Earticle

Purpose: This study was conducted to obtain data for the development of an effective fall risk assessment tool for pediatric inpatients through a systematic review and meta-analysis of the diagnostic test accuracy of existing scales. Methods: A literature search using Medline, Science Direct, CINAHL, EMBASE, and the Cochrane Library was performed between March 1 and 31, 2018. Of 890 identified papers, 10 were selected for review. Nine were used in the meta-analysis. Stata version 14.0 was used to create forest plots of sensitivity and specificity. A summary receiver operating characteristic curve was used to compare all diagnostic test accuracies. Results: Four studies used the Humpty Dumpty Falls Scale. The most common items included the patient's diagnoses, use of sedative medications, and mobility. The pooled sensitivity and specificity of the nine studies were .79 and .36, respectively. Conclusion: Considering the low specificity of the pediatric fall risk assessment scales currently available, there is a need to subdivide scoring categories and to minimize items that are evaluated using nurses' subjective judgment alone. Fall risk assessment scales should be incorporated into the electronic medical record system and an automated scoring system should be developed.

Purpose: The aim of this study was to conduct a systematic review of randomized controlled studies from 2012 to present that explore the diagnostic accuracy of clinical tests used for diagnosing anterior cruciate ligament (ACL) injury. Methods: Study design: Systematic review. Literature search of the PubMed and Scholar databases was conducted using keywords related to diagnostic accuracy of clinical tests for ACL injury. The PRISMA Guidelines were followed to conduct this study. The Cochrane Risk of Bias Tool was utilized to assess the quality of each included study. Results: As a result, 8 studies were included, and 6 clinical tests used in ACL tears were evaluated for diagnostic accuracy. The pivot shift test was reported as having the highest +LR (29.5) value with a sensitivity of 59% and a specificity of 98%. However, the test with the lowest -LR value was the lever test, and the values were as follows: -LR (0.08), +LR (4.7), specificity (80%), sensitivity (94%). Conclusion: In this study, it was concluded that a single clinical test is not sufficient to determine the presence of ACL injury. Test combinations have a higher diagnostic accuracy than a single test. In this study, the accuracy of the clinical tests was examined without considering the amount of ACL rupture and acute-chronic condition. Further research is required to examine the impact of these two factors on diagnostic accuracy of clinical test.

Objective To assess whether the order of test diets influences the results of swallowing studies with regard to their accuracy and safety.Methods Subjects with suspected dysphagia underwent a videofluoroscopic swallowing study (VFSS) and/or a fiberoptic endoscopic evaluation of swallowing (FEES) and repeated the study on the same day or within a week. The order of test diets comprised of two different sets: trial 1 with the fluid first and trial 2 with the semi-solid food first. Main outcome measurements were the modified penetration-aspiration scale (mPAS) and the pharyngeal residue severity scale (PRSS) for the vallecula and the pyriform sinus.Results Sixty-six patients (44 men and 22 women, aged 65.0±15.0 years) were enrolled in this study. Forty-three subjects were evaluated with VFSS only and 23 with both VFSS and FEES. As a result of the swallowing studies, there was no significant difference in each chosen diet sequence regarding mPAS and PRSS. Furthermore, there was no difference regarding the duration of studies, rate of premature study termination, rate of abnormal findings in post-study chest X-ray, and rate of fever or pneumonia post-study.Conclusion The accuracy and safety of the swallowing studies do not rely on the order of test diets.

One-dimensional arrays with subscripts formed by induction variables in real programs appear quite frequently. For most famous data dependence testing methods, checking if integer-valued solutions exist for one-dimensional arrays with references created by induction variable is very difficult. The I test, which is a refined combination of the GCD and Banerjee tests, is an efficient and precise data dependence testing technique to compute if integer-valued solutions exist for one-dimensional arrays with constant bounds and single increments. In this paper, the non-continuous I test, which is an extension of the I test, is proposed to figure out whether there are integer-valued solutions for one-dimensional arrays with constant bounds and non-sing ularincrements or not. Experiments with the benchmarks that have been cited from Livermore and Vector Loop, reveal that there are definitive results for 67 pairs of one-dimensional arrays that were tested.

