Earticle

본 연구는 세계화의 진전에 따른 국가 간 이동 장벽의 하향화와 국내 산업 및 인구구조의 변화에 따 라 국내에 장기체류하는 외국인 주민이 지속적으로 증가하는 상황 속에서, 이들, 장기체류 외국인 주 민 유입이 해당 지역총생산으로 측정되는 지역경제발달에 줄 수 있는 영향을 계층별 특성을 반영하 여 분석하였다. 외국인 주민은 그들의 입국 목적에 따라 단순기능 인력, 전문 인력, 결혼이주 및 영주 목적의 외국인, 유학생, 기타로 구분될 수 있는데, 지역의 성격에 따라 각 유형의 외국인들의 분포에 차이를 보이고 있다. 외국인 주민은 지역경제에 있어, 값싼 노동력 제공을 통한 생산요소비용절감, 핵심적 창조 인재로서 의 혁신의 주도, 다양하고 관용적인 사회분위기 형성을 통한 지역사회의 문화 융성, 소비를 통한 지 역시장규모의 확대의 역할을 할 수 있지만, 지역 주민들과의 갈등을 야기하거나 내국인 인력의 고용 을 대체하는 역기능을 할 수도 있어, 지역경제력에 대한 외국인 주민의 기여도를 측정할 때에는 그 순기능적 측면과 역기능적 측면을 모두 고려하여야 한다. 본 연구에서는 외국인 출입국 정책본부와 통계청의 각종 지역 통계자료를 바탕으로 하여, 2005년 외 국인 고용허가제가 시행되고, 2007년 재외동포를 대상으로 한 방문 취업제가 시행되면서, 출입국 관 련 정책의 틀이 완성된 2008년 이후 기초지자체 단위의 경제통계의 구득이 가능한 최신년도인 2010 년 자료를 대상으로 하여, 특성별 외국인 주민의 수와 함께 지역경제력의 발달에 영향을 줄 수 있는 인구 특성, 생활환경과 산업기반의 사회간접자본 특성, 지역의 기존 경제력 특성, 집적 경제특성, 혁 신창출 역량특성에 해당하는 변수들을 통제변수로 고려하여 다중회귀분석을 실시하였다. 분석은 매 모형마다 동일한 변수를 투입하되, 지역별 특성을 더미변수를 통해 모형에 반영하였으며, 변수 간 상 관성을 고려하여, 일부 변수에 한해 별도의 모형을 구성하였다. 분석결과, 전체적으로 전문 인력과 단순 기능 인력 외국인 주민이 지역경제발달에 영향을 주는 것으 로 나타났으며, 유학생의 경우 별다른 기여를 하지 못하는 것으로 나타났고, 결혼이주 및 영주 목적 외국인의 대도시에서의 지역경제 발달에 대한 영향은 유의미하지 않았지만, 농촌지역에서는 가장 큰 기여를 하는 것으로 나타났다. 유학생 외국인의 영향 유의미하지 않은 것은 그들의 소비시장 확장에 대한 기여도에 상응할 만큼 내 국인의 비정규 시간제 일자리를 대체함에 따른 영향이 큰 것으로 추정할 수 있으며, 결혼이주 및 영 주 목적 외국인 주민의 도시지역과 농촌지역에서의 지역경제 발달에 대한 기여도의 차이는 농촌지 역에서의 결혼이주 여성의 활동영역이 도시지역에 비해 더욱 건전하며, 그들의 남편의 사회경제적 능력 또한 도시지역에 비해 농촌지역이 더 건실하기 때문인 것으로 추정할 수 있다.

This study analyzed the effects of foreign residents staying for a long time on local economic growth in circumstance lowering the international migration’s barrier and change of domestic industrial and population structure. By their aims of entry, they are classified as 4 groups which are workforce, professionals, students, marriage migration and settler, and etc. The distribution of them shows clear distinctions by local characteristics. As for local economic growth, the foreign residents play a role as reducing cost of the elements of production through supplying of cheaper labor, taking leads in regional innovation as creative talented persons, making tolerant social atmosphere and expanding local consumer market. However, they also may come into conflict with domestic residents or replace domestic job occupation with themselves. Therefore when it comes to measuring the effects of the foreign residents as for local economic growth, we ought to consider the right function and the reverse of them as well. In the result of the regression analysis, both the professional and the workforce entirely and positively affected the local economic growth. However, the students did not affect the local economic growth. In the case of rural area, the marriage immigrants positively affected the local economic growth. As for interpretation of the results, the authors can presume the cause of the higher professional’s influence, and also presume the students’s neutral effect resulting from the loss of their replacement with domestic worker’s part time job as much as their consumption in their local region. Eventually, the difference of analysis result for marriage migration and settler in metropolis and rural municipality as well results from difference of soundness of employment opportunities in two different regions and their husbands’s ability maintaining their livelihood.

