Search

Maintaining a stable core body temperature is essential for endotherms. Cetaceans live in a highly thermally conductive medium, requiring special adaptations to reduce heat loss and maintain homeothermy. We employed a combination of aerial photogrammetry and existing data sources to estimate heat loss rates in five sympatric cetaceans of varying sizes, inhabiting the sub-arctic waters (∼3.7 °C) of NE-Iceland: harbour porpoises (Phocoena phocoena, 1.0 - 1.6 m, n=50), white-beaked dolphins (Lagenorhynchus albirostris, 1.1 - 2.9 m, n=294), minke whales (Balaenoptera acutorostrata, 4.4 - 8.6 m, n=30), humpback whales (Megaptera novaeangliae, 6.0 - 14.2 m, n=282) and blue whales (Balaenoptera musculus, 13.2 - 24.2 m, n=29). Further, we investigated the effect of body size (length), body shape (surface-area-to-volume ratio, SVR), body temperature, and blubber thermal conductivity and thickness on heat loss for all species. Smaller species had higher volume-specific heat loss compared to larger species due to their higher SVRs, a fundamental consequence of scaling. Apart from body size, blubber thickness had the largest effect on heat loss, followed by thermal conductivity. Smaller cetaceans seem to rely primarily on physiological and morphological adaptations to reduce heat loss, such as increased blubber thickness and lower thermal conductivity, whereas larger species offset heat loss by having larger bodies and lower SVRs. Our findings provide valuable insights into the thermal biology of these species and its implications for habitat use and prey requirements., (© 2025. Published by The Company of Biologists.)

Ovarian lesions are common and often incidentally detected. A critical shortage of expert ultrasound examiners has raised concerns of unnecessary interventions and delayed cancer diagnoses. Deep learning has shown promising results in the detection of ovarian cancer in ultrasound images; however, external validation is lacking. In this international multicenter retrospective study, we developed and validated transformer-based neural network models using a comprehensive dataset of 17,119 ultrasound images from 3,652 patients across 20 centers in eight countries. Using a leave-one-center-out cross-validation scheme, for each center in turn, we trained a model using data from the remaining centers. The models demonstrated robust performance across centers, ultrasound systems, histological diagnoses and patient age groups, significantly outperforming both expert and non-expert examiners on all evaluated metrics, namely F1 score, sensitivity, specificity, accuracy, Cohen's kappa, Matthew's correlation coefficient, diagnostic odds ratio and Youden's J statistic. Furthermore, in a retrospective triage simulation, artificial intelligence (AI)-driven diagnostic support reduced referrals to experts by 63% while significantly surpassing the diagnostic performance of the current practice. These results show that transformer-based models exhibit strong generalization and above human expert-level diagnostic accuracy, with the potential to alleviate the shortage of expert ultrasound examiners and improve patient outcomes., Competing Interests: Competing interests: E.E., K.S., F.C., E.K. and P.H. have applied for a patent (European patent application 23220765.4) that is pending to a company named Intelligyn. The patent covers methods for a computer-aided diagnostic system to improve generalization and protect against bias. E.E., K.S. and F.C. hold stock in Intelligyn, where E.E. also has an unpaid leadership role. N.C.P.’s institution has received payments for activities not related to this article, including lectures, presentations, expert testimonies, and service on speakers’ bureaus, as well as for travel support. N.C.P. has been an advisory board member of Mindray and GE Healthcare and has held unpaid leadership roles in the POGS Organization of Government Institutions (and the Rizal Medical Service Delivery Network, which are Philippine governmental institutions with the aim to facilitate smooth referral of patients. The other authors declare no competing interests., (© 2025. The Author(s).)

Improving our understanding of energy allocation in reproduction is key for accurately parameterizing bioenergetic models to assess population responses to environmental perturbations and anthropogenic disturbance. We quantified the energetic cost of gestation in humpback whales (Megaptera novaeangliae) using historical whaling records, non-invasive unoccupied aerial system (UAS) photogrammetry and post mortem tissue samples. First, we estimated relative birth size using body length measurements of 678 mother-fetus pairs from historical whaling records and 987 mother-calf pairs measured in situ using UAS-photogrammetry. The total energetic cost of gestation includes fetal growth (FG), heat increment of gestation and placental tissue development. FG was modelled from conception to birth, with fetal volume and mass estimated using the volume-to-length relationship of perinatal calves and published humpback whale tissue composition estimates. Tissue-specific energy content was quantified using post mortem bone, muscle, viscera and blubber samples from a neonatal humpback whale. Placental tissue development was estimated using humpback whale placental tissue and published equations. Relative birth length was found to be 33.75% (95% CI: 32.10-34.61) of maternal length. FG rates and absolute birth size increased with maternal length, with exponential growth in fetal length, volume and mass resulting in minimal energetic costs over the first two quadmesters (0.01-1.08%) before increasing significantly in the final quadmester (98.92%). Gestational heat constituted the greatest energetic cost (90.42-94.95%), followed by fetal (4.58-7.76%) and placental (0.37-1.83%) tissue growth. Our findings highlight the energetic costs endured by capital breeding females preceding parturition, with the most substantial energetic costs of gestation coinciding with migration and fasting. KEY POINTS: We quantified the energetic cost of gestation using body length measurements of mother-fetus pairs from historical whaling records, length estimates of mother-calf pairs measured in situ using aerial photogrammetry and post mortem tissue samples. Fetal growth rates and birth size increased with maternal length, with fetal length, volume and mass increasing exponentially over gestation. Energetic costs over the first two quadmesters were negligible (0.01-1.08%) before increasing significantly in the final quadmester (98.92%). Though larger females incur nearly twice the energetic cost of smaller females, they are likely buffered by greater absolute energy reserves, suggesting smaller females may be less resilient to perturbations in energy balance. We demonstrate the significant energetic costs incurred by pregnant humpback whales, with most of the energetic expenditure occurring over the final 100 days of gestation. Late-pregnant females are, therefore, particularly vulnerable to disruptions in energy balance, given periods of greatest energetic stress coincide with fasting and migration., (© 2024 The Authors. The Journal of Physiology © 2024 The Physiological Society.)