Inside out: transforming images of lab-grown plants for machine learning applications in agriculture.

Introduction: Machine learning tasks often require a significant amount of training data for the resultant network to perform suitably for a given problem in any domain. In agriculture, dataset sizes are further limited by phenotypical differences between two plants of the same genotype, often as a result of different growing conditions. Synthetically-augmented datasets have shown promise in improving existing models when real data is not available.

Generating accurate tip angles for NMR outside the rotating-wave approximation.

The generation of accurate tip angles is critical for many applications of nuclear magnetic resonance. In low static field, with a linear rather than circular polarized rf field, the rotating-wave approximation may no longer hold and significant deviations from expected trajectories on the Bloch sphere can occur. For rectangular rf pulses, the effects depend strongly on the phase of the rf field and can be further compounded by transients at the start and end of the pulse.

Leveraging Guided Backpropagation to Select Convolutional Neural Networks for Plant Classification.

The development of state-of-the-art convolutional neural networks (CNN) has allowed researchers to perform plant classification tasks previously thought impossible and rely on human judgment. Researchers often develop complex CNN models to achieve better performances, introducing over-parameterization and forcing the model to overfit on a training dataset. The most popular process for evaluating overfitting in a deep learning model is using accuracy and loss curves.

An embedded system for the automated generation of labeled plant images to enable machine learning applications in agriculture.

A lack of sufficient training data, both in terms of variety and quantity, is often the bottleneck in the development of machine learning (ML) applications in any domain. For agricultural applications, ML-based models designed to perform tasks such as autonomous plant classification will typically be coupled to just one or perhaps a few plant species. As a consequence, each crop-specific task is very likely to require its own specialized training data, and the question of how to serve this need for data now often overshadows the more routine exercise of actually training such models.

A fast MOSFET rf switch for low-field NMR and MRI.

TRansmit Array Spatial Encoding (TRASE) MRI uses trains of rf pulses alternately produced by distinct transmit coils. Commonly used coil switching involving PIN diodes is too slow for low- and ultra-low-field MRI and would introduce wait times between pulses typically as long as each individual pulse in a few mT. A MOSFET-based rf switch is described and characterised.

Assessing the accuracy of using oscillating gradient spin echo sequences with AxCaliber to infer micron-sized axon diameters.

Objective: Current magnetic resonance imaging (MRI) axon diameter measurements rely on the pulsed gradient spin-echo sequence, which is unable to provide diffusion times short enough to measure small axon diameters. This study combines the AxCaliber axon diameter fitting method with data generated from Monte Carlo simulations of oscillating gradient spin-echo sequences (OGSE) to infer micron-sized axon diameters, in order to determine the feasibility of using MRI to infer smaller axon diameters in brain tissue.

Materials And Methods: Monte Carlo computer simulation data were synthesized from tissue geometries of cylinders of different diameters using a range of gradient frequencies in the cosine OGSE sequence .

4D MR phase and magnitude segmentations with GPU parallel computing.

The increasing size and number of data sets of large four dimensional (three spatial, one temporal) magnetic resonance (MR) cardiac images necessitates efficient segmentation algorithms. Analysis of phase-contrast MR images yields cardiac flow information which can be manipulated to produce accurate segmentations of the aorta. Phase contrast segmentation algorithms are proposed that use simple mean-based calculations and least mean squared curve fitting techniques.

Measurement of alveolar oxygen partial pressure in the rat lung using Carr-Purcell-Meiboom-Gill spin-spin relaxation times of hyperpolarized 3He and 129Xe at 74 mT.

Regional measurement of alveolar oxygen partial pressure can be obtained from the relaxation rates of hyperpolarized noble gases, (3) He and (129) Xe, in the lungs. Recently, it has been demonstrated that measurements of alveolar oxygen partial pressure can be obtained using the spin-spin relaxation rate (R(2) ) of (3) He at low magnetic field strengths (<0.1 T) in vivo.

In vivo NMR of hyperpolarized 3He in the human lung at very low magnetic fields.

We present NMR measurements of the diffusion of hyperpolarized 3He in the human lung performed at fields much lower than those of conventional MRI scanners. The measurements were made on standing subjects using homebuilt apparatus operating at 3mT. O(2)-limited transverse relaxation (T(2) up to 15-35s) could be measured in vivo.