Pyridoxamine as a potential scavenger of neurotoxic dopamine quinone: Inhibition of dopamine-induced α-synuclein oligomerization

Research Square · 2025-05-12

Development of Customized Learning Support System with LLM

Intelligent Information Convergence and Future Education · 2025-05-23 · 1 citations

기존 교사 주도형 교육 방식은 개별 학습자의 수준과 필요를 충분히 반영하지 못하며, 맞춤형 학습 지원이 부족하다. 인공지능 기술을 활용 하면 학습자의 패턴을 분석하고 최적화된 교육 콘텐츠를 제공할 수 있어, 자기주도적 학습을 촉진할 수 있다. 본 논문은 인공지능 기반의 LLM을 이용한 맞춤형 학습 지원 서비스를 개발한다. 본 시스템은 ChatGPT를 이용한 자동 문제 생성 및 채점, BERT 모델을 통한 유사 문 제 추천, GPT와 BERT 모델을 혼합한 유사 문제 생성으로 구성된다. 이를 활용하여, 기존의 교사 주도형 교육 방식의 한계를 극복하고, 더 나아가 학생 개인별 학업성취도를 고려한 맞춤형 학습 환경을 조성하는 데 중점을 두었다. 검증 실험을 위해, 개발한 BERT 기반 문제 추천 시스템과 기존 ChatGPT 기반 문제 추천 모델을 비교하였고, 학교 수업에 적용하여 효과성을 분석하였다. 실험 결과, BERT 모델의 유사도 점수는 0.99, ChatGPT 모델의 유사도 점수는 0.90으로 나타났다. 또한, BERT 기반 문제 추천을 받은 학습자의 평균 점수는 88점, ChatGPT 기반 문제 추천을 받은 학습자의 평균 점수는 77점으로, BERT 모델이 학습 성취도 향상에 더 효과적인 것으로 분석되었다. 만족 도 조사 결과, 학습자의 80% 이상이 추천된 문제가 학습 목표에 부합하며, 자기주도적 학습에 도움이 되었다고 응답하였다. 본 연구를 통 해, 개발된 맞춤형 학습 지원 시스템이 학생들의 개별 학습 수준을 고려한 효과적인 교육 환경을 제공할 수 있을 것으로 기대된다.

Real-world outcomes of venoarterial extracorporeal membrane oxygenation in cardiogenic shock patients related to acute myocardial infarction: a single-center experience

European Heart Journal Acute Cardiovascular Care · 2025-04-01

Abstract Background The benefit of venoarterial extracorporeal membrane oxygenation (VA-ECMO) for cardiogenic shock (CS) patients is still on controversy and the previous results regarding clinical outcome according to etiology are conflicting. Objectives This study aimed to report real-world clinical outcome of VA-ECMO treatment for CS patients according to the presence of acute myocardial infarction (AMI). Methods Patients treated with peripheral VA-ECMO from 2008 to 2023 at a single tertiary cardiovascular center were included in this study. Included patients were classified into two groups based on the etiology of CS (AMI-CS and non-AMI-CS groups). In-hospital mortality was compared using logistic regression models and prognostic predictors were identified. Results Of the total 667 patients, 264 (39.6%) were classified as AMI-CS. The rate of cardiac arrest before VA-ECMO initiation was higher in the AMI-CS group than in the non-AMI-CS group (69.7% vs. 55.8%; P<0.001). The patients in the AMI-CS group were older (66 years vs. 61 years; P<0.001), more likely to be male (82.6% vs. 57.3%; P<0.001), and had a lower left ventricular (LV) ejection fraction (20% vs. 25%; P<0.001), compared with those in the non-AMI-CS group. Among 24-hour survivors, the AMI-CS group had a lower rate of in-hospital mortality (49.2% vs. 60.5%; odds ratio, 0.50; 95% confidence interval, 0.29–0.84; P=0.01) as compared to the non-AMI-CS group. Independent predictors for favorable clinical outcomes after VA-ECMO included younger age, shorter cardiac arrest duration, absence of severe LV dysfunction, higher hemoglobin levels, absence of renal replacement therapy, and lower lactate levels. Conclusions In this single-center experience, AMI-CS was associated with a lower in-hospital mortality compared with non-AMI-CS after VA-ECMO treatment.

