Convolutional neural networks have been applied to a wide variety of computer vision tasks. Recent advances in semantic segmentation have enabled their application to medical image segmentation. While most CNNs use two-dimensional kernels, recent CNN-based publications on medical image segmentation featured three-dimensional kernels, allowing full access to the three-dimensional structure of medical images. Though closely related to semantic segmentation, medical image segmentation includes specific challenges that need to be addressed, such as the scarcity of labelled data, the high class imbalance found in the ground truth and the high memory demand of three-dimensional images. In this work, a CNN-based method with three-dimensional filters is demonstrated and applied to hand and brain MRI. Two modifications to an existing CNN architecture are discussed, along with methods on addressing the aforementioned challenges. While most of the existing literature on medical image segmentation focuses on soft tissue and the major organs, this work is validated on data both from the central nervous system as well as the bones of the hand.
Current neuroethological experiments require sophisticated technologies to precisely quantify the behavior of animals. In many studies, solutions for video recording and subsequent tracking of animal behavior form a major bottleneck. Three-dimensional (3D) tracking systems have been available for a few years but are usually very expensive and rarely include very high-speed cameras; access to these systems for research is limited. Additionally, establishing custom-built software is often time consumingespecially for researchers without high-performance programming and computer vision expertise. Here, we present an open-source software framework that allows researchers to utilize low-cost high-speed cameras in their research for a fraction of the cost of commercial systems. This software handles the recording of synchronized high-speed video from multiple cameras, the offline 3D reconstruction of that video, and a viewer for the triangulated data, all functions previously also available as separate applications. It supports researchers with a performance-optimized suite of functions that encompass the entirety of data collection and decreases processing time for high-speed 3D position tracking on a variety of animals, including snakes. Motion capture in snakes can be particularly demanding since a strike can be as short as 50 ms, literally twice as fast as the blink of an eye. This is too fast for faithful recording by most commercial tracking systems and therefore represents a challenging test to our software for quantification of animal behavior. Therefore, we conducted a case study investigating snake strike speed to showcase the use and integration of the software in an existing experimental setup.
Knowledge about body motion kinematics and underlying muscle contraction dynamics usually derives from electromyographic (EMG) recordings. However, acquisition of such signals in snakes is challenging because electrodes either attached to or implanted beneath the skin may unintentionally be removed by force or friction caused from undulatory motion, thus severely impeding chronic EMG recordings. Here, we present a reliable method for stable subdermal implantation of up to eight bipolar electrodes above the target muscles. The mechanical stability of the inserted electrodes and the overnight coverage of the snake body with a “sleeping bag” ensured the recording of reliable and robust chronic EMG activity. The utility of the technique was verified by daily acquisition of high signal-to-noise activity from all target sites over four consecutive days during stimulus-evoked postural reactions in Amazon tree boas and Western diamondback rattlesnakes. The successful demonstration of the chronic recording suggests that this technique can improve acute experiments by enabling the collection of larger data sets from single individuals.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.