Objective To assess the accuracies and validities of popular smart bands for heart rate (HR) measurement in cardiovascular disease (CVD) patients during a graded exercise test (GXT).Methods Seventy-eight patients were randomly assigned to wear two different smart bands out of three possible choices: Samsung Galaxy Fit 2, Xiaomi Mi Band 5, or Partron PWB-250 on each wrist. A 12-lead exercise electrocardiogram (ECG) and patch-type single-lead ECG were used to assess the comparative HR accuracy of the smart bands. The HR was recorded during the GXT using the modified Bruce protocol.Results The concordance correlation coefficients (rc) were calculated to provide a measure of agreement between each device and the ECG. In all conditions, the Mi Band 5 and Galaxy Fit 2’ correlations were rc>0.90, while the PWB-250 correlation was rc=0.58 at rest. When evaluating the accuracy according to the magnitude of HR, all smart bands performed well (rc>0.90) when the HR was below 100 but accuracy tended to decrease with higher HR values.Conclusion This study showed that the three smart bands had a high level of accuracy for HR measurements during low-intensity exercise. However, during moderate-intensity and high-intensity exercise, all the three smart bands performed less accurately. Further studies are needed to find a more optimal smart band for HR measurement that can be used for precise HR monitoring during formal cardiac rehabilitation exercise training, including at high and maximal intensity (Clinical Trial Registration No. cris.nih.go.kr/KCT0007036).

.

본 논문은 F-16 전투기 축소모델을 사용한 위상비교 방향 탐지 장치의 방위각 정확도 시험에 대해서 기술하였다. 전투기 하부면에 안테나를 배치하면 항공기 구조물에 의한 전파 반사가 일어나서 방위각 측정오차가 발생한다. 본 연구에서는 F-16 전투기를 5:1로 축소 모델링하고 전투기 하부면에 5개 안테나를 원형으로 설치한 다음 0-360o 방위각에서 1o 간격으로 전파를 수신하였다. 이 때 5개 안테나에 수신되는 전파의 위상을 수치해석으로 계산하여 방향 탐지정확도 시험을 실시하였다. 축소모델을 사용해서 시험한 위상비교 방향 탐지 방식의 방위각 오차는 신호잡음세기가 0dB 이상인 경우 평균 0.5o 이하로 측정되어서 전투기의 위상비교 방향 탐지 장치 설계에 유용하게 활용할 수 있을 것으로 판단된다.

This paper describes the azimuth accuracy test of phase comparison direction finding method using F-16 fighter scale-down model. When the antennas are placed on the bottom of a fighter, reflection signals caused by an aircraft structure arises and an azimuth error occurs. In this research, the F-16 fighter scale-down model was made to 5:1, and five antennas were placed on the bottom of the model, then the radio waves of emitters were received by the antennas in the 0-360o azimuth angles. The accuracy test was performed by numerically analyzing the phases of the radio waves received by the five antennas. The azimuth error of the phase comparison direction finding with scale-down model was measured to be less than 0.5o when the signal noise ratio was larger then 0dB, and it could be very useful for the design of the phase comparison direction finding method of the fighter.

GPS 안전성에 대한 문제가 제기 되면서 ADS-B 대안 기술로 다변측정감시(MLAT; multilatration)시스템이 각광 받고 있다. MLAT는 항공기에 탑재된 트랜스폰더에서 송출되는 Mode A,C,S는 물론 1090ES(ADS-B) 신호를 지상의 여러 개의 수신기로 수신하여 항공기를 탐지하고 위치를 계산하며, GPS를 이용하는 ADS-B 수준의 위치 정밀도를 갖고 있으면서도 ADS-B와 달리 독립적으로 표적 위치를 계산 할 수 있다. 전 세계적인 환경변화에 따라 우리나라에서도 공항지역 이동차량 및 항공기 감시용으로 지역다변측정(LAM; local area multilateration)감시 시스템 개발 하고 있다. LAM시스템에 대한 지상 필드 시험을 태안 비행장에서 진행하고 있다. 본 논문에서는 태안 비행장에서 필드 시험 중인 MLAT의 위치 정확도를 확인하기 위하여, 지상에서 고정된 타겟과 움직이는 타겟에 대하여 MLAT 트렌스폰더와 RTK Rover를 설치하여 MLAT의 위치 정보와 RTK의 위치정보를 확인하여, RTK에서 보정된 위치 정보와 MLAT에서 계산되어진 위치정보를 비교하여 성능을 분석 하였다.