To deal with the problems of occlusion, pose variations and illumination changes in the object tracking system, a regression model weighted multi-templates mean-shift (MS) algorithm is proposed in this paper. Target templates and occlusion templates are extracted to compose a multi-templates set. Then, the MS algorithm is applied to the multi-templates set for obtaining the candidate areas. Moreover, a regression model is trained to estimate the Bhattacharyya coefficients between the templates and candidate areas. Finally, the geometric center of the tracked areas is considered as the object's position. The proposed algorithm is evaluated on several classical videos. The experimental results show that the regression model weighted multi-templates MS algorithm can track an object accurately in terms of occlusion, illumination changes and pose variations.

기후변화와 산업 발전으로 인해 자원 고갈 위기가 커지면서, 물, 에너지, 식량 자원을 연계하여 통합적으로 관리하는 것이 중요해지고 있다. 기존의 개별 관리 방식으로는 복잡한 상호작용을 반영하지 못해, 복합적인 자원 위기에 효과적으로 대응하기 어렵다. 본 연구에서는 충청남도 지역을 대상으로 자원 간의 상호작용을 정량적으로 평가하기 위하여, 장단기 메모리(LSTM)와 다중 출력 회귀(MOR) 모형을 결합한 복합 물-에너지-식량(WEF) 넥서스 시스템을 구축하였다. 2014년~2022년 기간의 강수량, 저수량, 용수 생산량 및 이용량, 에너지 발전 소비량, 농축산 생산·소비량과 부문별 공급 및 생산량으로부터 산정된 스트레스 지수를 시스템의 입력 및 출력 인자로 활용하였다. 그 결과, 충청남도 WEF 넥서스 시스템에서는 물과 식량 부문이 가장 지배적인 상호작용을 보였으며, 강수량은 식량작물 생산뿐만 아니라 에너지 생산에도 크게 영향을 미치는 것으로 나타났다. 특히, 물 스트레스가 전체의 45% 이상을 차지하며, 충청남도의 시스템 취약성을 결정하는 핵심 인자로 확인되었다. LSTM을 활용하여 과거 기간의 부문별 자원의 수요량 및 공급량 변동성을 모의하고 모형의 재현성을 검증한 결과, R_a² 는 0.49~0.78, KGE는 0.47~0.89의 범위로 나타났다. 반면, LSTM과 MOR 모형이 결합된 복합 넥서스 시스템의 R_a² 와 KGE는 LSTM만 적용한 결과에 비해 높게 나타났다. 따라서 본 연구에서 개발한 LSTM-MOR 기반의 WEF 넥서스 시스템은 기후변화 시나리오별 민감도 분석을 통해 지역 단위의 넥서스 평가를 가능케 하며, 향후 자원 관리 및 정책 결정을 지원하는 핵심적인 도구로 활용될 수 있을 것이다.

As climate change and rapid industrialization exacerbate resource scarcity, the importance of an integrated management approach that links and manages water, energy, and food resources has become increasingly critical. The existing fragmented management methods are ill-equipped to address the intricate interactions between resources, making it challenging to respond effectively to complex crises. To quantitatively assess resource interactions in the Chungcheongnam-do, this study developed a hybrid water-energy-food (WEF) nexus system that combines long short-term memory (LSTM) and multioutput regression (MOR). The system employed various input and output factors spanning 2014 to 2022, including precipitation, reservoir storage, water supply and usage, energy generation and consumption, and agricultural and livestock production and consumption, along with stress indices calculated from sectoral supply and production. The results indicated that the water and food sectors exhibited the most dominant interaction within the Chungcheongnam-do WEF nexus system. Furthermore, it was found that precipitation significantly influenced not only crop production, but also energy generation. Especially, the water stress index accounted for over 45% of the total nexus stress, identifying it as the critical determinant of system vulnerability in the region. The simulation of historical resource supply and demand fluctuations by sector using LSTM verified the model's reproducibility, yielding R_a² values ranging from 0.49 to 0.78 and KGE values ranging from 0.47 to 0.89. In contrast, the hybrid nexus system combining LSTM-MOR yielded higher R_a² and KGE than the standalone LSTM model. Consequently, the LSTM-MOR hybrid model developed in this study enables comprehensive nexus evaluations through sensitivity analysis for climate change scenarios, making it an essential asset for informed resource management and policy-making.

