More Like this

1. How Reproducibility Will Accelerate Discovery Through Collaboration in Physio-Logging

   https://doi.org/10.3389/fphys.2022.917976  

   Czapanskiy, Max F.; Beltran, Roxanne S. (July 2022, Frontiers in Physiology)

   What new questions could ecophysiologists answer if physio-logging research was fully reproducible? We argue that technical debt (computational hurdles resulting from prioritizing short-term goals over long-term sustainability) stemming from insufficient cyberinfrastructure (field-wide tools, standards, and norms for analyzing and sharing data) trapped physio-logging in a scientific silo. This debt stifles comparative biological analyses and impedes interdisciplinary research. Although physio-loggers (e.g., heart rate monitors and accelerometers) opened new avenues of research, the explosion of complex datasets exceeded ecophysiology’s informatics capacity. Like many other scientific fields facing a deluge of complex data, ecophysiologists now struggle to share their data and tools. Adapting to this new era requires a change in mindset, from “data as a noun” (e.g., traits, counts) to “data as a sentence”, where measurements (nouns) are associate with transformations (verbs), parameters (adverbs), and metadata (adjectives). Computational reproducibility provides a framework for capturing the entire sentence. Though usually framed in terms of scientific integrity, reproducibility offers immediate benefits by promoting collaboration between individuals, groups, and entire fields. Rather than a tax on our productivity that benefits some nebulous greater good, reproducibility can accelerate the pace of discovery by removing obstacles and inviting a greater diversity of perspectives to advance science and society. In this article, we 1) describe the computational challenges facing physio-logging scientists and connect them to the concepts of technical debt and cyberinfrastructure , 2) demonstrate how other scientific fields overcame similar challenges by embracing computational reproducibility, and 3) present a framework to promote computational reproducibility in physio-logging, and bio-logging more generally. 

   more » « less
2. Drivers of the dive response in trained harbour porpoises (Phocoena phocoena )

   https://doi.org/10.1242/jeb.208637  

   Elmegaard, S. L.; McDonald, B. I.; Madsen, P. T. (October 2019, Journal of Experimental Biology)

   null (Ed.)

   Pronounced dive responses through peripheral vasoconstriction and bradycardia enables prolonged apnoea in marine mammals. For most vertebrates, the dive response is initiated upon face immersion, but little is known about the physical drivers of diving and surfacing heart rate in cetaceans whose faces are always mostly submerged. Using two trained harbour porpoises instrumented with an ECG-measuring DTAG-3, we investigate the initiation and progression of bradycardia and tachycardia during apnoea and eupnoea for varying levels of immersion. We show that paranasal wetting drives bradycardia initiation and progression, whereas apnoea leads to dive-level bradycardia eventually, but not instantly. At the end of dives, heart rate accelerates independently of lung expansion, perhaps in anticipation of surfacing; however, full tachycardia is only engaged upon inhalation. We conclude that breathing drives surface tachycardia, whereas blowhole wetting is an important driver of bradycardia; although, anticipatory/volitional modulation can overrule such responses to sensory inputs. 

   more » « less
3. Stroke effort and relative lung volume influence heart rate in diving sea lions

   https://doi.org/10.1242/jeb.214163  

   McDonald, Birgitte I.; Tift, Michael S.; Hückstädt, Luis A.; Jeffko, Michael; Ponganis, Paul J. (March 2020, Journal of Experimental Biology)

   null (Ed.)

   The dive response, bradycardia (decreased heart rate) and peripheral vasoconstriction, is the key mechanism allowing breath-hold divers to perform long-duration dives while actively swimming and hunting prey. This response is variable and modulated by factors such as dive duration, depth, exercise and cognitive control. This study assesses the potential role of exercise and relative lung volume in the regulation of heart rate (fH) during dives of adult female California sea lions instrumented with ECG, depth, and 3-axis acceleration data loggers. A positive relationship between activity (minimum specific acceleration) and fH throughout dives suggested increased muscle perfusion associated with exercise. However, apart from late ascent, fH during dives was still less than or equal to resting heart rate (on land). In addition, the activity-fH relationship was weaker in long, deep dives consistent with prioritization of blood oxygen conservation over blood oxygen delivery to muscle in those dives. Pulmonary stretch receptor reflexes may also contribute to fH regulation as fH profiles generally paralleled changes in relative lung volume, especially in shallower dives and during early descent and late ascent of deeper dives. Overall, these findings support the concept that both exercise and pulmonary stretch receptor reflexes may influence the dive response in sea lions. 

