• Korean Journal of Artificial Intelligence
• /
• v.12 no.2
• /
• pp.1-7
• /
• 2024

In this paper, we compare three models (logistic regression, Random Forest, and XGBoost) for predicting stroke occurrence using data from the Korea National Health and Nutrition Examination Survey (KNHANES). We evaluated these models using various metrics, focusing mainly on recall and F1 score to assess their performance. Initially, the logistic regression model showed a satisfactory recall score among the three models; however, it was excluded from further consideration because it did not meet the F1 score threshold, which was set at a minimum of 0.5. The F1 score is crucial as it considers both precision and recall, providing a balanced measure of a model's accuracy. Among the models that met the criteria, XGBoost showed the highest recall rate and showed excellent performance in stroke prediction. In particular, XGBoost shows strong performance not only in recall, but also in F1 score and AUC, so it should be considered the optimal algorithm for predicting stroke occurrence. This study determines that the performance of XGBoost is optimal in the field of stroke prediction.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.07a
• /
• pp.397-398
• /
• 2017

정보 공유를 통해 국민들은 의료 기관 및 치료방법 등을 합리적으로 선택하려고 노력하고 있다. 의료 기관의 컴퓨터와 연결 기계의 사용으로 의료 관련 데이터는 규모가 급격히 늘어났다. 이에 따라 정부 3.0에 맞춰 공개되는 의료 데이터가 확대됨에 따라 의료 빅데이터를 통해 여러 정보들을 만들어내 의료 분야에 도움이 될 것이라는 기대가 커져가고 있다. 이 연구에서는 의료 빅데이터를 어떻게 활용하여 의료 기관, 국민, 정부, 보험사 등 여러 기관에게 제공할 수 있는 지에 대해 설명한다. 현재 빅 데이터를 사용해 연령 별 잘 걸리는 질병이나 질병 별 성비를 나타내는 것 등 단순 사실을 알아내는 정도가 아니라 실질적으로 다양한 목적으로 사용될 수 있는 정보를 만들어낼 수 있는 시스템을 구축하였다는 점에서 이 연구는 강점을 갖는다.

• Nuclear Engineering and Technology
• /
• v.53 no.6
• /
• pp.2046-2055
• /
• 2021

This paper analyzed the primary concerns about radiation among the Korean public with a big data analysis of questions posted at the section of "Knowledge iN" on the portal site NAVER in Korea from January 2010 to August 2020. First, we extracted questions about radiation and categorized them into the three categories with TF-IDF analysis: "Medical," "Career Counseling," and "General Interest". The "Medical" category includes questions about radiation diagnosis or treatment. The "Career Counseling" category includes questions about entering college and the prospect of finding jobs in radiation-related fields. The "General Interest" category includes questions about terminology and the basic knowledge of radiation or radioisotopes. Second, we extracted common questions for each category. Finally, we analyzed the temporal change in the numbers of questions for each category to confirm whether there is any correlation between radiation-related events and the number of questions. The analysis results demonstrate that major radiation-related events have little relevance to the number of questions except during March 2011.

• Korean Journal of Digital Imaging in Medicine
• /
• v.16 no.1
• /
• pp.35-42
• /
• 2014

Heart related special images are classified as Cardiac US, XA, CT, MRI. Several Problem is caused by image compression, control and medical support point, so most big hospitals have created a Cadiac PACS differentially in past years. For this reason, create a conflict in inner colleague and patient, protector that result from 2 data processing server operating independently in 1 medical center area. For this reason, we sugges an alternative model of best medical control process together with understand the current situation on medical facility.

• Journal of Digital Convergence
• /
• v.20 no.5
• /
• pp.681-693
• /
• 2022

The purpose of this study is to suggest a plan to utilize atypical data in the health care field by inferring standard medical terms related to the musculoskeletal system through keyword network analysis of medical records of patients hospitalized for musculoskeletal disorders. The analysis target was 145 summaries of discharge with musculoskeletal disorders from 2015 to 2019, and was analyzed using TEXTOM, a big data analysis solution developed by The IMC. The 177 musculoskeletal related terms derived through the primary and secondary refining processes were finally analyzed. As a result of the study, the frequent term was 'Metastasis', the clinical findings were 'Metastasis', the symptoms were 'Weakness', the diagnosis was 'Hepatitis', the treatment was 'Remove', and the body structure was 'Spine' in the analysis results for each medical terminology system. 'Oxycodone' was used the most. Based on these results, we would like to suggest implications for the analysis, utilization, and management of unstructured medical data.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.10
• /
• pp.1179-1187
• /
• 2019

