait analysis. Several methods, based on the use of a single IMU, have been proposed for their estimate, showing in most cases a sufficiently low number of missed and extra events and a satisfactory accuracy when applied to the gait of healthy subjects. In this study, we hypothesized that the same does not hold for abnormal gait. Some of the methods were applied to the gait of a group of hemiparetic subjects and to a group of healthy elderly. The results confirmed the hypothesis since, when applied to the hemiparetic subjects gait, all methods showed a) a higher number of missed and extra gait events detected by all tested methods, b) a significantly lower accuracy in estimating both gait events and temporal parameters. Errors in estimating gait temporal parameters were not significantly different among methods.

This paper presents a novel wearable sensor system based on the integration of miniaturised IMUs for fine hand movement analysis. The system, named SensHand V1, is composed of full 9-axis inertial sensors, placed on the fingers and wrist, which are managed by a cortex-M3 microcontroller. The acquired data are sent to a data logger through the use of Bluetooth communication. In this paper, the system is used for the objective diagnosis of Parkinson's disease, which is commonly assessed by neurologists through visual examination of motor tasks and semi-quantitative rating scales. Here, these motor tasks are also assessed using the SensHand V1, and then compared with the subjective metrics. Results demonstrate that the system is adequate to support neurologists in diagnostic procedures and allows for an objective evaluation of the disease.

Noise cancellation and detail preservation are two conflicting goals in the design of any denoising filter. In order to specifically measure these features, vector metrics have been recently introduced for grayscale and color images. This paper provides, for the first time, a study of the accuracy of the main algorithms in the literature focusing on the case of Gaussian noise. A novel vector method is also proposed to overcome the inaccuracies of current techniques.