• The KIPS Transactions:PartC
• /
• v.18C no.6
• /
• pp.379-388
• /
• 2011

Ensuring the security of medical records is becoming an increasingly important problem as modern technology is integrated into existing medical services. As a consequence of the adoption of EMR(Electronic Medical Records) in the health care sector, it is becoming more and more common for a health professional to edit and view a patient's record. In order to protect the patient's privacy, a secure authentication model to access the electronic medical records system must be used. A traditional identity based digital certificate for the authenticity of EMR has private key management and key escrow of a user's private key. In order to protect the EMR, The traditional authentication system is based on the digital certificate. The identity based digital certificate has many disadvantages, for example, the private key can be forgotten or stolen, and can be easily escrow of the private key. Nowadays, authentication model using fingerprint recognition technology for EMR has become more prevalent because of the advantages over digital certificate -based authentication model. Because identity-based fingerprint recognition can eliminate disadvantages of identity-based digital certificate, the proposed authentication model provide high security for access control in EMR.

• Information Systems Review
• /
• v.25 no.1
• /
• pp.165-187
• /
• 2023

The continuous increase in life expectancy and high interest in health has brought about significant changes in the use of health information by the public according to the development of information technology represented by the Internet and smartphones. As the medical market expands to the mobile health environment, many health-related apps have been created and distributed, but the acceptance rate is slow as it has become challenging to provide services due to various regulations. In this study, perceived value, perceived risk factors (psychological risk, risk of time-loss, legal risk), and perceived benefits (usefulness, interaction, autonomy) were derived and verified as factors that affect the acceptance resistance of personal health record apps based on the privacy calculation model. In addition, by analyzing the moderating effect of trust in the manufacturer, how the perceived risk and perceived benefit affect the perceived value was verified. A survey was conducted on Korean college students who recognized the personal health record apps but did not use them, and 127 samples were analyzed using structural equations. As a result of hypothesis verification, perceived value has a negative effect on acceptance resistance, perceived risk (risk of time-loss) has a negative effect on perceived value, and perceived benefits (usefulness, interaction, autonomy) were found to have a positive effect on perceived value. Trust in manufacturers has weakened the impact of perceived risks (legal risk) on perceived values. This study is expected to play an important role in maintaining a competitive advantage in the personal health record app market environment by identifying and proposing detailed criteria for reducing the acceptance resistance of personal health record apps.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.302-310
• /
• 2022

Efficient use of limited blood products is becoming very important in terms of socioeconomic status and patient recovery. To predict the appropriateness of patient-specific transfusions for the intensive care unit (ICU) patients who require real-time monitoring, we evaluated a model to predict the possibility of transfusion dynamically by using the Medical Information Mart for Intensive Care III (MIMIC-III), an ICU admission record at Harvard Medical School. In this study, we developed an explainable machine learning to predict the possibility of red blood cell transfusion for major medical diseases in the ICU. Target disease groups that received packed red blood cell transfusions at high frequency were selected and 16,222 patients were finally extracted. The prediction model achieved an area under the ROC curve of 0.9070 and an F1-score of 0.8166 (LightGBM). To explain the performance of the machine learning model, feature importance analysis and a partial dependence plot were used. The results of our study can be used as basic data for recommendations related to the adequacy of blood transfusions and are expected to ultimately contribute to the recovery of patients and prevention of excessive consumption of blood products.