As a GPS stability problem arises, MLAT system is spotlighted as an alternative technology of ADS-B. MLAT system has a high position accuracy as much as ADS-B. Also, MLAT receives the mode A,C,S, and 1090ES(ADS-B) signals from the mounted aircraft transponder. MLAT receives signals from several receiver units and calculates aircraft positions. MLAT has ADS-B level positioning accurarcy using GPS and can calculate the position information with objects independently. According to global environment changes, Local area multiltilateration(LAM) surveillance system is under development for moving vehicles and aircraft detection in airport. These are still under testing in Tae-an Airfield. In the paper, we analyzed the performance by comparing the calculated position data from MLAT to RTK. In order to confirm the position accuracy of MLAT and the deviation of position data between fixed target and moving target on the ground during the field test in Tae-an Airfield.

본 연구의 목적은 P300 숨긴정보검사에서 저역통과필터나 100 ms 구간의 P300 평균전위를 사용하더라도 관련자극과 무관련자극의 시행수 차이로 인하여 관련자극의 P300 진폭이 과대추정된다는 것을 밝히고, 부트스트랩 방법의 1 종 오류율은 유의수준보다 크게 나타나지만 순열 검증의 1 종 오류율은 적절하게 통제된다는 것을 증명하는 것이다. 몬테 카를로 연구를 이용하여 30 Hz, 10 Hz, 또는 5 Hz 저역통과필터를 사용하는 경우와 P300 정점 또는 100 ms 구간의 P300 평균전위를 사용하는 경우에 부트스트랩 방법과 순열 검증의 1 종 오류율과 통계적 검증력을 산출하였다. 연구 결과, 저역통과필터를 사용하거나 100 ms 구간의 P300 평균전위를 사용하여도 관련자극의 P300 진폭이 과대추정되는 정도가 무관련자극의 P300 진폭이 과대추정되는 정도보다 약 1 ㎶ 이상 더 컸다. 이로 인하여 부트스트랩 방법의 1 종 오류율은 유의수준보다 더 높게 나타났다. 그러나 순열 검증의 1 종 오류율은 유의수준과 동일하였다. 순열 검증의 검증력은 5 Hz의 저역통과필터를 사용하고 100 ms 구간의 P300 평균전위를 사용하는 경우에 가장 높았다. 본 연구결과는 P300 숨긴정보검사 뿐만 아니라 시행수가 서로 다른 자극의 P300 진폭을 비교하는 모든 경우에도 활용될 수 있을 것이다.

본 연구에서는 실체적 진실을 알지 못하는 상황에서 두 가지의 통계적인 방법을 이용하여 폴리그라프 검사의 정확도를 추정하였다. 한 가지는 대검찰청과 지방검찰청에서 2000년부터 2004년까지 범죄피의자를 대상으로 수행한 폴리그라프 검사와 검찰처분(기소여부)의 일치율을 이용하여 폴리그라프 검사의 정확도를 추정하였으며, 다른 한 가지는 대검찰청과 지방검찰청에서 2006년에 수행한 폴리그라프 검사결과와 검찰처분의 빈도분포를 가지고 잠재계층분석을 이용하여 정확도를 추정하였다. 판단불능으로 결정된 사례를 제외하고 산출한 일치율로 폴리그라프 검사의 정확도를 추정한 결과, 2000-2004년 자료의 경우는 정확도가 .914 (SE=.004)로 산출되었으며, 2006년 자료의 경우는 정확도가 .885(SE=.021)인 것으로 산출되었다. 2006년 자료에 판단불능 사례를 포함한 후 잠재계층분석을 이용하여 폴리그라프 검사의 정확도를 추정한 결과, 정확도가 .707~.734(SE=.027~.031)인 것으로 산출되었으며, 오류긍정율과 오류부정율은 각각 .078~.087(SE=.019~.023)과 .029~.078(SE=.010~.023)으로 추정되었다. 거짓을 말하는 사람을 정확하게 판별할 확률은 .912~.925(SE=.013~.016)로 높게 나타났고, 진실을 말하는 사람을 정확하게 판별할 확률은 .867~.955(SE=.011~.040)이었다.

