Leveraging VQ-VAE tokenization for autoregressive modeling of medical time series.

In this work, we present CodeAR, a medical time series generative model for electronic health record (EHR) synthesis. CodeAR employs autoregressive modeling on discrete tokens obtained using a vector quantized-variational autoencoder (VQ-VAE), which addresses key challenges of accurate distribution modeling and patient privacy preservation in the medical domain. The proposed model is trained with next-token prediction instead of a regression problem for more accurate distribution modeling, where the autoregressive property of CodeAR is useful to capture the inherent causality in time series data. In addition, the compressive property of the VQ-VAE prevents CodeAR from memorizing the original training data, which ensures patient privacy. Experimental results demonstrate that CodeAR outperforms the baseline autoregressive-based and GAN-based models in terms of maximum mean discrepancy (MMD) and Train on Synthetic, Test on Real tests. Our results highlight the effectiveness of autoregressive modeling on discrete tokens, the utility of CodeAR in causal modeling, and its robustness against data memorization.

Effects of Cool-Down Exercise and Cold-Water Immersion Therapy on Basic Fitness and Sport-Specific Skills among Korean College Soccer Players.

BACKGROUND: We aimed to examine the effects of cool-down exercise and cold-water immersion therapy on agility, speed, power, balance, and sport-specific skill performance in college soccer players, and to provide baseline data for the development of effective recovery programs. METHODS: In August 2020, 21 male college soccer players in Seoul, Korea, were randomly divided into the following groups: control group (CG, n=7), cool-down exercise group (CDG, n=7), and cool-down exercise plus cold-water immersion group (CDCWG, n=7). Agility, speed, power, balance, and sport-specific skill performance were assessed before and after the intervention. RESULTS: No significant differences in Southeast Missouri (SEMO) Agility Test, 20-m sprint test, vertical jump test, or Y-balance test (right) were observed among the groups; however, there was a significant effect of time (P<0.05) and a significant time × group interaction (P<0.05). Significant effects of time (P<0.001), group (P=0.043), and a time × group interaction (P=0.009) were observed in the Y-balance test (left). There were no between-group differences in the 22-m dribble test, shooting test (left), or shooting test (right); however, there were significant effects of time (P<0.05) and significant time × group interactions (P<0.05). No significant effects of group or time × group interactions were observed for the kicking test (left or right); however, there were significant effects of time (P<0.001). CONCLUSION: Addition of cold-water immersion therapy to a recovery program including cool-down exercise can promote recovery of basic and sport-specific abilities among college soccer players.

Neuroprotective effects of N-acetylcysteine via inhibition of matrix metalloproteinase in a mouse model of transient global cerebral ischemia.

N-acetylcysteine (NAC) is known to serve many biological functions including acting as an antioxidant, and electing antiinflammatory effects. Previous reports have revealed that NAC may have neuroprotective effects against the deleterious effects of brain ischemia. Despite of this, the mechanism by which NAC prevents neuronal damage after brain ischemia remains unclear. The current study aimed to investigate this mechanism in a mouse model of transient global brain ischemia. In the present study, mice were subjected to 20 min of transient global brain ischemia, proceeded by intraperitoneal administration of NAC (150 mg/kg) in one group. The mice were then euthanized 72 h after this ischemic insult for collection of experimental tissues. The effect of NAC on neuronal damage and matrix metalloproteinase (MMP)-9 activity were assessed and immunofluorescence, and hippocampal terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assay experiments were conducted and results compared between NAC- and vehicle-treated groups. Neuronal damage was primarily observed in the hippocampal CA1 and CA2 regions. In NAC-treated mice, neuronal damage was significantly reduced after ischemia when compared to vehicle-treated animals. NAC also inhibited increased MMP-9 activity after global brain ischemia. NAC increased laminin and NeuN expression and inhibited increases in TUNEL-positive cells, all in the hippocampus. These results suggest that NAC reduces hippocampal neuronal damage following transient global ischemia, potentially via reductions in MMP-9 activity.