Household appliances, climate control machines, vehicles, elevators, cash counting machines, etc., are complex machines with key contributions to the smart city. Those devices have limited memory and processing power, but they are not just actuators; they embed tens of sensors and actuators managed by several microcontrollers and microprocessors communicated by control buses. On the other hand, predictive maintenance and the capability of identifying failures to avoid greater damage of machines is becoming a topic of great relevance in Industry 4.0, and the large amount of data to be processed is a concern. This article proposes a layered methodology to enable complex machines with automatic fault detection or predictive maintenance. It presents a layered structure to perform the collection, filtering and extraction of indicators, along with their processing. The aim is to reduce the amount of data to work with, and to optimize them by generating indicators that concentrate the information provided by data. To test its applicability, a prototype of a cash counting machine has been used. With this prototype, different failure cases have been simulated by introducing defective elements. After the extraction of the indicators, using the Kullback–Liebler divergence, it has been possible to visualize the differences between the data associated with normal and failure operation. Subsequently, using a neural network, good results have been obtained, being able to correctly classify the failure in 90% of the cases. The result of this application demonstrates the proper functioning of the proposed approach in complex machines.

This study presents a novel system for diagnosing and evaluating soccer performance using wearable inertial sensors integrated into players’ insoles. Designed to meet the needs of professional podiatrists and sports practitioners, the system focuses on three key soccer-related movements: passing, shooting, and changes of direction (CoDs). The system leverages low-power IMU sensors, Bluetooth Low Energy (BLE) communication, and a cloud-based architecture to enable real-time data analysis and performance feedback. Data were collected from nine professional players from the SD Huesca women’s team during controlled tests, and bespoke algorithms were developed to process kinematic data for precise event detection. Results indicate high accuracy rates for detecting ball-striking events and CoDs, with improvements in algorithm performance achieved through adaptive thresholds and ensemble neural network models. Compared to existing systems, this approach significantly reduces costs and enhances practicality by minimizing the number of sensors required while ensuring real-time evaluation capabilities. However, the study is limited by a small sample size, which restricts generalizability. Future research will aim to expand the dataset, include diverse sports, and integrate additional sensors for broader applications. This system offers a valuable tool for injury prevention, player rehabilitation, and performance optimization in professional soccer, bridging technical advancements with practical applications in sports science.

A novel low-power sensor wireless system for continuous and simultaneous measurement of the electrocardiogram (ECG) and the impedance plethysmogram (IPG) from a single limb has been designed and tested. This system is made up of an analog subsystem composed of two signal-conditioning channels, one for the ECG and one for the IPG that share the same voltage pickup electrodes, and a digital subsystem based on the BGM220PC22WGA microcontroller that is responsible for digitizing the signals and sending their data via Bluetooth Low Energy (BLE) to a processing unit. The consumption of the system has been analyzed using different settings, obtaining values that would allow a battery duration of up to 131 h, by performing four 30-s measurements every hour. The system was tested in one arm and one leg in a group of healthy volunteers to verify the functionality and to identify the best electrode positions to get high quality signals. Non-standard ECG leads are obtained in both one arm and one leg with an SNR of more than 30 dB. In the worst case, for measurements on the legs, the mean error of the RR intervals from each recording was 1.7 ms and the 95% confidence interval (±SD) was ±7.3 ms versus a reference system. This has made it possible to accurately obtain the Pulse Arrival Time (PAT) between the ECG and the simultaneously measured IPG signal, which allows more cardiovascular information to be obtained from the patient than with systems that acquire only the ECG signal.

Scalable tools for Parkinson's disease (PD) detection are needed beyond laboratory-bound assessments. We present and evaluate a deliberately minimalist approach: a single six-axis IMU per shoe coupled with a lightweight 1D-CNN, tested during a six-minute walk test (6MWT) conducted indoors and outdoors under semi-controlled conditions. Eighty-nine participants (44 PD; 45 healthy controls) were instrumented bilaterally at 50 Hz. Signals (accelerometer and gyroscope) were standardized to a common shoe-referenced frame and ingested as raw time series using 0.5-s non-overlapping windows. The network used three tapered convolutional blocks (32-16-8 filters) with global average pooling (2,818 parameters). Evaluation employed participant-level, stratified 5-fold cross-validation repeated across seeds. At the foot level, overall accuracy was 80.58% (sensitivity 74.76%, specificity 85.55%). Applying a clinically motivated one-foot-positive criterion (OFPC), classifying a participant as PD if either foot was positive, improved patient-level performance to 82.1% accuracy, with 84.78% sensitivity and 80.29 specificity. Indoor performance (accuracy 81.48%, sensitivity 91.89%, specificity 72.73%,) exceeded outdoor sensitivity, whereas outdoor testing maintained higher specificity (accuracy 82.17%, sensitivity 80%, specificity 83.91%), reflecting expected context variability. Compact architecture and short, non-overlapping windows minimize memory, computation and latency, supporting on-device deployment on resource-constrained wearables. These results indicate that a foot-mounted, single-modality IMU set-up combined with a compact CNN can deliver clinically meaningful PD discrimination under ecologically relevant conditions, providing a practical pathway toward real-time case identification and follow-up in routine care while preserving scalability and usability.

Developing intelligent sensor-based systems becomes methodologically more demanding as inference moves from external infrastructure towards the Edge, Embedded ML and Ti-nyML, where a model with strong offline accuracy can be unviable on its target device through limits of memory, latency, energy or hardware–software integration. Established machine learning life cycles such as CRISP-DM, CRISP-ML(Q) and MLOps remain largely agnostic to where inference runs and give limited attention to problem definition, sen-sor-data generation and validation under real conditions. This article proposes a method-ological framework for sensor-based systems in which inference is brought close to, or on-to, the device. It was derived through a retrospective cross-case analysis of twelve projects spanning systems without machine learning, Cloud ML, Embedded ML and TinyML, and organises development into four macro-areas, nine phases and four checkpoints. The loca-tion of inference is treated as an explicit axis that reweights decisions, checkpoints act as decision gates, and three assessment levels are distinguished: offline evaluation, on-device technical validation and end-to-end validation. Thirteen international experts appraised the framework: across 114 verification questions, mean relevance was 3.55/4 and the over-all S-CVI/Ave was 0.92, with checkpoints rated highest. The results support its content va-lidity rather than field effectiveness, which remains for prospective evaluation.