The present study examined the accuracy of polygraph tests through two types of statistical methods with unknown ground truth. One method evaluated the accuracy based on the rates of agreements between polygraph test results of crime suspects and prosecutors' indictment decisions for them. Those crime suspects were tested with polygraph by the Prosecutors' Office of the Republic of Korea between 2000 and 2004. The other method estimated the accuracy by using the latent class analysis based on the frequency distributions of the polygraph results and indictments during 2006. Excluding cases that were ‘inconclusive’ on the polygraph test, the study showed that the accuracy of the polygraph tests is .914 (SE=.004) for the 2000-2004 data, and .885 (SE=.021) for the 2006 data. With the inclusion of ‘inconclusive’ cases in the 2006 data, the results from the latent class analysis showed the accuracy in the range between .707 and .734 (SE=.027~.031), with false positives between .078 and .087 (SE=.019~.023), and false negatives between .029 and .078 (SE=.010~.023). The probability that the polygraph test correctly classifies subjects appeared to be in the range between .912 and .925 (SE=.013-.016) for those who lie, and in the range between .867 to .955 (SE=.011-.040) for those who tell the truth.

최근 다축 가공을 통한 정밀가공의 수요가 크게 증대되고 있다. 그러나 5축 이상의 다축 공작기계는 기하학적 관계 가 복잡하여 공작기계의 기하학적 정밀도 평가와 분석이 어렵다. 본 연구에서는 먼저, 여러 가지 규정으로 분산되어 있는 KS/ISO 규격을 정리하여 다축 공작기계의 기하학적 정밀도 평가 항목을 체계화하였다. 또한, 다축 공작기계의 정밀도를 평가하고 분석하기 위해 표준공작물을 제안하였고, 표준 공작물을 사전에 정해진 방법과 절차에 따라 가공한 후, 가공면을 3차원 측정하여 분석하였다. 그 결과 표준공작물의 가공 결과와 공작기계의 정밀도가 정성적 및 정량적으로 매우 유사함을 확인하였다. 이 결과로부터 다축 공작기계의 정밀도 분석이 표준공작물의 가공만으로 가능함을 확인할 수 있었다.

Recently, the demand for precision machining through multi-axis machining has been greatly increased. However, it is difficult to evaluate the geometrical accuracy of the machine tool because of its complicated geometric relationship. In this study, we organized the KS/ISO specifications which are distributed in various regulations, and re-organized the geometrical precision evaluation items of multi-axis machine tools. In addition, a test workpiece was proposed to evaluate and analyze the accuracy of a multi-axis machine tool, and a test workpiece was machined according to predetermined methods and procedures, and then the machined surfaces were measured using CMM. As a result, it was verified that the machining results of the standard workpiece and the precision of the machine tool were very similar qualitatively and quantitatively. From these results, it can be confirmed that the precision analysis of the multi-axis machine tool is possible only by machining the test workpiece.