   more » « less
4. Effect of Sampling Rate on ECG Deep Learning Classifiers

   https://doi.org/10.1109/ICHI64645.2025.00043  

   Dakshit, Sristi; Dakshit, Sagnik; Prabhakaran, Balakrishnan (June 2025, IEEE)

   With recent advances in Deep Learning (DL) models, the healthcare domain has seen an increased adoption of neural networks for clinical diagnosis, monitoring, and prediction. Deep Learning models have been developed for various tasks using 1D (one-dimensional) time-series signals. Time-series healthcare data, typically collected through sensors, have specific structures and characteristics such as frequency and amplitude. The nature of these features, including varying sampling rates that depend on the instruments used for sensing, poses challenges in handling them. Electrocardiograms (ECG), a class of 1D time-series signals representing the electrical activity of the heart, have been used to develop heart condition classification decision support systems. The sampling rate of these signals, influenced by different ECG instruments as well as their calibrations, can greatly impact the learning functions of deep learning models and subsequently, their decision outcomes. This hinders the development and deployment of generalized, DL-based ECG classifiers that can work with data from a variety of ECG instruments, particularly when the sampling rate of the training data remains unknown to users. Moreover, DL models are not designed to recognize the sampling rate of the testing data on which they are being deployed, further complicating their effective application across diverse clinical settings. In this study, we investigated the effect of different sampling rates of time-series ECG signals on DL-based ECG classifiers. To the best of our knowledge, this is the first work to understand how varying sampling rates affect the performance of DL-based models for classifying 1D time-series ECG signals. Through our comprehensive experiments, we showed that accuracy can drop by as much as 20% when the training and testing sampling rates are different. We provide visual explanations to understand the differences in learned model features through activation maps when the sampling rates for training and testing data are different. We also investigated potential strategies to address the challenges posed by different sampling rates: (i) transfer learning, (ii) resampling, and (iii) training a DL model using ECG data at different sampling rates. 

   more » « less
5. Dive heart rate in harbour porpoises is influenced by exercise and expectations

   https://doi.org/10.1242/jeb.168740  

   McDonald, Birgitte I.; Johnson, Mark; Madsen, Peter T. (January 2018, Journal of Experimental Biology)

   null (Ed.)

   The dive response, a decrease in heart rate (ƒH) and peripheral vasoconstriction, is the key mechanism allowing breath-hold divers to perform long duration dives. This pronounced cardiovascular response to diving has been investigated intensely in pinnipeds, but comparatively little is known for cetaceans, in particular in ecologically relevant settings. Here we studied the dive ƒH response in one the smallest cetaceans, the harbour porpoise (Phocoena phocoena). We used a novel multi-sensor data logger to record dive behaviour, ƒH, ventilations and feeding events in three trained porpoises, providing the first evaluation of cetacean ƒH regulation while performing a variety of natural behaviours, including prey capture. We predicted that tagged harbour porpoises would exhibit a decrease in ƒH in all dives, but the degree of bradycardia would be influenced by dive duration and activity, i.e., the dive ƒH response will be exercise modulated. In all dives, ƒH decreased compared to surface rates by at least 50% (mean maximum surface=173 beats min−1, mean minimum dive=50 beats min−1); however, dive ƒH was approximately 10 beats min−1 higher in active dives due to a slower decrease in ƒH and more variable ƒH during pursuit of prey. We show that porpoises exhibit the typical breath-hold diver bradycardia during aerobic dives and that the heart rate response is modulated by exercise and dive duration; however, other variables such as expectations and individual differences are equally important in determining diving heart rate. 

   more » « less