In today's technology environment, most big data-based applications and solutions are based on real-time processing of streaming data. Real-time processing and analysis of big data streams plays an important role in the development of big data-based applications and solutions. In particular, in the maritime data processing environment, the necessity of developing a technology capable of rapidly processing and analyzing a large amount of real-time data due to the explosion of data is accelerating. Therefore, this paper analyzes the characteristics of NiFi, Kafka, and Druid as suitable open source among various open data technologies for processing big data, and provides the latest information on external linkage necessary for maritime service analysis in Korean e-Navigation service. To this end, we will lay the foundation for applying open data framework technology for real-time data processing.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.730-737
• /
• 2018

With the increasing number of health consumers aiming for a happy quality of life, the O2O medical marketing market is activated by choosing reliable health care facilities and receiving high quality medical services based on the medical information distributed on web's blog. Because unstructured text data used on the Internet, mobile, and social networks directly or indirectly reflects authors' interests, preferences, and expectations in addition to their expertise, it is difficult to guarantee credibility of medical information. In this study, we propose a blog reading system that provides users with a higher quality medical information service by classifying medical information blogs (medical blog, ad blog) using bigdata and MLP processing. We collect and analyze many domestic medical information blogs on the Internet based on the proposed big data and machine learning technology, and develop a personalized health information recommendation system for each disease. It is expected that the user will be able to maintain his / her health condition by continuously checking his / her health problems and taking the most appropriate measures.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.4
• /
• pp.15-20
• /
• 2023

Machine learning is comprised of supervised learning, unsupervised learning and reinforcement learning as the type of data and processing mechanism. In this paper, as input and output are unclear and it is difficult to apply the concrete modeling mathematically, reinforcement learning method are applied for crawling robot in this paper. Especially, Q-Learning is the most effective learning technique in model free reinforcement learning. This paper presents a method to implement a crawling robot that is operated by finding the most optimal crawling method through trial and error in a dynamic environment using a Q-learning algorithm. The goal is to perform reinforcement learning to find the optimal two motor angle for the best performance, and finally to maintain the most mature and stable motion about EV3 Crawling robot. In this paper, for the production of the crawling robot, it was produced using Lego Mindstorms with two motors, an ultrasonic sensor, a brick and switches, and EV3 Classroom SW are used for this implementation. By repeating 3 times learning, total 60 data are acquired, and two motor angles vs. crawling distance graph are plotted for the more understanding. Applying the Q-learning reinforcement learning algorithm, it was confirmed that the crawling robot found the optimal motor angle and operated with trained learning, and learn to know the direction for the future research.

• The Journal of Korean Medicine
• /
• v.36 no.3
• /
• pp.98-110
• /
• 2015

Objectives: The purpose of this study was to analyze the medical cost of facial paralysis in payer perspective and to estimate the practice pattern of patient using 2011 Health Insurance Review & Assessment Service-National Patients Sample(HIRA-NPS). Methods: Basic statistical system was used for descriptive analysis of NPS dataset. A table for general information (table20) was extracted by disease code, and social demographic characteristics, distribution of the use among inpatients and outpatients, utilization of each kind of medical care institutions, medical cost were analyzed. Subgroup analysis was conducted for assuming the practice pattern of korean medicine and western medicine. Results: A total of 8,219 people and 64,345 claims data were identified as having facial paralysis. Proportion of outpatient was 95.23%, inpatient 0.84% and patient using both services 3.93%. Mean patient charges was 44,229 won per outpatient, 178,886 won per inpatient and 523,542 won per patient using both services. Utilization of korean medical care institutions was 68.81%(claims), 40.46%(patients), utilization of western medical care institutions was 31.19%(claims), 59.54%(patients). The amount charged by korean medical care institutions was 52.61% and western medical care institutions was 47.39%. Cost per claim was higher than those of the korean treatment and cost per patient of western treatment was lower than those of the korean treatment. Conclusions: The research assessed the medical cost and practice pattern associated with facial paralysis. These findings could be used in health care policy and subsequent studies.

• Journal of Computing Science and Engineering
• /
• v.10 no.1
• /
• pp.21-31
• /
• 2016

With the application and development of biomedical techniques such as next-generation sequencing, mass spectrometry, and medical imaging, the amount of biomedical data have been growing explosively. In terms of processing such data, we face the problems surrounding big data, highly intensive computation, and high dimensionality data. Fortunately, cloud computing represents significant advantages of resource allocation, data storage, computation, and sharing and offers a solution to solve big data problems of biomedical research. In order to improve the efficiency of resource management in cloud computing, this paper proposes a clustering method and adopts Radial Basis Function in order to compress comprehensive data sets found in biology and medicine in high quality, and stores these data with resource management in cloud computing. Experiments have validated that with such a data-compression-based resource management in cloud computing, one can store large data sets from biology and medicine in fewer capacities. Furthermore, with reverse operation of the Radial Basis Function, these compressed data can be reconstructed with high accuracy.