Bridging the Gap between Field Experiments and Machine Learning: The EC H2020 B-GOOD Project as a Case Study towards Automated Predictive Health Monitoring of Honey Bee Colonies.

Honey bee colonies have great societal and economic importance. The main challenge that beekeepers face is keeping bee colonies healthy under ever-changing environmental conditions. In the past two decades, beekeepers that manage colonies of Western honey bees (Apis mellifera) have become increasingly concerned by the presence of parasites and pathogens affecting the bees, the reduction in pollen and nectar availability, and the colonies' exposure to pesticides, among others. Hence, beekeepers need to know the health condition of their colonies and how to keep them alive and thriving, which creates a need for a new holistic data collection method to harmonize the flow of information from various sources that can be linked at the colony level for different health determinants, such as bee colony, environmental, socioeconomic, and genetic statuses. For this purpose, we have developed and implemented the B-GOOD (Giving Beekeeping Guidance by computational-assisted Decision Making) project as a case study to categorize the colony's health condition and find a Health Status Index (HSI). Using a 3-tier setup guided by work plans and standardized protocols, we have collected data from inside the colonies (amount of brood, disease load, honey harvest, etc.) and from their environment (floral resource availability). Most of the project's data was automatically collected by the BEEP Base Sensor System. This continuous stream of data served as the basis to determine and validate an algorithm to calculate the HSI using machine learning. In this article, we share our insights on this holistic methodology and also highlight the importance of using a standardized data language to increase the compatibility between different current and future studies. We argue that the combined management of big data will be an essential building block in the development of targeted guidance for beekeepers and for the future of sustainable beekeeping.

Bacterially synthesized ferrite nanoparticles for magnetic hyperthermia applications.

Magnetic hyperthermia uses AC stimulation of magnetic nanoparticles to generate heat for cancer cell destruction. Whilst nanoparticles produced inside magnetotactic bacteria have shown amongst the highest reported heating to date, these particles are magnetically blocked so that strong heating occurs only for mobile particles, unless magnetic field parameters are far outside clinical limits. Here, nanoparticles extracellularly produced by the bacteria Geobacter sulfurreducens are investigated that contain Co or Zn dopants to tune the magnetic anisotropy, saturation magnetization and nanoparticle sizes, enabling heating within clinical field constraints. The heating mechanisms specific to either Co or Zn doping are determined from frequency dependent specific absorption rate (SAR) measurements and innovative AC susceptometry simulations that use a realistic model concerning clusters of polydisperse nanoparticles in suspension. Whilst both particle types undergo magnetization relaxation and show heating effects in water under low AC frequency and field, only Zn doped particles maintain relaxation combined with hysteresis losses even when immobilized. This magnetic heating process could prove important in the biological environment where nanoparticle mobility may not be possible. Obtained SARs are discussed regarding clinical conditions which, together with their enhanced MRI contrast, indicate that biogenic Zn doped particles are promising for combined diagnostics and cancer therapy.

Detection of virgin olive oil adulteration using low field unilateral NMR.

The detection of adulteration in edible oils is a concern in the food industry, especially for the higher priced virgin olive oils. This article presents a low field unilateral nuclear magnetic resonance (NMR) method for the detection of the adulteration of virgin olive oil that can be performed through sealed bottles providing a non-destructive screening technique. Adulterations of an extra virgin olive oil with different percentages of sunflower oil and red palm oil were measured with a commercial unilateral instrument, the profile NMR-Mouse. The NMR signal was processed using a 2-dimensional Inverse Laplace transformation to analyze the transverse relaxation and self-diffusion behaviors of different oils. The obtained results demonstrated the feasibility of detecting adulterations of olive oil with percentages of at least 10% of sunflower and red palm oils.

Reducing peripheral nerve stimulation due to gradient switching using an additional uniform field coil.

This study shows that larger rates of change of gradient with time (dG/dt) can be achieved at the threshold for peripheral nerve stimulation by applying a uniform concomitant field varying synchronously with a transverse field gradient and that this increase may be achieved without significant reduction of the spatial extent of the region over which imaging can be carried out. Realization of similar benefits through application of a uniform, z-directed field varying synchronously with an axial gradient is also demonstrated. The design and construction of transverse and axial coil arrangements is described, along with the results of volunteer studies that were carried out on 20 subjects, with the subjects positioned with four different regions (head, heart, hips, and knees) centered in the coils. These experiments were carried out at zero-field on a prototype system in which the coils were not actively shielded. The uniform concomitant field coil was not torque balanced. The increase in the rate of change of gradient at the threshold for peripheral nerve stimulation that could be achieved by the addition of the uniform field depends on body position and was larger for the transverse coils (head = × 1.9 ± 0.6; heart = × 0.9 ± 0.3; hips = × 1.4 ± 0.4; knees = × 1.5 ± 0.4) than for the axial coils (head = × 1.5 ± 0.6; heart = × 0.8 ± 0.3; hips = × 1.3 ± 0.4; knees = × 1.1 ± 0.3).