Slips and falls are leading causes of injury-related incidents, and one of the most effective ways to prevent slips is to use slip-resistant footwear. To evaluate footwear slip resistance, two main experiments are conducted: (1) human-centered tests, and (2) mechanical tests. Human-centered slip resistance tests better capture real-world locomotion dynamics than mechanical tests, but they are costly and time-consuming. This study aimed to develop a predictive model that estimates human-centered slipperiness score called Maximum Achievable Angle (MAA) outcomes using mechanical slip resistance measurements, enabling accessible footwear evaluation. Thirty-seven winter footwear samples were tested using both human-centered MAA testing and the SATRA STM 603 mechanical slip resistance tester on wet and dry ice surfaces. Ten linear and non-linear regression models were trained using a nested 5-fold cross-validation approach with unseen footwear in each fold to support generalizability. Model performance was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), coefficient of determination (R2), and Bland-Altman analysis. Across all models, the Ridge model produced the most conservative and consistent results, achieving an RMSE of 2.73° and MAE of 2.27° on wet ice, and an RMSE of 1.58° and MAE of 1.33° on dry ice. Although the results are close to the ± 1° acceptable range of MAA, slight deviations in some cases highlight opportunities for future model improvement. Mechanical slip resistance metrics can be used to reasonably predict human-centered MAA outcomes. While prediction accuracy is promising, further refinement is needed to fully meet the precision requirements for footwear slip resistance certification.

Synchronous measures of head kinematics and in vivo brain deformation during rapid head rotation are needed to advance understanding of traumatic brain injury (TBI) mechanics and enhance computational modeling as a tool for injury risk assessment and prevention. The aims of this study were to produce repeatable rapid rotation of the in vivo sheep head, to assess the viability of the sonomicrometry method for measuring multipoint brain displacement, and to quantify the in vivo brain deformation response to rapid rotation. In three anaesthetized adult sheep, arrays of sonomicrometry transceivers were implanted into the brain and rigidly attached to the inner skull surface. Repeatable, rapid, sagittal plane head rotation (nominally about the second cervical vertebra) was induced with a non-impact head rotation apparatus. Computed tomography imaging was performed to assess relative motion between transceivers and brain tissue. Three-dimensional brain displacement, strain, and head kinematics were assessed for repeatability. The location of up to 13 (mean = 11) brain transceivers were tracked in each of 11 rapid head rotation tests. Peak head angular acceleration and velocity were up to 38.56 krad/s2 and 30.43rad/s, respectively, and average duration of head motion was 241.9 ± 23.1ms. Pre-to-post-test transceiver displacement in brain tissue was less than the spatial resolution of the measurement system, and brain displacements measured during rapid head rotation had excellent repeatability (CORA score 0.99). Brain displacements and strains up to 2.47mm and 18%, respectively, were observed. The brain exhibited decaying sinusoidal rotational deformation in the sagittal plane with oscillating tension-compression waves. Sonomicrometry was reliably applied in vivo and provided repeatable measurements of brain deformation in a non-survival large animal model of rapid head rotation.

Woodpeckers often withstand extreme high acceleration during the pecking process, but show no signs of eye injuries. This study seeks to explore what biomechanical adaptations woodpeckers adopt to protect their eyes under such high impact. The morphometric parameter of eyes in the Great Spotted Woodpecker and its control, Eurasian Hoopoe, were observed based on frozen sectioning and Nissl staining. A series of protein expression including β-APP, p-Tau, p-CRMP2, HSP70, and mTOR of the birds' eyes were also obtained using immunofluorescence staining. The Great Spotted Woodpecker exhibits a denser, ossified sclera and a more complex pecten. β-APP was primarily localized in the outer nuclear layer while p-CRMP2 was predominantly expressed in the GCL, and HSP70 were co-localized with p-Tau in these birds' eyes. Meanwhile, mTOR exhibited distinct expression patterns corresponding to different phases of injury. Two layers of adaptations were found in eyes of woodpeckers. Structurally, ossified sclera and complex pecten contribute to resistance against pecking-induced eye injuries. Molecularly, abnormal aggregation of β-APP and p-Tau in optic nerve cells were prevented by regulating expression of CRMP2, HSP70, and mTOR, thereby reducing damage in the eyes. These results provide a theoretical foundation for the identification, prevention, and mitigation of acceleration-induced eye injuries.