Recently, a demand for precision 5-axis machine tools is significantly increasing, and the maintenance of machine tool accuracy becomes more important. it is very difficult to evaluate to accuracy of 5-axis M/C in the production site since it needs expensive measuring equipment and skilled engineer. On the other hand, evaluation items of 5-axis M/C are not systematically organized in the existing KS and ISO standards. In this study, the evaluation items for 5-axis M/C were derived systematically and a test workpiece was developed to evaluate the machine tool accuracy more easily. The error sources of machine tool can be estimated by machining and measuring of the test workpiece. The correlation between the machine tool accuracy and the accuracy of machined test workpiece was analyzed. As a result, the accuracy of machined test workpiece represented the accuracy of machine tool and the error sources very effectively.

한국의 검찰에서는 거짓말탐지검사(polygraph test)의 최종판정을 위한 기준 점수로써 관행적으로 -12점을 사용하고 있다. 이 판정기준 점수는 검사기법을 개발한 Backster(1963)가 제안한 판정기준 점수(-13점)와는 약간 다른 것으로 한국에서 거짓말탐지 검사를 표준화하기 위해서는 판정기준 점수에 대한 과학적 근거와 그 정확성에 대한 검증이 필요하다. 본 연구에서는 실제 범죄수사를 위하여 검찰에서 이루어진 거짓말탐지 검사 자료를 이용해서 거짓말탐지 검사의 판정기준 점수를 신호탐지이론에 기초하여 설정하고자 하였다. 또한 Backster가 제안한 판정기준 점수와 현재 검찰에서 사용하는 판정기준 점수, 신호탐지이론에 기초해서 본 연구가 설정한 판정기준 점수의 정확성을 비교하여 범죄수사를 위한 거짓말탐지검사의 검사 상황에 가장 적절한 판정기준 점수를 제시하고자 하였다. 신호탐지이론에 기초해서 거짓말탐지검사의 정확성을 최대화할 수 있는 판정기준 점수를 산출한 결과 -8점으로 나타났다. 또한 판정기준 점수를 도출한 표본자료와 다른 별도의 표본자료에서 판정기준 점수에 따른 정확성을 비교한 결과, -8점을 판정기준 점수로 사용했을 경우의 정확성(83.17%)이 가장 높았고, 검찰의 판정기준 점수(80.20%), Backster의 판정기준 점수(76.24%) 순으로 정확성이 감소하였다. 그러나 오류긍정과 오류부정의 비율을 비교한 결과에서는 -8점을 판정기준으로 사용할 때에 오류긍정이 가장 높은 것으로 나타났다. 따라서 범죄수사를 위한 거짓말탐지검사에서 전체적인 판단 정확률을 확보하는 것이 중요하다면 -8점을 기준 점수로 사용하고, 진실을 말하는 사람을 거짓이라고 판단하지 않는 것이 중요하다면 -12 또는 -13점을 판정기준 점수로 사용할 것이 권고 되었다.

The polygraph test administered by the Korean Prosecutors Office for crime investigations customarily uses the score of -12 as the cutoff point separating the subjects who lie from those who tell the truth. The criterion used by the KPO is different from the one (-13) suggested by Backster (1963) who invented the particular method for lie detection. Based on the signal detection theory applied to the real polygraph test data obtained from real crime suspects by the KPO, the present study identified the score of -8 as an optimal criterion resulting in the highest overall accuracy of the polygraph test. The classification of the subjects with the score of -8 as the criterion resulted in the highest accuracy (83.17%) compared with the accuracies of classifications with the Backster's criterion (76.24%) and the KPO's criterion (80.20%). However, the new criterion was also found to result in more false-positive cases. Based on the results from the present study, it was recommended to use the score of -8 as the criterion when the overall accuracy is important but the score of -12 or -13 when avoiding false-positive is more important than securing the overall accuracy.

