Long Short-Term Memory-Based Deep Learning Models for Screening Parkinson’s Disease Using Sequential Diagnostic Codes

Seokjoon Yoon,Minki Kim,Woong-Woo Lee 대한신경과학회 2023 Journal of Clinical Neurology Vol.19 No.3

Background and Purpose It is challenging to detect Parkinson’s disease (PD) in its early stages, which has prompted researchers to develop techniques based on machine learning methods for detecting PD. However, previous studies did not fully incorporate the slow progression of PD over a long period of time nor consider that its symptoms occur in a time-sequential manner. Contributing to the literature on PD, which has relied heavily on cross-sectional data, this study aimed to develop a method for detecting PD early that can process time-series information using the long short-term memory (LSTM) algorithm. Methods We sampled 926 patients with PD and 9,260 subjects without PD using medicalclaims data. The LSTM algorithm was tested using diagnostic histories, which contained the diagnostic codes and their respective time information. We compared the prediction power of the 12-month diagnostic codes under two different settings over the 4 years prior to the first PD diagnosis. Results The model that was trained using the most-recent 12-month diagnostic codes had the best performance, with an accuracy of 94.25%, a sensitivity of 82.91%, and a specificity of 95.26%. The other three models (12-month codes from 2, 3, and 4 years prior) were found to have comparable performances, with accuracies of 92.27%, 91.86%, and 91.81%, respectively. The areas under the curve from our data settings ranged from 0.839 to 0.923. Conclusions We explored the possibility that PD specialists could benefit from our proposed machine learning method as an early detection method for PD.

Diethylamino-curcumin mimic with trizolyl benzene enhances TRAIL-mediated cell death on human glioblastoma cells

안용철,Seokjoon Lee,Cheon Soo Park,Hyuk-Jai Jang,이지환,Byong-Gon Park,Yoon Sun Park,Woon-Seob Shin,Daeho Kwon 대한독성 유전단백체 학회 2018 Molecular & cellular toxicology Vol.14 No.2

Backgrounds: Glioblastoma multiforme is one of the most aggressive human malignant brain tumors. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is known as the death ligand, which induces preferential apoptosis of transformed cancer cells. In this study, we demonstrated that the newly synthesized diethylamino-curcumin mimic with trizolyl benzene (YM-4) enhances cytotoxicity in combination with TRAIL in human glioblastoma cells. Methods: We synthesized diethylamino-curcumin mimic with trizolyl benzene (YM-4) and investigated possible apoptotic cell signaling by co-treatment with YM-4 and TRAIL on human glioblastoma cells. Results: Caspase-8, 9, and 3 and poly (ADP-ribose) polymerase were more efficiently cleaved with cotreatment of YM-4 and TRAIL than treatment with each alone in human glioblastoma cells. Co-treatment with YM-4 and TRAIL significantly increased the expression of Bax and Smac/Diablo and also inhibited the expression of the X-linked inhibitor of apoptosis protein and Survivin in human glioblastoma cells. Conclusion: These results demonstrated that YM-4 can be an anticancer candidate that can be effective on human glioblastoma cells in combination with TRAIL.

Sensitive Detection of Nucleic Acid Biomarkers Using Transcription Mediated Isothermal Amplification

Taehwi YOON,Seokjoon KIM,Jiye SHIN,Ki Soo PARK 한국생물공학회 2021 한국생물공학회 학술대회 Vol.2021 No.10

Nucleic acid biomarkers such as microRNAs (miRNA) involved in various biological processes can be used for early diagnosis and prognosis of various cancers. Quantitative reverse transcription PCR is the gold standard method to detect nucleic acid biomarkers in clinical settings, but it requires expensive equipment and complicated primer designs, thus hindering its use in the field. Herein, we developed a simple, cost-effective strategy for the multiplex detection of nucleic acid biomarkers using light-up RNA aptamers as the key detection component. In principle, the presence of target nucleic acids produces a large amount of light-up RNA aptamers such as Spinach and Mango aptamers through strand displacement amplification in combination with transcription amplification. Consequently, light-up RNA aptamers bind to their corresponding fluorogens and emit highly enhanced fluorescence signals at different wavelengths, which enables the multiplex detection of target nucleic acids. With the proposed strategy, target miRNAs such as miR21 and miR141 were simultaneously determined with limits of detection of 0.955 fM and 0.195 fM, respectively. This assay was successfully applied for the quantification of miRNAs in different cell lines, demonstrating its practical utility in clinical diagnosis and prognosis.