• Child Health Nursing Research
• /
• v.7 no.1
• /
• pp.85-95
• /
• 2001

The purpose of this study was to reveal what influences the divergent methodological researches have brought the nursing practice in during the past 3 decades. The nursing record sheets ie, the nursing discourses were analyzed to know the knowledges that were recorded, accepted and communicated in nursing practice at pediatric intensive care units, and unclosed the philosophical and methodological position of that knowledges. The texts were 13 sheets, 3 kinds of nursing record(7 24hours flow charts, 4 nursing information record sheets and 2 transfer record sheets) used at 4 hospitals. The unit of analysis was 'word'. First, all words of the sheets were listed up, clustered into categories based on their contents. And then, the larger conceptual themes were drawn to elucidate the effect of the knowledge/power and the philosophical and methodological position of that knowledges. To enhanced the validity of the analysis, the data were analyzed by two researchers. The 'words' were classified into 3 categories; 'general information', 'assessment' and 'inter-vention'. The conceptual themes of the texts were 'the gaze for quantification and objectification' and 'technical/assimilated caring'. This themes reflected the logic positivistic and biomedical view that had dominated at clinical practice. Nursing has endeavored to resist the logic-positivistic knowledge/power and to established the nursing knowledge/power based on multiple philosophies and methodologies, especially phenomenological-interpretative. But the results of this study revealed that such efforts in nursing theory and research couldn't influenced the knowledge of practice. Logic positivism was yet so strong and the biomedical model yet dominated in the clinical practice. It identified that the borrowed theory and the knowledge from the received view gave nursing the power. But they were modalities that reinforced the dominant, medical power. Nursing has investigate the other positions (feminism, Habermas' critical social theory and Foucault's discourse theory). This positions suggest different assumptions but share the common concepts; equality, emancipation and freedom. The important point is how make these concepts the practical for nursing knowledge/power in practice. We must recognize that the praxis at clinical setting take place at the field unlike theoretical praxis. The change of clinical practice is the social, economic and political change.

• Quality Improvement in Health Care
• /
• v.27 no.2
• /
• pp.57-72
• /
• 2021

Purpose: At present, there are a variety of serious patient safety incidents related to problems in health information technology (HIT), specifically involving electronic medical records (EMRs). This emphasizes the need for an enhanced electronic medical record system (EMRS). As such, this study analyzed both the nature of and potential to prevent incidents associated with HIT/EMRS based on data from the Korea Patient Safety Reporting and Learning System (KOPS). Methods: This study analyzed patient safety incidents submitted to KOPS between August 2016 and December 2019. HIT keywords were used to extract HIT/EMRS incidents. Each case was reviewed to confirm whether the contributing factors were related to HIT/EMRS (HIT/EMRS-related incidents) and if the incident could have been prevented (HIT/EMRS-preventable incidents). The selected reports were summarized for general clarity (e.g., incident type, and degree of harm). Results: Of the 25,515 obtained reports, 2,664 incidents (10.4%) were HIT-related, while 2,525 (9.9%) were EMRS-related. HIT/EMRS-related incidents were the third largest type of incident followed by 'fall' and 'medication incidents.' More than 80% of HIT/EMRS-related incidents were medication-related, accounting for approximately one-third of the total number of medication incidents. Approximately 10% of HIT/EMRS-related incidents resulted in patient harm, with more than 94% of these deemed as preventable; further, sentinel events were wholly preventable. Conclusion: This study provides basic data for improving EMR use/safety standards based on real-world patient safety incidents. Such improvements entail the establishment of long-term plans, research, and incident analysis, thus ensuring a safe healthcare environment for patients and healthcare providers.

• The Journal of Korean Physical Therapy
• /
• v.10 no.2
• /
• pp.13-32
• /
• 1998

