Search

Introduction With similar prevalence to injuries from fires, stings, and natural disasters, soft tissue injuries may occur from fireworks, industrial accidents, or other explosives. Surgeons are less familiar with treating high-velocity penetration from small debris, which may increase the chance of infection and subsequent fatality. Penetration risk curves have been developed to predict V50, the velocity with 50% probability of penetration, for various sized projectiles. However, there has been limited research using nonmetallic materials to achieve lower density projectiles less than 1 g cm−2, such as sand or rocks. Material and Methods To emulate the size and density of these energized particles, 14 ball bearings of stainless steel, silicon nitride, or Delrin acetal plastic ranging from 1.59 mm (1/16") to 9.53 mm (3/8") with sectional densities between 0.3 g cm−2 and 5  g cm−2 were launched toward porcine legs at a range of velocities to determine the penetration thresholds. High-speed videography was captured laterally at 40 kHz and impact velocity was captured using a physics-based tracking software. A generalized linear model with repeated measures and a logit link function was used to predict probability of penetration for each projectile. A total of 600 impacts were conducted to achieve at least 15 penetrating impacts for each projectile over a range of velocities. Results Higher impact velocities were required to penetrate the skin as sectional density of the projectile decreased, and the relationship between velocity and sectional density exhibited an exponential relationship (V 50, |$ = 184.6*S{D^{ - 0.385}}$|⁠ , R 2 = 0.95) with substantial change for nonlinearity in sectional densities ranging from 0.3 g cm−2 to 1  g cm−2. Compared to previous studies, the empirical relationship was consistent in the linear region (2-5  g cm−2), and novel experimentation filled in the gaps for sectional densities less than 1  g cm−2, which expressed more nonlinearity than previously estimated. For low-density projectiles with diameters of 1.59 (1/16") or 3.18 (1/8"), 32 impacts were lodged into the epidermis but did not penetrate through the dermis; however, penetration was defined as displacement into or through the dermis. Conclusions These experimental results may be used to develop and validate finite element simulations of low-density projectile impacts to address complex, multivariate loading conditions for the development of protective clothing to reduce wounding and subsequent infection rates. [ABSTRACT FROM AUTHOR]

We propose masked particle modeling (MPM) as a self-supervised method for learning generic, transferable, and reusable representations on unordered sets of inputs for use in high energy physics (HEP) scientific data. This work provides a novel scheme to perform masked modeling based pre-training to learn permutation invariant functions on sets. More generally, this work provides a step towards building large foundation models for HEP that can be generically pre-trained with self-supervised learning and later fine-tuned for a variety of down-stream tasks. In MPM, particles in a set are masked and the training objective is to recover their identity, as defined by a discretized token representation of a pre-trained vector quantized variational autoencoder. We study the efficacy of the method in samples of high energy jets at collider physics experiments, including studies on the impact of discretization, permutation invariance, and ordering. We also study the fine-tuning capability of the model, showing that it can be adapted to tasks such as supervised and weakly supervised jet classification, and that the model can transfer efficiently with small fine-tuning data sets to new classes and new data domains.