Characterization of ene reaction products of Cookson-type fluorescent reagents and their application for recognizing amyloid β aggregation

Analytical Sciences · 2025-05-23

Accelerating Battery Material Optimization through iterative Machine Learning

ArXiv.org · 2025-05-12

The performance of battery materials is determined by their composition and the processing conditions employed during commercial-scale fabrication, where raw materials undergo complex processing steps with various additives to yield final products. As the complexity of these parameters expands with the development of industry, conventional one-factor-at-a-time (OFAT) experiment becomes old fashioned. While domain expertise aids in parameter optimization, this traditional approach becomes increasingly vulnerable to cognitive limitations and anthropogenic biases as the complexity of factors grows. Herein, we introduce an iterative machine learning (ML) framework that integrates active learning to guide targeted experimentation and facilitate incremental model refinement. This method systematically leverages comprehensive experimental observations, including both successful and unsuccessful results, effectively mitigating human-induced biases and alleviating data scarcity. Consequently, it significantly accelerates exploration within the high-dimensional design space. Our results demonstrate that active-learning-driven experimentation markedly reduces the total number of experimental cycles necessary, underscoring the transformative potential of ML-based strategies in expediting battery material optimization.

Transcreation Strategic Considerations for Realizing Contextual Equivalence in Video Translation: Focusing on Korean and Japanese Video Media

The Japanese Language Association Of Korea · 2025-03-28

Based on the relevance theory of translation, this study examines aspects and strategies of transcreation through subtitles inserted into Korean and Japanese audiovisual media. Through the Korean and Japanese audiovisual media used as examples of transcreation, the fact that substitution strategies were most frequently used in the process of transcreation to reduce the linguistic and cultural gap between the source and target languages was confirmed. Additionally, generalization (paraphrase), specification (addition), omission, and direct translation (shifted) strategies were partially used. The substitution strategy was used in the highest percentage because it overcomes the linguistic and cultural differences between the departure and arrival languages and connects the speaker and the audience by achieving equivalence between the two languages. It means that strategies to completely rewrite specific elements in the departure language into expressions that are more appropriate in the target language's cultural sphere are essential.<br/> This paper confirmed that transcreation plays a role in effectively communicating wordplay (puns), bridging cultural gaps, maintaining emotional and affective coherence, and increasing audience immersion. As a result, transcreation transcends the scope of simple language conversion. It is an essential tool for expanding the approachability of global content as a creative translation strategy that bridges the language and cultural gap.

Personalized Prediction of Survival Rate with Combination of Penalized Cox Models and Machine Learning in Patients with Colorectal Cancer

Research Square · 2024-03-12

Tyrosine modifications of insulin-degrading enzyme enable favorable control of substrate specificity for both Alzheimer’s disease and type-2 diabetes mellitus

Bioorganic Chemistry · 2024-10-25 · 4 citations

• 4-Phenyl-1,2,4-triazoline-3,5-dione (PTAD) was used for tyrosine modification. • The modification sites on insulin-degrading enzyme (IDE) were identified. • The substrate specificity of tyrosine-modified IDE was examined. • The modified IDE digested amyloid beta, but not insulin. • That is favorable for both Alzheimer’s disease and type-2 diabetes mellitus. Insulin-degrading enzyme (IDE) cleaves amyloid beta (Aβ), insulin, and other bioactive peptides. Because Aβ and insulin are closely related to Alzheimer’s disease (AD) and type-2 diabetes mellitus (T2DM), respectively, IDE is a candidate drug target for treating both AD and T2DM. However, the activity of IDE has opposing effects, including decreasing AD risk by degrading Aβ and increasing T2DM risk by degrading insulin. The opposed substrate specificity is associated with the exo- and active sites containing Tyr 314 and Tyr 831 residues, the plausible modification targets for controlling substrate specificity. In this study, we used a tyrosine-specific modification regent, Cookson reagent (4-phenyl-1,2,4-triazoline-3,5-dione, PTAD), for IDE and examined the degradation activities on Aβ 40 and insulin. Fifteen tyrosine residues, including Tyr 314 and Tyr 831 , were modified by PTAD. After incubation with PTAD-modified IDE for 3 days, insulin remained intact, whereas Aβ 40 was completely degraded. This favorable change of substrate specificity was also observed in the mixture of Aβ 40 and insulin, suggesting that tyrosine modification of IDE might be a therapeutic strategy for AD and T2DM.