Evaluation of Large-scale Data to Detect Irregularity in Payment for Medical Services : An Extended Use of Benford’s Law

Park, Junghyun,Kim, Minki,Yoon, Seokjoon Schattauer GmbH (Methods of Information in Medicin 2016 Methods of information in medicine. Vol.55 No.3

<B>Summary</B><P> Background: Sophisticated anti-fraud systems for the healthcare sector have been built based on several statistical methods. Although existing methods have been developed to detect fraud in the healthcare sector, these algorithms consume considerable time and cost, and lack a theoretical basis to handle large-scale data.</P><P> Objectives: Based on mathematical theory, this study proposes a new approach to using Benford’s Law in that we closely examined the individual-level data to identify specific fees for in-depth analysis.</P><P> Methods: We extended the mathematical theory to demonstrate the manner in which large-scale data conform to Benford’s Law. Then, we empirically tested its applicability using actual large-scale healthcare data from Korea’s Health Insurance Review and Assessment (HIRA) National Patient Sample (NPS). For Benford’s Law, we considered the mean absolute deviation (MAD) formula to test the large-scale data.</P><P> Results: We conducted our study on 32 diseases, comprising 25 representative diseases and 7 DRG-regulated diseases. We performed an empirical test on 25 diseases, showing the applicability of Benford’s Law to large-scale data in the healthcare industry. For the seven DRG-regulated diseases, we examined the individual-level data to identify specific fees to carry out an in-depth analysis. Among the eight categories of medical costs, we considered the strength of certain irregularities based on the details of each DRG-regulated disease.</P><P> Conclusions: Using the degree of abnormality, we propose priority action to be taken by government health departments and private insurance institutions to bring unnecessary medical expenses under control. However, when we detect deviations from Benford’s Law, relatively high contamination ratios are required at conventional significance levels.</P>

Status of Health and Safety Management in Occupational Hygiene Laboratories in Korea

Yang, Hyukseung,Choi, Jaewook,Yoon, Seokjoon,FARR, Terry Korean Industrial Hygiene Association 2007 한국산업보건학회지 Vol.17 No.1

목적: 1990년대 초반부터 시작된 우리나라의 산업 보건에 대한 사회적 관심의 증대와 시장의 요구에 따라 산업위생기관의 양적인 팽창이 이루어지긴 하였으나 실험실 관련 안전보건규정이 미비한 상태에서 추진되어 산업위생 실험실에서 다양한 유해 위험성 인자를 취급함에도 불구하고 사용되는 량이 소규모로 인해 안전보건과 관련된 사항은 대부분 간과하고 있으며, 산업위생 실험실의 안전보건관리 실태는 선진외국에 비해 상당히 낙후되어 있는 실정이다. 따라서 본 연구는 우리나라 산업위생 실험실과 관련한 안전보건관리 실태를 평가함으로써 향후 실험실의 안전보건관리 수준을 향상 시키는데 효과적인 자료로 활용하는데 기여하고자 한다. 방법: 산업위생실험실의 안전보건 실태를 파악하기 위해 설문조사를 수행하였으며, 설문은 실험실의 전반적인 안전보건, 화학물질 저장 및 용기, 가연성 및 인화성 화학물질, 가스 실린더, 의사전달, 응급처치 및 비상사태 장비, 정리정돈, 소방안전, 전기안전, 개인보호구, 흄후드 및 환기, 및 이황화탄소 취급 사례로서 12개 영역으로 구성되었다. 연구대상은 2001년 한국산업안전공단 정도관리프로그램에 참여한 기관으로서 총 대상은 119개 기관이었다. 연구기간은 2002년 7월 01일부터 8월 30일까지 약 60일 이었다. 설문은 반송봉투에 넣어 설문완성 후 연구자에게 보낼 수 있도록 배려하였으며, 1차 설문을 보낸 후 2주 후에 설문 참여를 독려하기 위해 엽서를 발송하였고, 다시 2주 후에 각 산업위생기관에 개별 전화 연락을 취하였다. 그 결과 63% (75개 기관)의 완성된 설문을 얻을 수 있었다. 설문의 총 조사항목은 79문항으로 구성되어 있으며, 산업위생실험실의 안전보건 수준을 정량적으로 평가하기 위해 안전보건 매뉴얼, 교육훈련프로그램 및 한국산업안전공단 실험실 지침서를 보유한 기관과 그렇지 않은 기관간의 차이는 선별된 67문항에 대해 각 문항에 합당하거나 적절할 경우 1점을 부여하는 방식으로 점수화하였다. 이들의 관련성을 파악하기 위하여 Microsoft-Excel 2000 프로그램을 이용하여 two-tailed t-test 분석을 실시하였다. 결과 1. 산업위생실험실 운영과 관련한 67개 항목(항목별로 각 1점 부여)에 대한 안전보건 성과지수화 (100점으로 점수 환산) 결과 안전보건 매뉴얼을 보유한 기관은 $42.98{\pm}13.36$(p<0.001)점, 교육훈련프로그램을 보유한 기관은 $50.75{\pm}14.12$(p<0.01)점, 한국산업안전공단 실험실 지침서를 보유한 기관은 $43.58{\pm}11.92$(p<0.01)점으로 그렇지 않은 기관에 비해 통계적으로 유의하게 높은 점수분포를 보였다. 2. 화학물질 보관 캐비닛이 있는 기관 중 64.8%의 기관은 화학물질을 알파벳 순으로, 27.8%는 분류 기준에 따라, 그리고 7.4%는 무작위로 보관하고 있었다. 3. 단지 8.0% (6개)의 기관만이 실험실내에 눈세정 분수(2 개), 샤워기(3개) 및 눈세정물병(3개, 2개 중복 응답 기관)과 같은 응급처치 장치 및 물품을 갖추고 있었다. 4. 89.0%의 기관이 흄후드내에 화학물질을 보관하고 있었다. 5. 물질안전보건자료, 흄후드 기록지, 비상사태 절차서 및 한국산업안전공단 실험실 지침서와 같은 문서관리는 부적절하게 관리 및 기록되고 있었다. 6. 대부분의 산업위생 실험실은 응급처치 장비, 화학물질 또는 가스용기 보관실과 같은 실험실 안전설비가 부족할 뿐만 아니라 정리정돈, 화학물질 저장 캐비닛, 안전보호구 및 흄후드와 같은 관