최대자율수축의 백분율(%MVC)과 같은 개인요소는 정적인 근육작업에서 필요한 근지구력시간과 유의성 있게 연관되어 있다고 알려져 왔다. 본 연구에서는 등장성 수축 운동 시 상완 이두근의 근지구력시간을예측하는 회귀모델에 관하여 연구하였다. 8명의 건강한 피검자가 자원하여 10%MVC강도의 등장성 수축운동을 소진할 때까지 각각 5회분씩 수행하였다. 개인요소와 신체측정치수를 근지구력시간의 예측자로 하는회귀모델들에서 산출된 결정계수와 p-값을 비교하였다. 그 결과 다중회귀모델에서 유의성 있는 결정계수가있었으며 근지구력시간을 예측하기에 유용함을 알 수 있었다.

Previous studies have shown that personal factors such as the percentage of maximal voluntary contraction(%MVC) were significantly related to endurance time for static muscular work. The present study investigated regression models to predict endurance time of biceps Brachii muscle during isotonic contractions. Eight healthy subjects volunteered for this study, and performed five test sessions of isotonic contraction exercises at 10% of MVC until their exhaustion, respectively. The determinant coefficients and p-values were compared among regression models using personal factors and anthropometrical data as predictors of endurance time. The results demonstrated that multi-regression model had significant coefficients of determination, and was useful to predict endurance time.

Making a contract of small private construction works is very often disregarded important or subsidiary contract details which strongly recommended in standard contract sheet. To have a well fulfilled contract sheet with detailed agreements are very symbolic to promising their rights and obligations to each other until they meet the final objectives of completion, delivery, and receipt of construction payments. Therefore, a almost perfect contract give us less disputes for payments during construction or after completion of work. In this study, a logistic regression model was used to determine the effect of subcontract items from survey. The results show that contract sheet, design alternation and cancellation of contract are the most important contract conditions, and derived a logistic regression model for reducing a dispute between both personal contract.

Accurate and continuous environmental monitoring is a key element in implementing smart cities to protect civic health, optimize urban services, and make data-driven policy decisions. However, real-time, city-wide measurement of multidimensional environmental indicators such as CO₂, PM2.5, and VOCs requires the installation of large-scale physical sensors, which poses practical limitations such as cost, maintenance, and spatial constraints. To address this issue, this study proposes a multi-target regression-based soft sensor framework that simultaneously predicts multiple environmental indicators using readily available auxiliary data in cities, such as traffic volume, weather information, and population density. Using techniques such as Random Forest, LightGBM, and Multi-Output Regressor, we construct an integrated prediction model that considers the correlation between various output variables. Even in areas with limited monitoring stations, we achieve an average R² of over 0.80. The proposed model can be integrated with smart city public services, environmental policies, and real-time alert systems to enhance the efficiency and responsiveness of urban environmental management.

Significant color discrepancies between the different camera views can be observed in multi-view video sequences. In this paper, a color adjustment algorithm based on support vector regression is proposed. A mapping function is established by extracted feature points from original-image and target-image. Then the mapping function is applied to the original image to obtain the corrected image. Experimental results show that the proposed method can produce good correction result. It also shows that the color differences between multi-view video can be effectively reduced by SVR.

Offline signature verification system finds several applications in monitory transaction systems like banks. However one of the major challenges in this direction is the capability of the system to detect skilled and unskilled forgery. Many cases of bank check forgeries have been reported. Most of the offline signature verification system adopts recognition based technique where the system classifies a given signature sample as one of the samples from the database. However detection of a forgery in a given sample is challenging as the input sample looks similar to one of the samples in the database. In this paper we propose an innovative approach for offline signature verification with polar feature descriptor for signature that contains Radon Transform and Zernike Moments. Verification is performed using Multiclass Support Vector Machine. Once a signature is verified as being of a registered class, PLS Regression is applied on the sample against all samples in the database of the verified user to obtain regression score. Log Likelihood of the sample against all sample of the user is calculated using Hidden Markov Model. Authenticity of the classification is justified if the regression score and Log Likelihood distance deviation is less than 5%. Results show that the system verifies signature with an accuracy of 98% with false acceptance rate of .8%. Proposed system also detects skilled forgery with an accuracy of 71% and Random forgery with an accuracy of 76%.

