Beehive health monitoring has gained interest in the study of bees in biology, ecology, and agriculture. As audio sensors are less intrusive, a number of audio datasets (mainly labeled with the presence of a queen in the hive) have appeared in the literature, and interest in their classification has been raised. All studies have exhibited good accuracy, and a few have questioned and revealed that classification cannot be generalized to unseen hives. To increase the number of known hives, a review of open datasets is described, and a merger in the form of the “BeeTogether” dataset on the open Kaggle platform is proposed. This common framework standardizes the data format and features while providing data augmentation techniques and a methodology for measuring hives’ extrapolation properties. A classical classifier is proposed to benchmark the whole dataset, achieving the same good accuracy and poor hive generalization as those found in the literature. Insight into the role of the frequency of the classification of the presence of a queen is provided, and it is shown that this frequency mostly depends on a colony’s belonging. New classifiers inspired by contrastive learning are introduced to circumvent the effect of colony belonging and obtain both good accuracy and hive extrapolation abilities when learning changes in labels. A process for obtaining absolute labels was prototyped on an unsupervised dataset. Solving hive extrapolation with a common open platform and contrastive approach can result in effective applications in agriculture.

This paper aims to provide a review of the electrically assisted bicycles (also known as e-bikes) used for recovery of the rider’s physical and physiological information, monitoring of their health state, and adjusting the “medical” assistance accordingly. E-bikes have proven to be an excellent way to do physical activity while commuting, thus improving the user’s health and reducing air pollutant emissions. Such devices can also be seen as the first step to help unhealthy sedentary people to start exercising with reduced strain. Based on this analysis, the need to have e-bikes with artificial intelligence (AI) systems that recover and processe a large amount of data is discussed in depth. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were used to complete the relevant papers’ search and selection in this systematic review.

This paper presents the development of an electronic system that converts an electrically assisted bicycle into an intelligent health monitoring system, allowing people who are not athletic or who have a history of health issues to progressively start the physical activity by following a medical protocol (e.g., max heart rate and power output, training time). The developed system aims to monitor the health state of the rider, analyze data in real-time, and provide electric assistance, thus diminishing muscular exertion. Furthermore, such a system can recover the same physiological data used in medical centers and program it into the e-bike to track the patient’s health. System validation is conducted by replicating a standard medical protocol used in physiotherapy centers and hospitals, typically conducted in indoor conditions. However, the presented work differentiates itself by implementing this protocol in outdoor environments, which is impossible with the equipment used in medical centers. The experimental results show that the developed electronic prototypes and the algorithm effectively monitored the subject’s physiological condition. Moreover, when necessary, the system can change the training load and help the subject remain in their prescribed cardiac zone. This system allows whoever needs to follow a rehabilitation program to do so not only in their physician’s office, but whenever they want, including while commuting.