A Search Algorithm for the Worst Operation Scenario of a Cross-Point Phase-Change Memory Utilizing Particle Swarm Optimization

Kim, Kwangmin,Kang, Seokjoon,Sim, Jae-Yoon,Park, Hong-June,Kim, Byungsub IEEE 2018 IEEE transactions on very large scale integration Vol.26 No.12

<P>In this paper, we propose a search algorithm to find the worst operation scenario of a cross-point array of a phase-change random access memory to enable a precise read margin evaluation. The search algorithm utilizes a particle swarm optimization method to find the worst scenario quickly and efficiently. In an experiment, the proposed algorithm improves the search speed by <TEX>$39.3\times $</TEX> compared with the previous algorithm. With the improved search speed, the proposed algorithm could find the worst operation scenarios of large arrays whose worst operation scenarios had been only guessed before. In the experiment with a large array, the proposed algorithm proved that the worst high-resistance state read current can be <TEX>$36\times $</TEX> larger than the previous best guess. In the reliability test, the evaluation error of the worst read current found by the proposed algorithm is less than 0.2% with 99% probability. These results show that the proposed search algorithm can improve the precision and efficiency of the read margin evaluation in designing a cross-point phase-change memory array.</P>

가구 내 미디어 기기 구매와 폐기 분석: 집단 의사결정의 개인화에 대하여

박진환(Jinhwan Park),윤석준(Seokjoon Yoon),윤성욱(Sungwook Yoon) 한국경영과학회 2023 한국경영과학회지 Vol.48 No.2

Tumor-suppressor Protein p53 Sensitizes Human Colorectal Carcinoma HCT116 Cells to 17α-estradiol-induced Apoptosis via Augmentation of Bak/Bax Activation

Cho Rong Han(한초롱),Ji Young Lee(이지영),Dongki Kim(김동기),Hyo Young Kim(김효영),Se Jin Kim(김세진),Seokjoon Jang(장석준),Yoon Hee Kim(김윤희),Do Youn Jun(전도연),Young Ho Kim(김영호) 한국생명과학회 2013 생명과학회지 Vol.23 No.10