In this paper, a novel approach to multi-hop range-free localization algorithm in wireless sensor network is proposed using principal component regression. The localization problem in the wireless sensor network is formulated as a multiple regression problem, which is resolved by principal component regression. The proposed methods are simple and efficient that no additional hardware is required for the measurements, and only hop-counts information and location information of the beacons are used for the localization. The proposed method consists of two phases: the offline training phase and the online localization phase. In offline training phase, the real distances and the hop-counts among sensor nodes are collected to build localization model. In online localization phase, each unknown sensor node finds its own location using the localization model. The experimental results show that compared with previous localization methods, the proposed method exhibits excellent and robust performances not only in the isotropic sensor networks but also in the anisotropic sensor networks.

CCTV는 범죄상황 발생시 보안과 증거확보를 위해 사용되어 왔다. 그러나 실제 상황에서 범죄가 발생하기 전 예방을 하는 것 보다 사후 처리에 용도를 두고 있으며, 범죄 예방의 목적에 대해 미미한 효과를 보이고 있다. 본 논문에서는 CCTV로 수집된 보행자의 데이터를 통해 객체의 행동을 분석하여 위험도로 행동의 위험여부를 추정하기 위한 Flexible Multi-level Regression 모델을 제안하였다. 제안된 모델을 통해 관찰된 객체의 행동이 이상행동이라고 판단될 시 위험을 받는 객체에게 알림을 주어 범죄 발생 전 즉각적인 대응이 가능하며 빠른 상황판단이 가능할 것으로 예상된다.

Plate(후판) design은 수요가 주문시 지정size(두께, 폭)로 부터 당사 압연 process를 거치면서 발생하는 지시대비 실적간의 차이를 보정하여 최종적으로 산출하게 되며, 이러한 과정은 제품생산시 size 부족으로 인한 불량 발생을 방지하는데 그 목적이 있다. Process진행중 size실적은 .gamma.-ray등 각종 측정기기로 부터 자동 측정되며 이는 process computer로 부터 main computer로 일별 전송되어 3개월 동안 조업관리 DATA BASE에 누적관리되고 있다. 본 연구는 이러한 조업실적을 근거로 제조과정에서 발생하는 size오차를 probability theory과 MULTI-REGRESSION 기법을 적용하여 DESIGN LOGIC을 개발, 제품 실수율을 향상하는데 그 목적이 있다.

A regression with multi-dimensional responses is quite common nowadays in the so-called big data era. In such regression, to relieve the curse of dimension due to high-dimension of responses, the dimension reduction of predictors is essential in analysis. Sufficient dimension reduction provides effective tools for the reduction, but there are few sufficient dimension reduction methodologies for multivariate regression. To fill this gap, we newly propose two fused slice-based inverse regression methods. The proposed approaches are robust to the numbers of clusters or slices and improve the estimation results over existing methods by fusing many kernel matrices. Numerical studies are presented and are compared with existing methods. Real data analysis confirms practical usefulness of the proposed methods.

This study develops a spatiotemporal geospatial-statistical framework to analyse the impacts of changes in atmospheric pollutants and greenhouse gases on Land Surface Temperature (LST) over Iraq during 2015-2024. It integrates satellite and ground-based observations, including major atmospheric gases, Normalized Difference Vegetation Index (NDVI), LST, and key climatic variables across three climatic regions(North, Middle, and South Iraq) classified according to the Köppen-Geiger climate system. The Air Pollution Composite Index (APCI), based on principal component analysis, and the Greenhouse Gas Index (GHGI) were developed. At the same time, temporal trends were assessed using the Mann-Kendall test, and linear and multiple linear regression models were applied to identify the factors controlling LST. The results revealed significant decreasing trends in APCI across northern and middle Iraq (Sen's slope = -0.189 and -0.163; p < 0.05), mainly associated with significant reductions in carbon monoxide (CO) concentrations across all regions(p ≤ 0.002).In contrast, GHGI and carbon dioxide (CO₂) exhibited highly significant increasing trends (p = 8.3 × 10^-5), with GHGI Sen's slope values ranging from 0.346 to 0.348. Methane (CH₄) also showed significant increases, particularly in southern Iraq (Sen's slope = 0.0069), whereas nitrogen dioxide (NO₂) and ozone (O₃)showed no significant long-term trends.Airtemperature was consistently identified as the dominant predictor of LST across all models, whereas APCI and GHGI showed weaker, regionally variable contributions, indicating that meteorological controls exceeded atmospheric composition effects in explaining seasonal LST variability. Northern Iraq exhibited more stable environmental responses, whereas middle and southern Iraq showed greater complexity due to contrasting climatic and surface conditions. Overall, the findings reveal a decoupling between improving air quality and increasing greenhouse gas accumulation, highlighting the need for region-specific environmental management strategies.