본 연구는 상관부트스트랩(bootstrapped correlation difference, BCD)법과 진폭부트스트랩(bootstrapped amplitude difference, BAD)법을 이용하여 P300-기반 유죄지식검사의 탐지정확률을 도출하고 이를 비교하였다. 강기영과 김영윤(2010)의 연구에서 유죄집단과 무죄집단으로 분류된 총 24명의 피험자들의 사건관련전위 데이터를 이용하여 부트스트랩 분석을 수행하였다. 개개 피험자들의 세 가지 자극 유형(목표자극, 탐침자극, 무관련자극)에 대한 두정중심영역(Pz) 혹은 전두중심영역(Fz)에서의 모든 시행의 사건관련전위 진폭값이 분석에 이용되었다. 목표자극은 유죄지식을 포함하지 않는 실험과제와 관련된 자극으로, 모든 피험자는 실험과제에서 목표자극과 목표자극이 아닌 자극을 분류하도록 지시받았다. 탐침자극은 유죄지식을 포함한 자극으로, 무관련자극은 유죄지식을 포함하지 않는 실험과제와 관련이 없는 자극으로 구성되었다. 부트스트랩 분석은 BCD법과 BAD법으로 나뉘어 실시하였다. BCD법은 목표자극과 탐침자극의 사건관련전위 사이의 상관이 탐침자극과 무관련자극의 사건관련전위 사이의 상관보다 큰지를 계산하는 방법이며, BAD법은 탐침자극의 P300 진폭이 무관련자극의 P300 진폭보다 큰지를 계산하여 한 피험자의 유/무죄 여부를 가려내는 방법이다. 부트스트랩 분석 결과 판정불능을 제외한 BCD법의 탐지정확률은 최대 80.0%, 판정불능을 제외한 BAD법의 탐지정확률은 최대 56.5%로 BCD법의 탐지정확률이 더 높게 나타났다. 본 연구에서는 탐지정확률을 계산하는 방법 중 BAD법보다 BCD법이 더 효율적인 방법으로 나타났다.

We compared detection accuracies of P300-based guilty knowledge test based on 2 bootstrap approaches: bootstrapped correlation difference (BCD) and bootstrapped amplitude difference (BAD). Event-related potential data of the guilty group (n=12) and the innocent group (n=12) in Kang and Kim(2010)'s study were subjected to bootstrap analysis. P300 amplitude from all single sweeps of each of target, probe, and irrelevant stimulus at the parietal midline (Pz) or the frontal midline (Fz) electrode site was collected for each participant. Target stimulus is relevant to the experimental task, but not related to the crime. Participants were asked to discriminate the target stimulus from other stimuli. Probe stimulus includes the critical information of the crime, the guilty knowledge, thus guilty participants are supposed to pay attention to it compared to irrelevant stimulus. Irrelevant stimulus is not relevant to the crime and the task. Two different bootstrap analysis were applied to determine individual's guilt or innocence. BCD method estimated double-centered correlation coefficients between the average of target and probe sweeps, and between the average of probe and irrelevant sweeps for a participant. If the former are greater than the latter in 600 trials out of 1,000 bootstrap iterations, the participant is regarded as guilty. BAD method estimated amplitude difference between probe and irrelevant sweeps, and if the difference is positive, the participant is found to be guilty. As a result, BCD method outperformed BAD method. The detection accuracies, without indeterminacy, of BCD and BAD were up to 80.0%, 56.5%, respectively. It seems that BCD method is more appropriate to determine individual's guilt or innocence, and to get high detection accuracy of the guilty knowledge test using event-related potentials.

본 연구는 투사적 그림검사인 빗속의 사람(Person-in-the-Rain, PITR) 해석에서 대규모 언어모델(LLM)의 활용가능성과 프롬프트 리터러시 수준에 따른 해석 정확도 및 내용 완성도의 차이를 탐색하였다. 이를 위해 197명의대학생 PITR 자료와 PDI 응답을 바탕으로 GPT-4o를 활용하여 총 4단계(자유 해석, 저숙련 기준 제시, 고숙련프롬프트, PDI 반영 종합 해석)의 프롬프트 조건에 따라 해석 결과를 수집하고 분석하였다. 해석 정확도는 전문가 채점과의 Cohen’s Kappa 계수로, 내용 완성도는 7개 항목의 코딩 프레임을 통해 평가하였다. 분석 결과, 프롬프트 리터러시가 높을수록 해석의 정확도와 일치도가 향상되었으며, 고숙련 프롬프트 및 PDI 반영 조건에서는 전문가 수준에 근접한 해석이 가능하였다. 본 연구는 LLM의 임상 보조 도구로써의 활용 가능성과 프롬프트 설계 교육의 중요성을 제시하였으며, 이러한 연구 결과를 바탕으로 본 연구의 의의와 제한점에 대해 논의하였다.

