According to the World Health Organization (WHO), Diabetes Mellitus (DM) is one of the most prevalent diseases in the world. It is also associated with a high mortality index. Diabetic foot is one of its main complications, and it comprises the development of plantar ulcers that could result in an amputation. Several works report that thermography is useful to detect changes in the plantar temperature, which could give rise to a higher risk of ulceration. However, the plantar temperature distribution does not follow a particular pattern in diabetic patients, thereby making it difficult to measure the changes. Thus, there is an interest in improving the success of the analysis and classification methods that help to detect abnormal changes in the plantar temperature. All this leads to the use of computer-aided systems, such as those involved in artificial intelligence (AI), which operate with highly complex data structures. This paper compares machine learning-based techniques with Deep Learning (DL) structures. We tested common structures in the mode of transfer learning, including AlexNet and GoogleNet. Moreover, we designed a new DL-structure, which is trained from scratch and is able to reach higher values in terms of accuracy and other quality measures. The main goal of this work is to analyze the use of AI and DL for the classification of diabetic foot thermograms, highlighting their advantages and limitations. To the best of our knowledge, this is the first proposal of DL networks applied to the classification of diabetic foot thermograms. The experiments are conducted over thermograms of DM and control groups. After that, a multi-level classification is performed based on a previously reported thermal change index. The high accuracy obtained shows the usefulness of AI and DL as auxiliary tools to aid during the medical diagnosis.

Induction motor is a ubiquitous machine. In industrial settings, online monitoring of motors' health status in order to schedule maintenance operations with the goal of damage prevention has become an essential necessity. Broken rotor bar is one of the most common failures in the rotor of a squirrel cage motor. Motor current signature analysis (MCSA) has become a popular method for the detection of this failure because of its high reliability. Recent works have performed the MCSA with a combination of different signal processing techniques to identify the presence of broken bars. In this paper, the MCSA is done with empirical mode decomposition from which a set of intrinsic mode functions (IMFs) is obtained. The extracted features from two of the obtained IMFs form the basis of the proposed classification criterion; these are the samples between zero crossings (SBZCs) and the time between successive zero crossings (TSZCs). The standard deviation from the SBZCs and the TSZCs is used as a classification feature. Experimental results using our method show high accuracy in the detection of a broken and a half-broken rotor bar.

This paper introduces a particle swarm optimization (PSO) algorithm that uses experimental measurements for the identification of the direct ($d$ ) and quadrature ($q$ ) stator inductances and the stator resistance of permanent-magnet synchronous machines (PMSMs). Strategies for the identification of stator resistance and flux linkage of permanent magnets are also presented. The developed PSO algorithm has the advantages of fast and stable convergence characteristics, and it is relatively easy to implement; moreover, neither voltage source inverters nor closed-loop control strategies are needed. In this paper, the experimental results obtained by PSO are compared with other ones obtained through a current decay test to demonstrate the effectiveness of the presented method. A 32-bit C2000 real-time microcontroller and a commercial surface-mounted PMSM have been used to carry out the experimental measurements.