The goal of multi-output regression is to predict multiple realvalued output variables. Research in this area is lacking compared to multi-output classification . In this approach multi-output kernel regression based on kernel regression in which different outputs are assumed to be correlated. We also introduce a modelselection method employs generalized crossvalidation function for choosing optimal values of hyperparameters. Numerical results from synthetic and real datasets are then obtained to illustrate that the proposed outperforms the other methods on multi-output regression problems.

In this paper, we prove that multi-variate fuzzy polynomials are universal approximators for multi-variate fuzzy functions which are the extension principle of continuous real-valued function under $T_W-based$ fuzzy arithmetic operations for a distance measure that Buckley et al.(1999) used. We also consider a class of fuzzy polynomial regression model. A mixed non-linear programming approach is used to derive the satisfying solution.

고차원 데이터는 기계학습에서 높은 시간 소모와 큰 메모리 요구로 학습에 어려움을 유발한다. 특히 다중 레이블 환경에서는 레이블의 개수만큼 더 높은 복잡도를 요구한다. 본 논문에서는 다중 레이블 환경에서 분류 성능 향상을 위한 특징 선별 기법을 제안한다. 중요한 특징을 선별하기 위해 특징과 특징, 특징과 레이블, 레이블과 레이블, 세 가지 관계를 고려했으며 이를 위해 회귀 기반 목적 함수를 설계하였다. 이 목적 함수는 특징과 레이블 사이의 선형적인 관계를 계산하고 상호 정보 기반의 특징과 특징, 레이블과 레이블 사이의 관계를 계산한다. 이 목적 함수를 최소화하는 가중치를 찾아 특징을 선별할 수 있다. 이 목적 함수 최적화를 위해 경사하강법 방식을 제시하여 빠르게 수렴할 수 있는 알고리즘을 제안하였다. 여섯 개의 다중 레이블 데이터에 대한 실험 결과 제안된 방법이 기존 다중 레이블 특징 선별 기법보다 높은 성능을 보여주었다. 여섯 개의 데이터 평균으로 본 제안 방법의 분류 성능은 해밍 로스 0.1285, 랭킹 로스 0.1811, 다중 레이블 정확도 0.6416으로, 비교 대상 알고리즘인 AMI(Approximating Mutual Information)에 비해 각각 0.0148, 0.0435, 0.0852 더 우수한 성능을 보였다.

High-dimensional data causes difficulties in machine learning due to high time consumption and large memory requirements. In particular, in a multi-label environment, higher complexity is required as much as the number of labels. This paper proposes a feature selection method to improve classification performance in multi-label settings. The method considers three types of relationships: between features, between features and labels, and between labels themselves. To achieve this, a regression-based objective function is designed. This objective function calculates the linear relationships between features and labels and uses mutual information to compute relationships between features and between labels. By minimizing this objective function, the optimal weights for feature selection are found. To optimize the objective function, a gradient descent method is applied to develop a fast-converging algorithm. The experimental results on six multi-label datasets show that the proposed method outperforms existing multi-label feature selection techniques. The classification performance of the proposed method, averaged over six datasets, showed a Hamming loss of 0.1285, a ranking loss of 0.1811, and a multi-label accuracy of 0.6416. Compared to the AMI(Approximating Mutual Information) algorithm, the performance was better by 0.0148, 0.0435, and 0.0852, respectively.