17α-estradiol (17α-E₂)의 에폽토시스 유도활성에 미치는 종양억제단백질 p53의 조절효과를 조사하고자, 17α-E₂에 의해 유도되는 에폽토시스 현상들을 인체 대장암 세포주 유래 클론인 HCT116 (p53+/+) 및 HCT116 (p53-/-) 세포에서 비교하였다. HCT116 (p53+/+) 및 HCT116 (p53-/-) 세포를 17α-E₂ (2.5~10 μM)로 처리하거나 혹은 HCT116 (p53+/+) 및 HCT116 (p53-/-) 세포를 10 μM 17α-E₂로 시간 별로 처리한 결과, HCT116 (p53+/+)에 있어서는 세포독성과 에폽토시스-관련 sub-G₁ peak의 비율은 처리농도와 시간에 의존적으로 나타났다. 그러나 HCT116 (p53-/-) 세포의 경우는 이러한 현상이 미약하게 나타났다. 17α-E₂에 의해 유도되는 비정상적 유사분열방추사 형성, 중기판 염색체 배열의 미완성, 이에 따른 유사분열정지(G₂/M arrest) 등의 현상은 HCT116 (p53+/+) 및 HCT116 (p53-/-) 세포에서 유사한 수준으로 나타났다. 이에 반해, 17α-E₂에 의해 유도되는 Bak과 Bax의 활성화, 미토콘드리아의 막전위 상실(Δψm loss), 그리고 PARP 분해 등의 현상은 HCT116 (p53-/-) 세포에 비해 HCT116 (p53+/+) 세포에서 훨씬 높은 수준으로 확인되었다. 아울러 17α-E₂로 처리된 HCT116 (p53+/+) 세포에서 확인되는 p53 (Ser-15)의 인산화 및 p53 수준의 증가와 일치하여, 세포 내의 p21및 Bax 수준도 현저히 증가하였다. 이때 17α-E2로 처리된 HCT116 (p53-/-) 세포에서는 p21 및 Bax의 발현수준이 매우 낮았다. 한편, 에폽토시스 억제단백질인 Bcl-2 단백질수준은 HCT116 (p53-/-) 세포에 비해 HCT116 (p53+/+) 세포에서 다소 낮았으나, 이러한 Bcl-2 단백질 수준은 17α-E₂ 처리 후에도 크게 변화하지 않는 것으로 나타났다. 이러한 결과들은 17α-E₂ 처리에 의해 유도되는 에폽토시스 유도 경로의 구성원들의 변화, 즉 비정상적 유사분열방추사 형성 및 이에 따른 유사분열정지(G2/M arrest), 뒤이은 Bak 및 Bax의 활성화, 미토콘드리아의 막전위 상실, 그리고 이에 수반되는 caspase cascade 활성화 및 PARP 분해로 진행되는 에폽토시스 현상들 중에서, Bak 및 Bax의 활성화 단계가 종양억제단백질 p53의 에폽토시스 증진 활성에 의해 양성적으로 조절되는 작용 타켓임을 보여준다. The regulatory effect of the tumor-suppressor protein p53 on the apoptogenic activity of 17α-estradiol (17α-E₂) was compared between HCT116 (p53+/+) and HCT116 (p53-/-) cells. When the HCT116 (p53+/+) and HCT116 (p53-/-) cells were treated with 2.5~10 μM 17α-E₂ for 48 h or with 10 μM for various time periods, cytotoxicity and an apoptotic sub-G₁ peak were induced in the HCT116 (p53+/+) cells in a dose- and time-dependent manner. However, the HCT116 (p53-/-) cells were much less sensitive to the apoptotic effect of 17α-E₂. Although 17α-E₂ induced aberrant mitotic spindle organization and incomplete chromosome congregation at the equatorial plate, G₂/M arrest was induced to a similar extent in both cell types. In addition, 17α-E₂-induced activation of Bak and Bax, Δψm loss, and PARP degradation were more dominant in the HCT116 (p53+/+) than in the HCT116 (p53-/-) cells. In accordance with enhancement of p53 phosphorylation (Ser-15) and p53 levels, p21 and Bax levels were elevated in the HCT116 (p53+/+) cells treated with 17α-E₂. The HCT116 (p53-/-) cells exhibited barely or undetectable levels of p21 and Bax, regardless of 17α-E₂ treatment. On the other hand, although the level of Bcl-2 was slightly lower in the HCT116 (p53+/+) than in the HCT116 (p53-/-) cells, it remained relatively constant after the 17α-E₂ treatment. Together, these results show that among the components of the 17α-E₂-induced apoptotic-signaling pathway, which proceeds through mitotic spindle defects causing mitotic arrest, subsequent activation of Bak and Bax and the mitochondria-dependent caspase cascade, leading to PARP degradation, 17α-E₂-induced activation of Bak and Bax is the upstream target of proapoptotic action of p53.