This sutdy inquired the differences in interpretation accuracy and content completeness according to the level of prompt literacy of Person-in-the-Rain (PITR) as projective drawing test. To this end, based on PITR data and PDI responses from 197 college students, the study used GPT-4o to collect and analyze interpretation results according to four stages of prompt conditions (free interpretation, low-skilled standard presentation, high-skilled prompt, and comprehensive interpretation reflecting PDI). Interpretation accuracy was evaluated using the Cohen’s Kappa coefficient with expert scoring, and content completeness was evaluated using a coding frame of seven items. The analysis results showed that when the prompt literacy was high, showed more accurate and consistent interpretation, and the high-skilled prompt and PDI conditions enabled the interpretation of the expert level. This study suggested the possibility of using LLM as a clinical assistance tool and the importance of prompt design education, and discussed the significance and limitations based on these research results.

A common problem with neural network learning is that it is too suitable for the specificity of learning. In this paper, various methods were compared to avoid overfitting: regularization, drop-out, different numbers of data and different types of neural networks. Comparative studies of the above-mentioned methods have been provided to evaluate the test accuracy. I found that the more data using method is better than the regularization and dropout methods. Moreover, we know that deep convolutional neural networks outperform multi-layer neural networks and simple convolution neural networks.

Coring has been widely used to measure annual increment in temperate forest ecosystems. This method is attractive because cores can be taken in just one visit. However, the accuracy of this method has not been tested. We expected coring to be less accurate than band dendrometers because of the eccentricity of tree growth. We studied 41 trees at the Long Term Experimental Forest in Mt. Gyebang, which has been monitored with band dendrometers since 1996. We collected two tree cores from the south and north face of each tree, 10 cm below the growth band. Increment cores were measured to 0.01 mm under stereomicroscopy. Annual growth from 1997 to 2006 was 3.2 mm $yr^{-1}$ for Quercus mongolica, 3.5 mm $yr^{-1}$ for Kalopanax septemlobus, and 5.7 mm $yr^{-1}$ for Pinus densiflora. The difference between the two methods was 10% for Q. mongolica, 14% for K. septemlobus, and 4% for P. densiflora. Compaction in the corer and shrinkage during drying decreased diameter increment by 5.6% and 1.0% on P. densiflora, respectively. This study suggests that the two methods for annual increment measurement are very similar, but species specificity should be concerned for direct comparison.

본 논문은 MNIST 데이터셋을 활용한 손글씨 숫자 인식에서 합성곱 신경망(CNN)과 배치정규화(BN)를 결합한 모델을 제안한다. LeCun et al.의 LeNet-5 모델의 성과를 뛰어넘는 것을 목표로 6계층 신경망 구조를 설계하였다. 제안된 모델은 28×28 픽셀 이미지를 입력으로 받아 합성곱, 맥스 풀링, 완전연결계층을 거쳐 처리하며, 특히 배치정규화계층을 도입하여 학습 안정성과 성능을 향상시켰다. 실험에서는 60,000개의 훈련 이미지와 10,000개의 테스트 이미지를 사용하였으며, Momentum 최적화 알고리즘을 적용하였다. 모델 구성에서는 30개의 필터, 필터 사이즈 5×5, 패딩 0, 스트라이드 1을 사용하였고, ReLU 활성화 함수를 채택하였다. 훈련 과정에서는 미니배치 사이즈 100, 총 20 에포크, 학습률 0.1로 설정하였다. 결과적으로 제안된 모델은 99.22%의 테스트 정확도를 달성하여 LeNet-5의 99.05%를 상회하였으며, F1-score 0.9919를 기록하여 모델의 성능을 입증하였다. 또한, 본 논문에서 제안한 6계층 모델은 LeCun et al.의 LeNet-5(7계층 모델)와 Ji, Chun and Kim(10계층 모델)이 제안한 모델보다 더 단순한 구조로 모델의 효율성을 강조하였다. 본 연구의 결과는 AI 비전 검사기 등 실제 산업 응용에서 활용 가능성을 보여주며, 특히 스마트팩토리에서 부품의 불량 상태를 판별하는 데 효과적으로 적용될 수 있을 것으로 기대된다.

