• Journal of the Korea Society of Computer and Information
• /
• v.28 no.11
• /
• pp.53-63
• /
• 2023

This study developed a cognitive impairment predictive model as one of the screening tests for preventing dementia in the elderly by using Automated Machine Learning(AutoML). We used 'Wearable lifelog data for high-risk dementia patients' of National Information Society Agency, then conducted using PyCaret 3.0.0 in the Google Colaboratory environment. This study analysis steps are as follows; first, selecting five models demonstrating excellent classification performance for the model development and lifelog data analysis. Next, using ensemble learning to integrate these models and assess their performance. It was found that Voting Classifier, Gradient Boosting Classifier, Extreme Gradient Boosting, Light Gradient Boosting Machine, Extra Trees Classifier, and Random Forest Classifier model showed high predictive performance in that order. This study findings, furthermore, emphasized on the the crucial importance of 'Average respiration per minute during sleep' and 'Average heart rate per minute during sleep' as the most critical feature variables for accurate predictions. Finally, these study results suggest that consideration of the possibility of using machine learning and lifelog as a means to more effectively manage and prevent cognitive impairment in the elderly.

• Journal of Chest Surgery
• /
• v.43 no.6
• /
• pp.619-626
• /
• 2010

Background: The indications and the optimal time of surgery of infective endocarditis are controversial. We report the surgical results of our hospital during the last 10 years with literature review. Material and Method: Between January 2000 and December 2009, we enrolled 23 infective endocarditis patients who underwent surgery, and analyzed retrospectively. In the preoperative blood culture, 8 cases (34.8%) were positive. The average preoperative antibiotics treatment period was $20.78{\pm}16.00$ days. There were 12 (52.2%) urgent operations. The average follow up period was $49.26{\pm}33.21$ months. Result: 20 mechanical valve replacements were performed, 9 in aortic position, 8 in mitral position and 3 in the both positions. The other procedures were one mitral valvuloplasty, one infected myxoma extirpation, and one infected pacemaker lead removal with debridement. The average period of postoperative intravenous antibiotic treatment was $24.39{\pm}15.98$ days. There were 5 complications, including 2 cases of postoperative bleeding, one postcardiotomy syndrome, one cerebral ischemia, and a low cardiac output syndrome. There were statistically significant postoperative improvement in NYHA class, left ventricle end diastolic/end systolic volume, and left atrium size (p-value < 0.05). Conclusion: We could obtain the satisfactory results without any moftalities by using sufficient preoperative antibiotics in hemodynamically stable patients, and by prompt surgery in unstable patients.

• Journal of Korean Society of Forest Science
• /
• v.104 no.2
• /
• pp.277-284
• /
• 2015

This study aims to evaluate the impact of forest therapy on neuro-cognitive, psychosocial, and physiological aspect of adolescent internet addiction risk group. We have classified potential and high risk user group as internet addiction risk group according to the criteria of Korean Internet Addiction Proneness Scale(K Scale). Based on the results of k-scale from the adolescents in metropolitan area from May to July 2013, 25 people were selected as Internet addiction risk group. We have randomized 13 participants joining forest therapy camp and 12 participants not joining one, and analyzed the change of the two groups with Continuous Performance, Kimberly S. Young, Connor-Davidson Resilience, Relationship Change Scale, heart rate variability and cortisol. Statistically significant changes were observedd in neuro-cognitive, psychosocial, and physiological variables, Through this study, we can consider that the therapy healing may relieve the level of internet addiction and can be an alternative to control emotional stability and impulsive behavior.

• Journal of Chest Surgery
• /
• v.31 no.3
• /
• pp.233-241
• /
• 1998

To define the risk factors affecting the early major morbidity and mortality after multiple- valve operations, the preoperative, intraoperative and postoperative informations were retrospectively collected on 124 consecutive patients undergoing a multiple-valve operation between October 1985 and July 1996 at the department of Thoracic and Cardiovascular Surgery of Pusan Paik Hospital. The study population consists of 53 men and 71 women whose mean age was 37.9$\pm$11.5(mean$\pm$SD) years. Using the New York Heart Association(NYHA) classification, 41 patients(33.1%) were in functional class II, 60(48.4%) in class III, and 20(16.1%) in class IV preoperatively. Seven patients(5.6%) had undergone previous cardiac operations. Atrial fibrillations were present in 76 patients(61.3%), a history of cerebral embolism in 5(4.0%), and left atrial thrombus in 13(10.5%). The overall early mortality rate and postoperative morbidity was 8.1% and 21.8% respectively. Among the 124 cases of multiple-valve operation, there were 57(46.0%) of combined mitral valve replacement(MVR) and aortic valve replacement(AVR), 48(38.7%) of combined MVR and tricuspid annuloplasty(TVA), 12(9.7%) of combined MVR, AVR and TVA, 3(2.4%) of combined MVR and aortic valvuloplasty, 2(1.6%) of combined MVR and tricuspid valve replacement, and others. The patients were classified according to the postoperative outcomes; Group A(27 cases) included the patients who had early death or major morbidity such as low cardiac output syndrome, mediastinitis, cardiac rupture, ventricular arrhythmia, sepsis, and others; Group B(97 cases) included the patients who had the good postoperative outcomes. The patients were also classified into group of early death and survivor. In comparison of group A and group B, there were significant differences in aortic cross-clamping time(ACT, group A:153.4$\pm$42.4 minutes, group B:134.0$\pm$43.7 minutes, p=0.042), total bypass time(TBT, group A:187.4$\pm$65.5 minutes, group B:158.1$\pm$50.6 minutes, p=0.038), and NYHA functional class(I:33.3%, II:9.7%, III:20%, IV:50%, p=0.004). In comparison of early death(n=10) and survivor(n=114), there were significant differences in age(early death:45.2$\pm$8.7 years, survivor:37.2$\pm$11.6 years, p=0.036), sex(female:12.7%, male:1.9%, p=0.043), ACT(early death:167.1$\pm$38.4 minutes, survivor:135.7$\pm$43.7 minutes, p=0.030), and NYHA functional class(I:0%, II:4.9%, III:1.7%, IV:35%, p=0.001). In conclusion, the early major morbidity and mortality were influenced by the preoperative clinical status and therefore the earlier surgical intervention should be recommended whenever possible. Also, improved methods of myocardial protection and operative techniques may reduce the risk in patients with multiple-valve operation.

