arXiv subjects

Alexandra E. Porter

Publications and source records attributed to Alexandra E. Porter.

4 recordsLinked to original sources

arXiv2026

An AI-driven framework for the prediction of personalised health response to air pollution

Air pollution is a growing global health threat, exacerbated by climate change and linked to cardiovascular and respiratory diseases. While personal sensing devices enable real-time physiological monitoring, their integration with environmental data for individualised health prediction remains underdeveloped. Here, we present a modular, cloud-based framework that predicts personalised physiological responses to pollution by combining wearable-derived data with real-time environmental exposures. At its core is an Adversarial Autoencoder (AAE), initially trained on high-resolution pollution-health data from the INHALE study and fine-tuned using smartwatch data via transfer learning to capture individual-specific patterns. Consistent with changes in pollution levels commonly observed in the real-world, simulated pollution spikes (+100%) revealed modest but measurable increases in vital signs (e.g., +2.5% heart rate, +3.5% breathing rate). To assess clinical relevance, we analysed U-BIOPRED data and found that individuals with such subclinical vital sign elevations had higher asthma burden scores or elevated Fractional Exhaled Nitric Oxide (FeNO), supporting the physiological validity of these AI-predicted responses. This integrative approach demonstrates the feasibility of anticipatory, personalised health modelling in response to environmental challenges, offering a scalable and secure infrastructure for AI-driven environmental health monitoring.

cs.LG↗

arXiv2025

Advancing atom probe tomography capabilities to understand bone microstructures at the near-atomic scale

Bone structure is generally hierarchically organized into organic (collagen, proteins,...), inorganic (hydroxyapatite (HAP)) components. However, many fundamental mechanisms of the biomineralization processes such as HAP formation, the influence of trace elements, the mineral-collagen arrangement, etc., are not clearly understood. This is partly due to the analytical challenge of simultaneously characterizing the three-dimensional (3D) structure and chemical composition of biominerals in general at the nanometer scale, which can, in principle be achieved by atom probe tomography (APT). Yet, the hierarchical structures of bone represent a critical hurdle for APT analysis in terms of sample yield and analytical resolution, particularly for trace elements, and organic components from the collagen appear to systematically get lost from the analysis. Here, we applied in-situ metallic coating of APT specimens within the focused ion beam (FIB) used for preparing specimens, and demonstrate that the sample yield and chemical sensitivity are tremendously improved, allowing the analysis of individual collagen fibrils and trace elements such as Mg and Na. We explored a range of measurement parameters with and without coating, in terms of analytical resolution performance and determined the best practice parameters for analyzing bone samples in APT. To decipher the complex mass spectra of the bone specimens, reference spectra from pure HAP and collagen were acquired to unambiguously identify the signals, allowing us to analyze entire collagen fibrils and interfaces at the near-atomic scale. Our results open new possibilities for understanding the hierarchical structure and chemical heterogeneity of bone structures at the near-atomic level and demonstrate the potential of this new method to provide new, unexplored insights into biomineralization processes in the future.

cond-mat.mtrl-sci↗

arXiv2020

Label-Free Chemical Nano-Imaging of Intracellular Drug Binding Sites

Optical microscopy has a diffraction limited resolution of about 250 nm. Fluorescence methods (e.g. PALM, STORM, STED) beat this, but they are still limited to 10 s of nm, and the images are an indirect pointillist representation of only part of the original object. Here we describe a way of combining a sample preparation technique taken from histopathology, with a probe-based nano-imaging technique, (s SNOM) from the world of Solid State Physics. This allows us to image subcellular structures optically, and at a nanoscale resolution that is about 100 x better than normal microscopes. By adding a tuneable laser source, we also demonstrate mid-infrared chemical nano-imaging (MICHNI) in human myeloma cells and we use it to map the binding sites of the anti cancer drug bortezomib to less than 10 zL sized intracellular components. MICHNI is label free and can be used with any biological material and drugs with specific functional chemistry. We believe that its combination of speed, cheapness, simplicity, safety and chemical contrast promises a transformative impact across the life sciences.

physics.bio-ph↗

arXiv2018

Mid-infrared Chemical Nano-imaging for Intra-cellular Drug Localisation

In the past two decades a range of fluorescence cell microscopy techniques have been developed which can achieve ~10 nm spatial resolution, i.e. substantially beating the usual limits set by optical diffraction. However, these methods rely on specialised labelling. This limits the applicability, risks perturbing the biology, and it also makes them so-called "discovery techniques" that can only be used when there is prior knowledge about the biological problem. The alternative, electron microscopy (EM), requires complex and time-consuming sample preparation, that risks compromising the sample's integrity. Samples have to withstand vacuum, and staining with heavy metals to make them conductive, and give usable electron-contrast. None of these techniques can directly map out drug distributions at a sub-cellular level. Recently infrared light-based scanning probe techniques have demonstrated a capability for ~1 nm spatial resolution. However, they need samples that are flat, dry and dimensionally stable and they only probe down to a depth commensurate with the spatial resolution, so they yield essentially surface chemical information. Thus far they have been applied only to artificially produced test samples, e.g. gold particles, or isolated proteins on silicon. Here we show how these probe-based techniques can be adapted for use with routinely prepared general biological specimens. This allows for "Mid-infrared Chemical Nano-imaging" (MICHNI) that delivers chemical analysis at a ~10 nm spatial resolution, suitable for studying cellular ultrastructure. We demonstrate its utility by performing label-free mapping of the anti-cancer drug Bortezomib (BTZ) within a single human myeloma cell. We believe that this MICHNI technique has the potential to become a widely applicable adjunct to EM across the bio-sciences.

physics.bio-ph↗