A Study on the Practice of Audiovisual Translation Education Using Translation Strategy: Focusing on the Analysis of Japanese Animation Retranslation Tasks

The Japanese Language Association Of Korea · 2024-09-27

This study examined effective educational methods for cultivating the next generation of Japanese audiovisual translators who combine young sensibilities and competence within university education and discover practical examples in the educational field.<br/> For this research, audiovisual translation strategies and practical translation education were conducted in the &quotJapanese Translating for Visual Media&quot class for undergraduate students majoring in Japanese interpretation and translation.<br/> The class was broadly divided into theory and practice. Students learned the rules and strategies of audiovisual translation in the theory class and then used them as the basis for practical translation assignments. The translation assignments were based on the Japanese animations &quotYour Name&quot and &quotSpirited Away&quot and presented their translation strategy and the translated sentences. After the presentation, the students presented the translated example sentences and exchanged various opinions through peer review by their fellow learners. Through this series of processes, the class was designed to produce audiovisual translation results of higher quality than existing translations.<br/> As a result of the class, it was confirmed that the learners&quot translation and analytical skills were remarkable, and that their speed of improvement was very fast. The learners corrected mistranslations of existing translations appropriately and proposed innovative and appropriate alternatives for areas that did not fall within the scope of mistranslations but still needed improvement. The strategies used in the translation were logical and convincing. The level of discussion among fellow students was significantly high, and the process of adopting the best alternative after several rounds of discussion was extremely rational.<br/> Compared to the rapidly increasing Japanese audiovisual market in Korea, the number of Japanese audiovisual translators is limited. To resolve this imbalance between supply and demand, it is necessary to systematically foster next-generation translators with an innovative sense and a deep understanding of the Japanese language and culture. The results of this study can be an example of supporting the legitimacy of fostering talented translators with young sensibilities in the Japanese audiovisual market in Korea.

Stable Isotope Dilution LC/ESI-SRM/MS Analysis for Highly Polar Bioactive Dipeptides Found in Fermented Brown Rice Product using a Porous Graphitic Carbon Column

Food Analytical Methods · 2023-09-05 · 3 citations

Abstract In this study, a validated analytical method based on liquid chromatography/electrospray ionization-selected reaction monitoring/mass spectrometry (LC/ESI-SRM/MS) was developed for dipeptides in a functional food, fermented brown rice and rice bran with Aspergillus oryzae (FBRA). The aim of this study was to screen and quantify dipeptides in the water-soluble fraction of FBRA. There are few studies focusing on dipeptides in FBRA, even though FBRA is expected to contain bioactive peptides: FBRA or its aqueous extract has shown variety of biological activities. An anti-hypertensive dipeptide was found in similar rice bran products digested by thermolysin. Dipeptides are the smallest class of peptides, and many show a variety of biological activities. Some bioactive peptides are widely found in fermented foods. However, because dipeptides are generally too polar to be retained on versatile LC columns, such as octadecylsilyl columns, dipeptides have been often overlooked and not quantified because of the analytical difficulties. The LC was performed using a porous graphitic carbon column with the mobile phases of 0.1% formic acid/0.1% trifluoroacetic acid/1% tetrahydrofuran in water and acetonitrile. The stable isotope-labeled internal standards and SRM for dipeptides made it possible to develop a validated method. Five dipeptides were identified and quantified in FBRA in much higher concentrations than related materials (malted brown rice, rice bran, brown rice, and white rice) as follows: Ile-Arg, 82.1 µg/g; Ala-Phe, 27.8 µg/g; Ala-Tyr, 31.6 µg/g; Val-Phe, 46.3 µg/g; and Ile-Tyr, 49.9 µg/g. This method was simple and robust and would be applicable for other functional foods. Graphical Abstract