Quantifying collagen in histological slides is essential for diagnosing and monitoring fibrosis. However, the combination of PicroSirius Red staining with polarized light microscopy, one of the standard techniques, requires expensive equipment and time-consuming procedures. This study investigated whether supervised machine learning algorithms applied to bright-field images could provide an automated and accurate method for collagen segmentation, reducing the dependence on polarized imaging. A total of 140 histological images of renal, cardiac, and tendinous tissues from mice and rats stained with PicroSirius Red were analyzed. The images were fragmented into smaller patches, and the green channel was selected for non-polarized analysis. The reference standard was generated from polarized images binarized using ImageJ. Two supervised models were evaluated: (1) a Multilayer Perceptron (MLP) trained using five features (RGB intensities, median filter, and Wavelet), and (2) a U-Net architecture adapted to the image dimensions. Both models performed well, with the best results observed in rat tendon samples (Dice and precision > 80%). Performance was lower in mouse kidney images, possibly due to weaker staining and thinner collagen fibers. Overall, the MLP slightly outperformed the U-Net, which still showed comparable results. The proposed methodology proved effective for collagen segmentation highlighting the potential of supervised learning to reduce costs and processing time in histological analysis. Future improvements include expanding the dataset and refining the reference standard to enhance robustness in biological tissue evaluation.

Accurate delineation of the prostate and surrounding pelvic structures is critical to successful treatment planning, accurate identification, and staging of prostate cancer. Segmentation of anatomically complex regions surrounding the prostate in CT imaging can be challenging due to low soft-tissue contrast and complex boundary delineations. In this work, we investigate three complementary paradigms of knowledge distillation-voxel-level, region-level, and a dynamically weighted combination of the two-to improve segmentation performance for the prostate and parailiac regions. The region-level approach imposes the semantic coherence of network predictions via the region-wise contrastive form of supervision, whereas the voxel-level distillation provides fine-tuned supervision in terms of Kullback-Leibler divergence on soft probabilistic outputs. We introduce a novel fusion approach that adds uncertainty-aware dynamic weighting, thus allowing the model to adjust the contribution of every distillation loss in an adaptive manner during training, taking advantage of the strengths of both approaches. The distillation methods are implemented within the dual-network architecture in terms of VNet (Milletari et al. in Proc. Int. Conf. 3D Vis. (3DV):565-571, 2016) and 3D-ResVNet (Wang et al. in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR)), thus allowing synergistic learning of different architectural biases. Experimental results on both an in-house collected and annotated CT dataset of prostate cancer patients-where parailiac regions and the prostate gland (including seminal vesicles) are manually segmented-and three public benchmark datasets demonstrate that each individual distillation method consistently improves segmentation accuracy over baseline models. These results indicate that the effectiveness of the proposed distillation strategies generalizes across different datasets and anatomical structures, highlighting their robustness and practical applicability. Complementary supervision at voxel and region levels can improve the delineation of complex pelvic structures in CT imaging of prostate cancer.

Deepfakes have posed severe challenges to healthcare systems as fake medical images and videos can be utilized to disseminate fake information about an organization or person. The challenges have open room for more solutions to address them. Therefore, this study provides a survey that highlights the considerable strides made in the development of deepfake detection technologies while showcasing various approaches, from advanced machine learning techniques to multi-modal systems. The progress made in identifying deepfakes, particularly with regard to deep learning and hybrid models, shows promise for detecting alterations in digital content and medical imaging. But the use of these technologies shows differing degrees of efficacy, suggesting the necessity for customized detection tactics that take into account the particular difficulties of certain domains, such as nuclear medicine and endoscopic videography. In addition, the application of these technologies raises significant ethical and legal questions, such as those pertaining to data security, privacy, and possible abuses of artificial intelligence. Therefore, it becomes critical to provide a survey on these issues in order to build and apply deepfake detection tools responsibly.

