Browse Articles

Effects of seamless care in the perioperative management of laparoscopic pancreatoduodenectomy on patients’ quality of life and postoperative complications

Attention-enhanced and integrated deep learning approach for fishing vessel classification based on multiple features

A robust deep learning approach for segmenting cortical and trabecular bone from 3D high resolution µCT scans of mouse bone

Abstract Recent advancements in deep learning have significantly enhanced the segmentation of high-resolution microcomputed tomography (µCT) bone scans. In this paper, we present the dual-branch attention-based hybrid network (DBAHNet), a deep learning architecture designed for automatically segmenting the cortical and trabecular compartments in 3D µCT scans of mouse tibiae. DBAHNet’s hierarchical structure combines transformers and convolutional neural networks to capture long-range dependencies and local features for improved contextual representation. We trained DBAHNet on a limited dataset of 3D µCT scans of mouse tibiae and evaluated its performance on a diverse dataset collected from seven different research studies. This evaluation covered variations in resolutions, ages, mouse strains, drug treatments, surgical procedures, and mechanical loading. DBAHNet demonstrated excellent performance, achieving high accuracy, particularly in challenging scenarios with significantly altered bone morphology. The model’s robustness and generalization capabilities were rigorously tested under diverse and unseen conditions, confirming its effectiveness in the automated segmentation of high-resolution µCT mouse tibia scans. Our findings highlight DBAHNet’s potential to provide reliable and accurate 3D µCT mouse tibia segmentation, thereby enhancing and accelerating preclinical bone studies in drug development. The model and code are available at https://github.com/bigfahma/DBAHNet .

Explaining trunk strength variation and improvement following resistance training in people with chronic low back pain: clinical and performance-based outcomes analysis

Abstract A multitude of variables contribute to the variation of trunk strength in individuals with chronic low back pain (CLBP). This study investigated a range of variables to determine which variables contribute most to variation in trunk isometric strength and gains in strength following resistance training in people with CLBP. Outcome measures were recorded from 20 participants with CLBP both at baseline and following resistance training. Regression analyses were applied with the average trunk maximum voluntary isometric torque as the dependent variable. Variance in baseline trunk flexion strength (R 2 = .66) was explained by demographic covariates and a measure of trunk muscle co-activation. The baseline trunk extension strength variance (R 2 = .65) was explained by demographic covariates and lumbar erector spinae (LES) activity during a maximum trunk extension contraction. Demographic variables, trunk muscle co-activation, baseline trunk flexion strength, level of physical function, and pain intensity over the past week influenced the change in trunk flexion strength after training (R 2 = .93). Demographic variables and LES muscle activity explained the variance in trunk extension strength at follow-up (R 2 = .64). This study supports the major influence of sex, physical function and baseline strength and muscle activity, on the variation in maximum trunk strength in participants with CLBP at baseline and gains in trunk muscle strength following progressive resistance training.

Immature forms of low density granulocytes are increased in acute myeloid leukemia and myelodysplastic syndromes

Acrolein exposure associated with kidney damage: a cross‑sectional study

Asymmetry in centrosome maturation revealed through AIR-1 dynamics in the early Caenorhabditis elegans embryo

Feature refinement and rethinking attention for remote sensing image captioning

Methane oxidation to ethanol by a molecular junction photocatalyst

Abstract Methane, the main component of natural and shale gas, is a significant carbon source for chemical synthesis. The direct partial oxidation of methane to liquid oxygenates under mild conditions1–3 is an attractive pathway, but the inertness of the molecule makes it challenging to achieve simultaneously high conversion and high selectivity towards a single target product. This difficulty is amplified when aiming for more valuable products that require C–C coupling4,5. Whereas selective partial methane oxidation processes1–3,6–9 have thus typically generated C1 oxygenates6,7, recent reports have documented photocatalytic methane conversion to the C2 oxygenate ethanol with low conversions but good-to-high selectivities4,5,8–12. Here we show that the intramolecular junction photocatalyst covalent triazine-based framework-1 with alternating benzene and triazine motifs13,14 drives methane coupling and oxidation to ethanol with a high selectivity and significantly improved conversion. The heterojunction architecture not only enables efficient and long-lived separation of charges after their generation, but also preferential adsorption of H2O and O2 to the triazine and benzene units, respectively. This dual-site feature separates C–C coupling to form ethane intermediates from the sites where •OH radicals are formed, thereby avoiding over-oxidation. When loaded with Pt to further boost performance, the molecular heterojunction photocatalyst generates ethanol in a packed-bed flow reactor with greatly improved conversion that results in an apparent quantum efficiency of 9.4%. We anticipate that further developing the ‘intramolecular junction’ approach will deliver efficient and selective catalysts for C–C coupling, pertaining, but not limited, to methane conversion to C2+ chemicals.

Enhancing cardiovascular monitoring: a non-linear model for characterizing RR interval fluctuations in exercise and recovery

Engineering a genomically recoded organism with one stop codon

Three-dimensional markerless surface topography approach with convolutional neural networks for adolescent idiopathic scoliosis screening

A semi-supervised domain adaptive medical image segmentation method based on dual-level multi-scale alignment

A neoantigen vaccine generates antitumour immunity in renal cell carcinoma

Increased bronchopulmonary dysplasia along with decreased mortality in extremely preterm infants

Multivitamin supplementation and its impact in metabolic dysfunction-associated steatotic liver disease

Diagnostic value of C-reactive protein/ albumin ratio and TMTC1 in intracranial atherosclerotic stenosis in patients with acute cerebral infarction

Impact of distal or pylorus preserving gastrectomy on postoperative quality of life in T1 stage middle third gastric cancer patients

Pulse width modulation for current source inverters with arbitrary number of phases

Specific visual expertise reduces susceptibility to visual illusions

Abstract Extensive exposure to specific kinds of imagery tunes visual perception, enhancing recognition and interpretation abilities relevant to those stimuli (e.g. radiologists can rapidly extract important information from medical scans). For the first time, we tested whether specific visual expertise induced by professional training also affords domain-general perceptual advantages. Experts in medical image interpretation (n = 44; reporting radiographers, trainee radiologists, and certified radiologists) and a control group consisting of psychology and medical students (n = 107) responded to the Ebbinghaus, Ponzo, Müller-Lyer, and Shepard Tabletops visual illusions in forced-choice tasks. Our results show that medical image experts were significantly less susceptible to all illusions except for the Shepard Tabletops, demonstrating superior perceptual accuracy. These findings could possibly be attributed to a stronger local processing bias, a by-product of learning to focus on specific areas of interest by disregarding irrelevant context in their domain of expertise.