It is possible to have a higher efficient use of water in a greenhouse because there is a better control of environmental conditions for crop production (^{Costa et al., 2018}), including tomato as one of the most cultivated vegetables due to its profitability and its consumption (^{Krause et al., 2017}).

One of the main challenges agricultural greenhouses face is accurately predicting environmental conditions to ensure optimal crop growth. However, the current prediction methods have limitations in handling large volumes of dynamic and nonlinear temporal data, which makes it difficult to make accurate early predictions. This paper aims to forecast a greenhouse’s internal temperature up to one hour in advance using supervised learning tools like Extreme Gradient Boosting (XGBoost) and Recurrent Neural Networks combined with Long-Short Term Memory (LSTM-RNN). The study uses the many-to-one configuration, with a sequence of three input elements and one output element. Significant improvements in the R², RMSE, MAE, and MAPE metrics are observed by considering various combinations. In addition, Bayesian optimization is employed to find the best hyperparameters for each algorithm. The research uses a database of internal data such as temperature, humidity, and dew point and external data such as temperature, humidity, and solar radiation, splitting the data into the year’s four seasons and performing eight experiments according to the two algorithms and each season. The LSTM-RNN model produces the best results for the metrics in summer, achieving an R² = 0.9994, RMSE = 0.2698, MAE = 0.1449, and MAPE = 0.0041, meeting the acceptability criterion of ±2 °C hysteresis.

Breast cancer is one of the diseases of most profound concern, with the most prevalence worldwide, where early detections and diagnoses play the leading role against this disease achieved through imaging techniques such as mammography. Radiologists tend to have a high false positive rate for mammography diagnoses and an accuracy of around 82%. Currently, deep learning (DL) techniques have shown promising results in the early detection of breast cancer by generating computer-aided diagnosis (CAD) systems implementing convolutional neural networks (CNNs). This work focuses on applying, evaluating, and comparing the architectures: AlexNet, GoogLeNet, Resnet50, and Vgg19 to classify breast lesions after using transfer learning with fine-tuning and training the CNN with regions extracted from the MIAS and INbreast databases. We analyzed 14 classifiers, involving 4 classes as several researches have done it before, corresponding to benign and malignant microcalcifications and masses, and as our main contribution, we also added a 5th class for the normal tissue of the mammary parenchyma increasing the correct detection; in order to evaluate the architectures with a statistical analysis based on the received operational characteristics (ROC), the area under the curve (AUC), F1 Score, accuracy, precision, sensitivity, and specificity. We generate the best results with the CNN GoogLeNet trained with five classes on a balanced database with an AUC of 99.29%, F1 Score of 91.92%, the accuracy of 91.92%, precision of 92.15%, sensitivity of 91.70%, and specificity of 97.66%, concluding that GoogLeNet is optimal as a classifier in a CAD system to deal with breast cancer.

Remote laboratories are essential in addressing access and quality challenges in technical education. They enable students from various locations to engage with real equipment, overcome geographic and economic constraints, and provide solutions during crises, such as pandemics, when in-person learning is limited. As a key element of Education 4.0, remote labs promote technical skill development, enhance engineering education, and support diverse learning approaches. This study presents a remote laboratory based on Field Programmable Gate Arrays (FPGAs), developed using a waterfall methodology integrating IoT and Cloud Computing technologies to facilitate close interaction between hardware and software. The lab focuses on controlling DC, servo, and stepper motors, allowing students to apply theoretical concepts such as digital signals, pulse-width modulation (PWM), and data representation in bits in a practical setting. The testing phase involved 50 robotics and mechatronics engineering students who participated in hands-on sessions for one month, followed by a structured survey evaluating their experience, interaction, and the educational relevance of the platform. The survey shows high student satisfaction, highlighting the platform’s strengths and identifying areas for improvement. The results also underscore the system’s potential to significantly enhance the educational experience in remote environments, aligning with the United Nations Sustainable Development Goals (SDGs).

The accurate measurement of solar radiation is essential for applications in agriculture, renewable energy, and environmental monitoring. Traditional pyranometers provide high-precision readings but are often costly and inaccessible for large-scale deployment. This study explores the feasibility of using low-cost ambient light sensors combined with statistical and machine learning models based on linear, random forest, and support vector regressions to estimate solar irradiance. To achieve this, an Internet of Things-based system was developed, integrating the light sensors with cloud storage and processing capabilities. A dedicated solar radiation sensor (Davis 6450) served as a reference, and results were validated against meteorological API data. Experimental validation demonstrated a strong correlation between sensor-measured illuminance and solar irradiance using the random forest model, achieving a coefficient of determination (R²) of 0.9922, a root mean squared error (RMSE) of 44.46 W/m², and a mean absolute error (MAE) of 27.12 W/m². These results suggest that low-cost light sensors, when combined with data-driven models, offer a viable and scalable solution for solar radiation monitoring, particularly in resource-limited regions.