Specialized domain knowledge is often necessary to accurately annotate training sets for in-depth analysis, but can be burdensome and time-consuming to acquire from domain experts. This issue arises prominently in automated behavior analysis, in which agent movements or actions of interest are detected from video tracking data. To reduce annotation effort, we present TREBA: a method to learn annotation-sample efficient trajectory embedding for behavior analysis, based on multi-task self-supervised learning. The tasks in our method can be efficiently engineered by domain experts through a process we call "task programming", which uses programs to explicitly encode structured knowledge from domain experts. Total domain expert effort can be reduced by exchanging data annotation time for the construction of a small number of programmed tasks. We evaluate this trade-off using data from behavioral neuroscience, in which specialized domain knowledge is used to identify behaviors. We present experimental results in three datasets across two domains: mice and fruit flies. Using embeddings from TREBA, we reduce annotation burden by up to a factor of 10 without compromising accuracy compared to state-of-the-art features. Our results thus suggest that task programming and self-supervision can be an effective way to reduce annotation effort for domain experts.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9766046 | PMC |
http://dx.doi.org/10.1109/cvpr46437.2021.00290 | DOI Listing |
Global disparities in neurosurgical care necessitate innovations addressing affordability and accuracy, particularly for critical procedures like ventriculostomy. This intervention, vital for managing life-threatening intracranial pressure increases, is associated with catheter misplacement rates exceeding 30% when using a freehand technique. Such misplacements hold severe consequences including haemorrhage, infection, prolonged hospital stays, and even morbidity and mortality.
View Article and Find Full Text PDFSci Rep
January 2025
Department of Sociology and Social Research, University of Trento, Trento, Italy.
This study investigates the dissemination of archaeological information on Twitter/X through the lens of cultural evolution. By analysing 132,230 tweets containing the hashtag #archaeology from 2021 to 2023, we examine how content and context-related factors influence retweeting behaviour. Our findings reveal that tweets with positive sentiment and non-threatening language are more likely to be shared, contrasting with the common negativity bias observed on social media.
View Article and Find Full Text PDFSci Rep
January 2025
Hive AI Innovation Studio, Department of Computer Science and Engineering, University of Louisville, Louisville, KY, 40292, USA.
Nailfold Capillaroscopy (NFC) is a simple, non-invasive diagnostic tool used to detect microvascular changes in nailfold. Chronic pathological changes associated with a wide range of systemic diseases, such as diabetes, cardiovascular disorders, and rheumatological conditions like systemic sclerosis, can manifest as observable microvascular changes in the terminal capillaries of nailfolds. The current gold standard relies on experts performing manual evaluations, which is an exhaustive time-intensive, and subjective process.
View Article and Find Full Text PDFTransl Vis Sci Technol
January 2025
New England Eye Center, Tufts Medical Center, Boston, MA, USA.
Purpose: To evaluate visibility of a sub-band posterior to the external limiting membrane (ELM) and assess its age-associated variation.
Methods: In a retrospective cross-sectional study, normal eyes were imaged using a high-resolution spectral-domain optical coherence tomography (SD-OCT) prototype (2.7-µm axial resolution).
J Neurol
January 2025
LUMC Department of Neurology, Albinusdreef 2, 2333 ZA, Leiden, The Netherlands.
Background And Objectives: The total functioning capacity (TFC) assessment has been integral to Huntington's disease (HD) research and clinical trials, measuring disease stage and progression. This study investigates the natural progression of function in HD, focusing on changes in TFC scores related to age and CAG-repeat length, and evaluates TFC's strengths and weaknesses in longitudinal studies.
Methods: Using Enroll-HD platform's clinical dataset version 5, including Registry-3, we analysed data from 21,079 participants, with 16,083 having an expanded CAG repeat.
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!