Search for: All records

Foraminifera play an important role in oceanographic and paleoceanographic research. The test morphology and chemistry within species, as well as the presence or absence of certain species, are affected by environmental conditions. Classification of different species of foraminifera is a crucial yet tedious task for researchers. Deep-learning approaches can help with morphological studies and aid in species classification; however, they require large-scale datasets that are challenging to obtain and annotate because of the extremely small size and delicate handling of these microorganisms. In this work, we expand on an existing mathematical model for foraminifera shell growth to generate 3D synthetic models to aid in these studies. We define parameter spaces for the model which are intended to approximate seven randomly chosen foraminifera taxa. Along with providing an open-source code base to support other researchers in generating models and studying growth patterns, we further extend the synthetic data generation to include a rendering component that mimics two existing robotic imaging systems. We provide two use cases for our synthetic dataset. First, we show how orientation can affect the automated classification of different species and how incorporating aleatoric uncertainty indicators can help select the next views of the samples to significantly improve classification accuracy from 82% to 89%. Next, we show how a sparse set of synthetic 2D images can be used to extract 3D morphology of foraminifera using Neural Radiance Fields (NeRFs). 

Dementia is primarily caused by neurodegenerative diseases like Alzheimer’s disease (AD). It affects millions worldwide, making detection and monitoring crucial. This study focuses on the detection of dementia from speech transcripts of controls and dementia groups. We propose encoding in-text pauses and filler words (e.g., “uh” and “um”) in text-based language models and thoroughly evaluating their impact on performance (e.g., accuracy). Additionally, we suggest using contrastive learning to improve performance in a multi-task framework. Our results demonstrate the effectiveness of our approaches in enhancing the model’s performance, achieving 87% accuracy and an 86% f1-score. Compared to the state of the art, our approach has similar performance despite having significantly fewer parameters. This highlights the importance of pause and filler word encoding on the detection of dementia. 

Environmental factors, such as drought stress, significantly impact maize growth and productivity worldwide. To improve yield and quality, effective strategies for early detection and mitigation of drought stress in maize are essential. This paper presents a detailed analysis of three imaging trials conducted to detect drought stress in maize plants using an existing, custom-developed, low-cost, high-throughput phenotyping platform. A pipeline is proposed for early detection of water stress in maize plants using a Vision Transformer classifier and analysis of distributions of near-infrared (NIR) reflectance from the plants. A classification accuracy of 85% was achieved in one of our trials, using hold-out trials for testing. Suitable regions on the plant that are more sensitive to drought stress were explored, and it was shown that the region surrounding the youngest expanding leaf (YEL) and the stem can be used as a more consistent alternative to analysis involving just the YEL. Experiments in search of an ideal window size showed that small bounding boxes surrounding the YEL and the stem area of the plant perform better in separating drought-stressed and well-watered plants than larger window sizes enclosing most of the plant. The results presented in this work show good separation between well-watered and drought-stressed categories for two out of the three imaging trials, both in terms of classification accuracy from data-driven features as well as through analysis of histograms of NIR reflectance. 

Sleep staging has a very important role in diagnosing patients with sleep disorders. In general, this task is very time-consuming for physicians to perform. Deep learning shows great potential to automate this process and remove physician bias from decision making. In this study, we aim to identify recent trends on performance improvement and the causes for these trends. Recent papers on sleep stage classification and interpretability are investigated to explore different modeling and data manipulation techniques, their efficiency, and recent advances. We identify an improvement in performance up to 12% on standard datasets over the last 5 years. The improvements in performance do not appear to be necessarily correlated to the size of the models, but instead seem to be caused by incorporating new architectural components, such as the use of transformers and contrastive learning.