Urban centers worldwide are facing changes in terms of environmental standards and an increase in temperatures. Recent significant developments, particularly in developing nations, have led to an increase in Land Surface Temperature (LST) and surface thermal variations in urban areas, accompanied by rising levels of atmospheric pollutants such as nitrogen dioxide (NO₂), ozone (O₃), carbon monoxide (CO), sulfur dioxide (SO₂), and aerosols. Despite growing concerns about urban warming and air pollution, very few studies have simultaneously integrated multisource atmospheric pollutants, LST, precipitation, and high-resolution Land Use/Land Cover (LULC) dynamics into an analytical workflow using two independently trained Deep Learning Neural Network (DLNN) models, particularly for rapidly growing cities in developing countries. The novelty of this study lies in two separate DLNN architectures-one optimized for LULC classification using PlanetScope multispectral imagery and another for LST estimation using atmospheric pollutant concentrations-and their outputs were synthesized into a cohesive analytical workflow. This dual-model approach, combined with multi-sensor data integration, enables a comprehensive assessment of the relationships between urban expansion, atmospheric pollution, and thermal environmental changes. The LULC change detection analysis reveals a net increase of 8.34% in urban built-up areas, coupled with increases in atmospheric pollutants and LST. This study utilizes high-resolution multispectral remote sensing data from PlanetScope and atmospheric pollutant concentration data from Sentinel-5P TROPOMI, along with LST data from MODIS, making it a multi-sensor and multi-parametric remote sensing approach to analyze urban environmental changes in Patna city, Bihar, India. The findings provide useful evidence to support urban environmental monitoring and planning decisions in Patna.

OBJECTIVES:We analyzed dietary patterns using reduced rank regression (RRR), and assessed how well the scores extracted by RRR predicted stroke in comparison to the scores produced by partial least squares and principal component regression models. METHODS: Dietary data at baseline were used to extract dietary patterns using the 3 methods, along with 4 response variables: body mass index, fibrinogen, interleukin-6, and low-density lipoprotein cholesterol. The analyses were based on 5,468 males and females aged 45-84 years who had no clinical cardiovascular disease, using data from the Multi-Ethnic Study of Atherosclerosis. RESULTS: The primary factor derived by RRR was positively associated with stroke incidence in both models. The first model was adjusted for sex and race and the second model was adjusted for the variables in model 1 as well as smoking, physical activity, family and sibling history of stroke, the use of any lipid-lowering medication, the use of any anti-hypertensive medication, hypertension, and history of myocardial infarction (model 1: hazard ratio [HR], 7.49; 95% confidence interval [CI], 1.66 to 33.69; p for trend=0.01; model 2: HR, 6.83; 95% CI, 1.51 to 30.87 for quintile 5 compared with the reference category; p for trend=0.02). CONCLUSIONS: Based primarily on RRR, we identified that a dietary pattern high in fats and oils, poultry, non-diet soda, processed meat, tomatoes, legumes, chicken, tuna and egg salad, and fried potatoes and low in dark-yellow and cruciferous vegetables may increase the incidence of ischemic stroke.

Purpose Studies comparing radical prostatectomy (RP) outcomes with those of radiotherapy with or without androgen deprivation therapy (RT±ADT) for prostate cancer (PCa) have yielded conflicting results. Therefore, we used propensity score-matched analysis and competing risk regression analysis to compare cancer-specific mortality (CSM) and other-cause mortality (OCM) between these two treatments. Materials and Methods The multi-center, Severance Urological Oncology Group registry was utilized to identify 3,028 patients with clinically localized or locally advanced PCa treated by RP (n=2,521) or RT±ADT (n=507) between 2000 and 2016. RT±ADT cases (n=339) were matched with an equal number of RP cases by propensity scoring based on age, preoperative prostate-specific antigen, clinical tumor stage, biopsy Gleason score, and Charlson Comorbidity Index (CCI). CSM and OCM were co-primary endpoints. Results Median follow-up was 65.0 months. Five-year overall survival rates for patients treated with RP and RT±ADT were 94.7% and 92.0%, respectively (p=0.105). Cumulative incidence estimates revealed comparable CSM rates following both treatments within all National Comprehensive Cancer Network risk groups. Gleason score ! 8 was associated with higher risk of CSM (p=0.009). OCM rates were comparable between both groups in the low- and intermediate-risk categories (p=0.354 and p=0.643, respectively). For high-risk patients, RT±ADT resulted in higher OCM rates than RP (p=0.011). Predictors of OCM were age ! 75 years (p=0.002) and CCI ! 2 (p < 0.001). Conclusion RP and RT±ADT provide comparable CSM outcomes in patients with localized or locally advanced PCa. The risk of OCM may be higher for older high-risk patients with significant comorbidities.