In light of our aging population, there is an immediate need for non-obtrusive, continuous, and ubiquitous health monitoring technologies that will enable our population to age with a higher quality of life and independence. Research has demonstrated that gait indicators, such as walking speed, can reflect cognitive and physical functioning. However, gradual changes in such indicators usually go undetected until critical problems arise; being able to detect changes in indicators, such as gait deterioration, of older adults while in their home environments would enable clinicians to tailor more effective and personalized interventions by better understanding user behaviour in real-world settings. Real-world data is essential to enabling our healthcare system to act where patients most need help and to optimize the effect of designed eHealth solutions.

Background/Objectives: Fenbendazole is a potential cancer treatment and a proven antiparasitic in veterinary applications. However, its poor water solubility limits its application. In this study, potential fenbendazole solubility enhancement was investigated through size reduction methods. The effect of the presence of Soluplus^® on solubility was investigated as well. Methods: Solubility enhancement was explored using microfluidization and ultrasonication techniques. These techniques were applied to fenbendazole alone and in combination with Soluplus^®. UV–Vis spectroscopy was used to determine solubility. Possible chemical reactions were checked using Fourier transform infrared spectroscopy (FT-IR). Differential scanning calorimetry (DSC) was conducted to analyze the physical structure and crystallinity of the samples. Scanning electron microscopy (SEM) was also utilized for characterization of the effect of the treated formulations and the size reduction method on morphology. The elements present in samples were identified with energy-dispersive X-ray spectroscopy (EDX) combined with SEM. A comparison of crystalline structure between the products was performed via X-ray powder diffraction (XRPD). Dynamic light scattering (DLS) was also used to measure the samples’ average particle size at different stages. Results: Both ultrasonication and microfluidization led to marginal increases in the solubility of neat fenbendazole. In contrast, formulations processed in the presence of Soluplus^® demonstrated a greater enhancement in solubility. However, solubility improvement was not retained in the dried samples. The post-drying samples, irrespective of the presence of Soluplus^®, showed nearly the same solubility as neat fenbendazole. Conclusions: Size-reduction methods, particularly when combined with Soluplus^®, improved the solubility of fenbendazole. However, drying appeared to reverse these gains, regardless of the method used.

Fenbendazole is an important anti-parasitic medicine widely used in the veterinary field and has recently been considered as a possible anti-cancer agent in humans by some researchers. Fenbendazole encounters challenges in its usage due to its limited aqueous solubility, which consequently impacts its therapeutic efficacy. In this work, an in vitro mechanistic investigation was conducted to evaluate the compatibility, amorphization behaviour and dissolution profile of fenbendazole dispersed in Soluplus^® using the solid dispersion approach via hot-melt extrusion. Three different fenbendazole/Soluplus^® ratios were formulated and characterised through systematic experimentation. Powder X-Ray Diffraction (PXRD), Differential Scanning Calorimetry (DSC), Scanning Electron Microscopy (SEM), Energy Dispersive X-Ray (EDX) and Fourier Transform Infrared Spectroscopy (FTIR) were employed for thermal, physical, chemical and morphological analyses. The solubility of the drug formulation during a dissolution test was investigated using Ultraviolet–Visible (UV–Vis) spectrophotometric measurements. In vitro dissolution testing in acidic and neutral media was employed as a controlled environment to compare dissolution behaviour among different loadings. The extrudates demonstrated markedly enhanced apparent solubility compared to neat fenbendazole, with the 5% formulation showing the highest dissolution rate (approximately 85% after 48 h). This improvement can be attributed to better wetting properties and drug dispersion within the Soluplus^® matrix. This innovative strategy holds promise in surmounting fenbendazole’s solubility limitations, presenting a comprehensive solution to enhance its therapeutic effectiveness.

Supervised dimensionality reduction strategies have been of great interest. However, current supervised dimensionality reduction approaches are difficult to scale for situations characterized by large datasets given the high computational complexities associated with such methods. While stochastic approximation strategies have been explored for unsupervised dimensionality reduction to tackle this challenge, such approaches are not well-suited for accelerating computational speed for supervised dimensionality reduction. Motivated to tackle this challenge, in this study we explore a novel direction of directly learning optimal class-aware embeddings in a supervised manner via the notion of supervised random projections (SRP). The key idea behind SRP is that, rather than performing spectral decomposition (or approximations thereof) which are computationally prohibitive for large-scale data, we instead perform a direct decomposition by leveraging kernel approximation theory and the symmetry of the Hilbert-Schmidt Independence Criterion (HSIC) measure of dependence between the embedded data and the labels. Experimental results on five different synthetic and real-world datasets demonstrate that the proposed SRP strategy for class-aware embedding learning can be very promising in producing embeddings that are highly competitive with existing supervised dimensionality reduction methods (e.g., SPCA and KSPCA) while achieving 1-2 orders of magnitude better computational performance. As such, such an efficient approach to learning embeddings for dimensionality reduction can be a powerful tool for large-scale data analysis and visualization.

The problem of feature disentanglement has been explored in the literature, for the purpose of image and video processing and text analysis. State-of-the-art methods for disentangling feature representations rely on the presence of many labeled samples. In this work, we present a novel method for disentangling factors of variation in data-scarce regimes. Specifically, we explore the application of feature disentangling for the problem of supervised classification in a setting where few labeled samples exist, and there are no unlabeled samples for use in unsupervised training. Instead, a similar datasets exists which shares at least one direction of variation with the sample-constrained datasets. We train our model end-to-end using the framework of variational autoencoders and are able to experimentally demonstrate that using an auxiliary dataset with similar variation factors contribute positively to classification performance, yielding competitive results with the state-of-the-art in unsupervised learning.