Determine the effect of mechanical and intrinsic (demographic and anthropometric) variables on anteroposterior shear stiffness in cervical functional spinal units (FSUs), and provide response corridors for computer model validation. Thirty-nine human FSUs were loaded non-destructively in statically rotationally unconstrained pure shear up to 200N, in anterior and posterior directions, at rates of 1, 10, and 100mm/s. Due to the inherent non-linear behavior, the resulting stiffness response was analyzed as bilinear (lower Phase I and higher Phase II). The effect of mechanical (direction and rate) and intrinsic (age, sex, anteroposterior endplate dimension, disc and facet joint degeneration, bone volume fraction) factors on stiffness were evaluated using a linear mixed models approach. In Phase I, specimens were 67% stiffer in anterior shear than posterior (p < 0.001), and 25% less stiff at 100mm/s than at 1 and 10mm/s (p < 0.001) across both directions. In Phase II, specimens were 42% stiffer in anterior shear than posterior (p < 0.001) and displacement rate was positively correlated with stiffness (p < 0.001). Males were 8% stiffer than females on average (p = 0.001), although a significant interaction between sex and direction (p = 0.008) showed that this was more prominent in the anterior direction, where males were 12% stiffer. Disc degeneration was positively correlated with stiffness (p = 0.003). The unique data measured in this study provides improved understanding of lower cervical spine shear response and provide critical data for validating computational modelsof the spine.

Quantify the long-term mechanical durability of Abbott stylet-driven leads for left bundle branch area pacing (LBBAP) using a combination of computational modeling and fatigue bench test. Intracardiac lead curvatures were simulated in a 3D computational heart model across a range of lead implant scenarios and cardiac electro-mechanical scenarios over a complete cardiac cycle to determine the range of induced stresses. Safety factor was determined using the bending curvature and mechanical properties of the conductor. A benchtop fatigue physical test was implemented at higher stress levels, thus lower safety factors, than the computational modeling cohort. Computer simulation modeling showed that maximum curvature and safety factor (SF) were comparable for LBBA implants (0.06 ± 0.02mm-1 and 4.7 ± 1.6) vs. other traditional implant locations (0.05 ± 0.03mm-1 and 5.0 ± 0.7). Lead durability was verified through successful fatigue bench testing at max curvatures of 0.12 ± 0.01mm-1 and safety factor of 2.2 ± 0.2. These test conditions represented approximately a twofold increase in maximum curvature and a 2.1-fold reduction in safety factor relative to simulated cases, over a test duration representative of 10years of service. Abbott stylet-driven leads were demonstrated to be safe for LBBAP, even at stresses far exceeding those which may be clinically observed for 400 million cardiac cycles. No differences in long-term safety were observed between LBBAP and traditional pacing locations.

Large quantities of human pluripotent stem cells (hPSCs) are required for clinical applications. 3D suspension cultures are suitable for large-scale manufacturing of hPSCs but yield, viability, and quality are affected by the hydrodynamic environment. This paper characterizes the hydrodynamic environment inside vertical-wheel bioreactors (VWBRs) as a function of size and agitation rates, measures its effect on cell aggregation and proliferation, and proposes the use of Lagrangian-based shear stress and energy dissipation rate (EDR) exposures to support scale-up. In silico: Transient, 3D, turbulent flow simulations are conducted for two VWBR sizes (100, 500mL) at five agitation rates between 20 and 80rpm. Trajectories of cell aggregates of sizes from 200 to 1,000 microns are calculated, and shear stress and EDR exposures are collected along these trajectories. In vitro: ESI-017 hPSCs were cultured in VWBRs for 6days. Aggregation efficiency and daily fold ratios were calculated based on cell counts and initial inoculation density. Aggregate size, agitation rate, and bioreactor size modulate cell aggregate exposures to EDR and shear stress, which significantly depart from maximum or volume average metrics used for scale-up. Combined in vitro/in silico results show EDR affects aggregation efficiency, cell counts, and aggregate size, and has a small effect on daily fold ratios but a significant effect on total fold ratio. History of trajectory-based cell aggregate exposures to EDRs provides a better scale-up basis for VWBRs than volume-averaged EDR. Shear stress does not significantly affect hPSC aggregation, proliferation and expansion in VWBRs under the tested conditions.