A questionaire was conducted to obtain ran satisfactions in information of the 325 disabled persons among the total 9,314 handicapped people in Taejon area, and was surveyed during the period of June 1 to August 31, 1997. The results are as follows: 1. Among the studied disabled persons, $54.5\%$ of male, and $45.5\%$ of female. 2. Before disabled in occupation, $32.0\%$ of out of work group were high, $6.5\%$ of farm, student group were low. Before disabled in occupation by gender, male group is $29.9\%$ of out of work group were high, $0.6\%$ of housework group were low. female group is$34.5\%$ of out of work group were high, $4.7\%$ of student group were low(P<0.001). 3. After disabled in occupation, $75.1\%$ of out of work group wert high, $10.8\%$ of in working group were low. After disabled in occupation by gender, male group is $87.6\%$ of out of work group were high, $1.7\%$ of housework group were low. female group is $60.1\%$ of out of work group were hgh, $10.8\%$ of in working group were low(P<0.001). 4. Medical security status, $64.9\%$ of medical aid group wore high, $35.1\%$ of medical insurance group were low. Medical security status by gender, male group is $71.2\%$ of medical aid group were high, $28.8\%$ of medical insurance group were Iew. female group is $57.4\%$ of medical aid group wan high, $42.6\%$ of medical insurance were low(P<0.01). 5. Disabled record status, $68.6\%$ of record group were high, $31.4\%$ of non group were low. Disabled record status by gender, male group is $78.5\%$ of record group were high, $21.5\%$ of non record group were low. female group is $56.6\%$ of record group were high, $43.4%$ of non record group were low(P<0.001). 6. Disabled duration status, $42.2\%$ of loss than 9 year group were high, $10.2\%\;of\;20-29,\;30-39$ year group were low. Disabled duration status by gender,'male group is $44.6\%$ of less than 9 year group were high, $6.2\%$ of 20-29 year group wert low. female group is $39.2\%$ of less than 9 year were high, $39.2\%$ of 30-39 year group were low (P<0.05). 7. Cause of disabled status, $26.5\%$ of other group, $23.7\%$ of congenital group were high. $9.2\%$ of unknown group, $6.8\%$ of industry accident, $2.5\%$ of drug poisoning group were low. Cause of disabled status by gender, male group is $27.7\%$ of other group, $23.7\%$ of congenital group were high, $2.3\%$ drug poisoning group were low. female group is $25.0\%$ of other group, $20.9\%$ of congenital group were high, $2.5\%$ of drug poisoning group were low (P<0.001). 8. Disabled type status, $19.4\%$ of double disabled group were high, $2.2\%$ of muscle paralysis group were low. Disabled type status by gender, male group is $22.0\%$ of double disabled group were high, $2.3\%$ of muscle paralysis group were low. female group is $23.3\%$ of rheumatism group were high, $0.7\%$ of amputation group were low(P<0.001). 9. Smoking status, $73.2\%$ of non smoking group were high, $26.8\%$ of smoking group were low. Smoking status by gender, male group is $59.9\%$ of double non smoking group were high, $40.1\%$ of Smoking group were low, female group is $89.2\%$ of non smoking group were high, $10.8\%$ of smoking group were low(P<0.001). 10. Drinking status, $80.0\%$ of non drinking group were high, $20.0\%$ of drinking group were low. Drinking status by gender, male group is $72.3\%$ of non drinking group were high, $27.7\%$ of drinking group were low. female group is $89.2\%$ of non drinking group were high, $10.8\%$ of drinking group were low(P<0.001). 11. Stress level status, $52.9\%$ of high stress group were high, $1.8\%$ of very severe stress group were low. Stress level status by gender, male group is $50.8\%$ of high stress group were high, $2.3\%$ of very severe stress group were low. female group is $55.4\%$of high stress group were high, $1.4\%$ of very severe stress group were low. 12. Heed status, $28.0\%$ of economic support were high, $4.6\%$ of speech therapy, brace group were low. Need status by Sender, male group is $2i2\%$ of economic support group were high, $4.5\%$ of bracegroup were low. female group is$27.7\%$ of economic support group were high, $3.4\%$ of speech therapy group were low. 13. Care satisfaction comparision, 3.09, 0.55 point of IBR, 4.01, 0.45 point of CHR(P<0.001). 14. The variables which had positive correlation with IBR were gender(r=0.1406, P<0.01), age(r=0.1872, p<0.001), economic level(r=0.1246, P<0.05), disabled record(r=0.1137, P<0.05), education level(r=-0.1122. p<0.05). 15. The variables which had positive : correlation with CBR were gender(r=0.1613, P<0.01), age(r=0.2255, P<0.001). list of family(r=0.12i3, P<0.01), disabled record(r=0.1273, P<0.05). education level(r=-0.1294, P<0.01).

