Abstract
Photoplethysmography (PPG) is a noninvasive way to monitor various aspects of the circulatory system, and is becoming more and more widespread in biomedical processing. Recently, deep learning methods for analyzing PPG have also become prevalent, achieving state of the art results on heart rate estimation, atrial fibrillation detection, and motion artifact identification. Consequently, a need for interpretable deep learning has arisen within the field of biomedical signal processing. In this paper, we pioneer novel explanatory metrics which leverage domain-expert knowledge to validate a deep learning model. We visualize model attention over a whole testset using saliency methods and compare it to human expert annotations. Congruence, our first metric, measures the proportion of model attention within expert-annotated regions. Our second metric, Annotation Classification, measures how much of the expert annotations our deep learning model pays attention to. Finally, we apply our metrics to compare between a signal based model and an image based model for PPG signal quality classification. Both models are deep convolutional networks based on the ResNet architectures. We show that our signal-based one dimensional model acts in a more explainable manner than our image based model; on average 50.78% of the one dimensional model's attention are within expert annotations, whereas 36.03% of the two dimensional model's attention are within expert annotations. Similarly, when thresholding the one dimensional model attention, one can more accurately predict if each pixel of the PPG is annotated as artifactual by an expert. Through this testcase, we demonstrate how our metrics can provide a quantitative and dataset-wide analysis of how explainable the model is.

Abstract

Abstract
Pressure ulcers are associated with significant morbidity, resulting in a decreased quality of life for the patient, and contributing to healthcare professional burnout, as well as an increase of health service costs. Their prompt diagnosis and treatment are important, and several studies have proposed solutions to help healthcare professionals in this process. This work analyzes studies that use machine-learning algorithms for risk assessment and management of preventive treatments for pressure ulcers. More specifically, it focuses on the use of machine-learning algorithms that combine information from intrinsic and extrinsic pressure-ulcer predisposing factors to produce recommendations/alerts to healthcare professionals. The review includes articles published from January 2010 to June 2021. From 60 records screened, seven articles were analyzed in full-text form. The results show that most of the proposed algorithms do not use information related to both intrinsic and extrinsic predisposing factors and that many of the approaches separately address one of the following three components: data acquisition; data analysis, and production of complementary support to well-informed clinical decision-making. Additionally, only a few studies describe in detail the outputs of the algorithm, such as alerts and recommendations, without assessing their impacts on healthcare professionals' activities.

Abstract
This experimental study focuses on the comparison between two different sensors for vibration signals: a magnetoresistive sensor and an accelerometer as a calibrated reference. The vibrations are collected from a variable speed inductor motor setup, coupled to a ball bearing load with adjustable misalignments. To evaluate the performance of the magnetoresistive sensor against the accelerometer, several vibration measurements are performed in three different axes: axial, horizontal and vertical. Vibration velocity measurements from both sensors were collected and analyzed based on spectral decomposition of the signals. The high cross-correlation coefficient between spectrum vibration signatures in all experimental measurements shows good agreement between the proposed magnetoresistive sensor and the reference accelerometer performances. The results demonstrate the potential of this type of innovative and non-contact approach to vibration data collection and a prospective use of magnetoresistive sensors for predictive maintenance models for inductive motors in Industry 4.0 applications.

Abstract
Although previous research on the e-learning system acceptance has been conducted usingUTAUT, no study followed the longitudinal approach. Accordingly, this research examines the engineering students' (N = 291) e-learning system acceptance by three years of study. The structural equation modelling analysis confirmed UTAUT relationships in each year. Effort expectancy and social influence resulted as significant predictors of behavioural intention in all three years. In contrast, performance expectancy influence got lower in later usage. Altogether, our longitudinal study presented that the UTAUT model has weakened over time. Therefore, we propose extending the UTAUT model in future research to better understand user satisfaction and positively contribute to system acceptance. Our research findings can be used for university leaders to investigate and evaluate any implemented information system acceptance through the years.

Abstract
This paper presents the initial developments of new hardware devices targeted for CAN (Controller Area Network) bus communications in forest machines. CAN bus is a widely used protocol for communications in the automobile area. It is also applied in industrial vehicles and machines due to its robustness, simplicity, and operating flexibility. It is ideal for forestry machinery producers who need to couple their equipment to a machine that allows the transportation industry to recognize the importance of standardizing communications between tools and machines. One of the problems that producers sometimes face is a lack of flexibility in commercialized hardware modules; for example, in interfaces for sensors and actuators that guarantee scalability depending on the new functionalities required. The hardware device presented in this work is designed to overcome these limitations and provide the flexibility to standardize communications while allowing scalability in the development of new products and features. The work is being developed within the scope of the research project "SMARTCUT-Remote Diagnosis, Maintenance and Simulators for Operation Training and Maintenance of Forest Machines ", to incorporate innovative technologies in forest machines produced by the CUTPLANT S.A. It consists of an experimental system based on the PIC18F26K83 microcontroller to form a CAN node to transmit and receive digital and analog messages via CAN bus, tested and validated by the communication between different nodes. The main contribution of the paper focuses on the presentation of the development of new CAN bus electronic control units designed to enable remote communication between sensors and actuators, and the main controller of forest machines.