In this paper, we proposes a Convolutional Neural Networks(CNN) equipped with Batch Normalization(BN) for handwritten digit recognition training the MNIST dataset. Aiming to surpass the performance of LeNet-5 by LeCun et al., a 6-layer neural network was designed. The proposed model processes 28×28 pixel images through convolution, Max Pooling, and Fully connected layers, with the batch normalization to improve learning stability and performance. The experiment utilized 60,000 training images and 10,000 test images, applying the Momentum optimization algorithm. The model configuration used 30 filters with a 5×5 filter size, padding 0, stride 1, and ReLU as activation function. The training process was set with a mini-batch size of 100, 20 epochs in total, and a learning rate of 0.1. As a result, the proposed model achieved a test accuracy of 99.22%, surpassing LeNet-5's 99.05%, and recorded an F1-score of 0.9919, demonstrating the model's performance. Moreover, the 6-layer model proposed in this paper emphasizes model efficiency with a simpler structure compared to LeCun et al.'s LeNet-5 (7-layer model) and the model proposed by Ji, Chun and Kim (10-layer model). The results of this study show potential for application in real industrial applications such as AI vision inspection systems. It is expected to be effectively applied in smart factories, particularly in determining the defective status of parts.

Background: Bacterial meningitis is a serious infection leading to increased morbidity and mortality every year due to delayed diagnosis and treatment. Previous literatures had shown that cerebrospinal fluid (CSF) procalcitonin outweighs serum procalcitonin to diagnose bacterial infections of the central nervous system. Current meta-analysis aims to find the diagnostic accuracy of serum and CSF C-reactive protein (CRP) to diagnose bacterial meningitis. Material and Methods: PubMed, Google Scholar, Cochrane Library and Google databases were searched from 1st January 1980 to 30th June 2022. Observational studies, prospective or retrospective focusing on C-reactive protein as a biomarker for bacterial meningitis in adult patients were searched. The articles related to serum and CSF CRP for diagnosing bacterial meningitis were explored and retrieved separately, by two independent experts from the published studies available in the electronic search engines. The risk of bias and scholarly quality of studies were evaluated by QUADAS-2. Results: Altogether 637 articles were recognized, out of which 22 studies selected. CSF CRP has shown better diagnostic value than serum CRP. Pooled sensitivity of CSF CRP was 0.89 (95% confidence interval [CI], 0.81?0.94), specificity 0.96 (95% CI, 0.92?0.97), area under the curve (AUC) 0.98 (95% CI, 0.96?0.99), diagnostic odds ratio (DOR) 175 (95% CI, 74?410), positive likelihood ratio (PLR) 20 (95% CI, 11.5?34.1) and negative likelihood ratio (NLR) 0.11 (95% CI, 0.06?0.21). While, pooled sensitivity of serum CRP was 0.80 (95% CI, 0.69?0.88), specificity 0.86 (95% CI, 0.74?0.93), AUC 0.89 (95% CI, 0.86?0.92), DOR 24 (95% CI, 9?62), PLR 6 (95% CI, 2.9?10.7) and NLR 0.23 (95% CI, 0.15?0.37). Heterogeneity was higher for serum CRP than CSF CRP. Conclusion: Our meta-analysis shows that CSF CRP had higher pooled sensitivity, specificity and PLR along with higher AUC and DOR for confirming bacterial meningitis in adults than serum CRP.