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.8
• /
• pp.89-98
• /
• 2010

Recently supply of Internet is bringing a important change in medical environments. The hospitals which had a different system is required the system that can efficiently share and exchange medical information. In order to transmission medical information between systems, the Health Level Seven(HL7) interface engine development that can convert medical data to HL7 messages is necessary. The HL7 is a standard protocol for data exchange in healthcare environments. In this paper, I implemented HL7 interface engine for Alzheimer's disease in elderly care facility. The interface engine is composed of the client system and the server system. The client system inputs user's medical care data for the aged, and builds them into HL7 message stream. HL7 messages in the client system transmitted over TCP/IP protocol to the server system. The server system parses and validates this messages stream to the segments and fields and then transmits acknowledgement to the client system. I implemented it using the Java and JavaCC. The study of interface engine implementation can be used meaningfully in electronic health record, telemedicine system, and medical information sharing among various healthcare institutions.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.11 no.4
• /
• pp.25-31
• /
• 2015

Smart healthcare environments which merge medical and IT technology are getting ready for the third generation centering EHR from current second generation. As a basic technology for the introduction and activation of EHR systems it requires heterogeneous network interworking techniques between various wired and wireless medical devices. Interworking technology for heterogeneous network among various medical devices is needed to introduce EHR system. The heterogeneous network interworking technology is needed for construction of a reliable data system to convert each of unstructured data into structured data. Therefore, in this paper, we identify the domestic and international trends of smart medical field and analyze the characteristics of wired and wireless communication technology that is used in a heterogeneous network. and also suggest requirements needed for interworking technology and provide interworking technology based on them. we expect that proposed method which is designed for smart healthcare environments would provide a basic architecture needed for third smart medical technology generation.

• Journal of Korean Society of Archives and Records Management
• /
• v.13 no.3
• /
• pp.151-171
• /
• 2013

To comprehend the importance and necessity of record management metadata standard implemented in an electronic medical records system, a survey was undertaken to 50 medical records managers in charge of 5 major hospitals in Seoul. Analysis of the survey results was performed by averaging the responses given by those who answered the survey. SPSS was utilized for statistical analysis. Managers of medical records placed importance on metadata that are related to security of records, such as "levels of security", "types of access to medical records", "levels of authorization granted to personnel", and "users accessing medical records". It shows that these managers need the functions of privacy protection in ERMS. Metadata on "external disclosure" had the lowest level but those surveyed with more than 7 years of experience placed greater importance in this area more those surveyed with less than 7 years of experience in a hospital. This shows that managers need the functions of external disclosure to meet the needs of third partiesfor medical research and medical education.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.520-522
• /
• 2022

의료 인공지능은 특정 진단에서 높은 정확도를 보이지만 모델의 신뢰성 문제로 인해 활발하게 쓰이지 못하고 있다. 이에 따라 인공지능 모델의 진단에 대한 원인 설명의 필요성이 대두되었고 설명가능한 의료 인공지능에 관한 연구가 활발히 진행되고 있다. 하지만 MRI 등 의료 영상 인공지능 분야에서 주로 진행되고 있으며, 이미지 형태가 아닌 전자의무기록 데이터 (Electronic Health Record, EHR) 를 기반으로 한 모델의 설명가능성 연구는 EHR 데이터 자체의 복잡성 때문에 활발하게 진행 되지 않고 있다. 본 논문에서는 전자의무기록 데이터인 MIMIC-III (Medical Information Mart for Intensive Care) 를 전처리 및 그래프로 표현하고, GCT (Graph Convolutional Transformer) 모델을 학습시켰다. 학습 후, 어텐션 흐름 그래프를 시각화해서 모델의 예측에 대한 직관적인 설명을 제공한다.