Limitations of effective medium models for tissue phantoms in the THz frequency range.

The absorption of terahertz (THz) radiation by water molecules facilitates its application to several biomedical applications such as cancer detection. Therefore, it is critical for the THz technologies to be characterised with water content in a sample. In this paper, we analyse gelatine phantoms in the THz frequency range, with continuously varying hydration levels as they dry over time.

Use of fidget and drinking behaviour in combination with facial infrared thermography for diagnosis of bovine respiratory disease in a spontaneous model.

Bovine respiratory disease (BRD) is a highly prevalent multi pathogen infectious disease (70-80%) in newly received feedlot cattle, causing significant economic losses and reduced animal welfare. Current BRD diagnosis involves stressful and invasive methods that can increase the incidence and transmission of BRD. An alternative is the use of an automated infrared thermography (IR) platform that can monitor facial temperature and behaviour traits to diagnose BRD in a non-invasive manner.

Large-scale electrophysiology and deep learning reveal distorted neural signal dynamics after hearing loss.

Listeners with hearing loss often struggle to understand speech in noise, even with a hearing aid. To better understand the auditory processing deficits that underlie this problem, we made large-scale brain recordings from gerbils, a common animal model for human hearing, while presenting a large database of speech and noise sounds. We first used manifold learning to identify the neural subspace in which speech is encoded and found that it is low-dimensional and that the dynamics within it are profoundly distorted by hearing loss.

Unsupervised segmentation of biomedical hyperspectral image data: tackling high dimensionality with convolutional autoencoders.

Information about the structure and composition of biopsy specimens can assist in disease monitoring and diagnosis. In principle, this can be acquired from Raman and infrared (IR) hyperspectral images (HSIs) that encode information about how a sample's constituent molecules are arranged in space. Each tissue section/component is defined by a unique combination of spatial and spectral features, but given the high dimensionality of HSI datasets, extracting and utilising them to segment images is non-trivial.

Evaluating automated infrared thermography and vulva exposure tracking as components of an estrus detection platform in a commercial dairy herd.

The primary objective of this study was to develop an automated infrared thermography platform (Estrus BenchMark) capable of measuring skin temperature and tail movements as a means of identifying cows in estrus. The secondary objective was to evaluate the accuracy of Estrus BenchMark to detect estrus compared to in-line milk progesterone (P) analysis (Herd Navigator System) in a commercial dairy herd managed under a robotic milking system. Data were collected on forty-six cows from 45 to 120 d after calving.