• Journal of Chest Surgery
• /
• v.30 no.2
• /
• pp.137-145
• /
• 1997

From July 1983 to December 1993, total 112 consecutive mitral valve replacements in 107 patients were performed in patient with mitral valvular abnormalities. To estimate the risk factor related to operative death, all patient's perioperative data were reviewed retrospectively. Except 20 patients received concomitant aortic valve replacement and 2 patients had incomplete data, 85 patients were included in this study. Mean age were $37.3\pm$ 13.1 years ranging from 13 to 72 years. Thirty-seven patients were male and fourty-eight patients were female. Mean follow-up durations were $51.1\pm33.8$ months ranging from 6 months t 11 years. Patients in this study showed improvement in mean NYHA functional clssification, from $3.02\pm0.73$ to 1 $78\pm0.55,$ and also in cardiothoracic ratio, from 0.61 $\pm0.09$ to $0.58\pm0.08$ at 6 months follow-up after operation. Operative complications were detected in 23 patients(27.1 %) and common postoperative complications were rhythm disturbance in 7 cases, pulmonary complications in 6 cases and low cardiac output syndrome in 6 cases. Early mortality was 10.6%(n=9) and the most common cause of death was a congestive heart failure due to low cardiac output syndrome. Main cause of our higher operative mortality than other study was that operative mortality in the initial period of our mitral surgery was high(5 operative deaths among 19 mitral valve replacement from July 1983 to December 1985). Actuarial survival was 80.8% at 5 years, 71.8% at 11 years including operative deaths. Actuarial freedom from anticoagulant-related bleeding was 85.3% at 5 years, 78.3% at 11 years. 95.1% at 5 years and 88.8% at 11 years among the patient in this study were free from thromboembolism, and 97.5% at 5 years and 75.1% at 11 years were free from reoperation. Preoperative cardiothoracic ratio and patient's age were statistically significant operative risk factors.

• Tuberculosis and Respiratory Diseases
• /
• v.53 no.6
• /
• pp.612-618
• /
• 2002

Background : Bronchoscopy is an important diagnostic and a therapeutic tool in chest medicine. However, most patients feel that a bronchoscopy is an unpleasant procedure, and it is important to sedate the patients appropriately, particularly where repetitive examinations are required. Midazolam is a sedative drug with amnestic qualities and a rapid 2 hour half-life. This study have attempted to determine the safety, appropriate dosage, and the effect of midazolam premedication in patients who underwent a bronchoscopy. Methods : One hundred and eighty consecutive patients undergoing bronchoscopy were enrolled in this study. The patients received a midzolam doses of 0.03 mg/kg, 0.06 mg/kg, or a placebo. An additional dose of lidocaine, the total number of coughs, and the duration of the procedures were recorded with monitoring the the blood pressure, heart rate, and oxygen saturation. The level of satisfaction was assessed by the patient, bronchoscopist, and the nurse. Results : The blood pressure, pulse rates, oxygen saturation, number of coughs, lidocaine dose, and procedure time in the 3 groups were similar. There was a trend for the midazolam 0.03 mg/kg group to satisfy bronchoscopists more than the other two groups. The nurses' acceptability was lower in the midazolam 0.06 mg/kg group than the other groups. The patients' acceptablity was greater in both the midazolam 0.03 mg/kg and 0.06 mg/kg groups than in the control group. Conclusion : Sedation with low doses of intravenous midazolam is a safe technique for fiberoptic bronchoscopy with a low morbidity and high acceptable to patients and bronchoscopists.

• Journal of Veterinary Clinics
• /
• v.26 no.6
• /
• pp.539-546
• /
• 2009

The effect of hemorrhage on the electroencephalogram(EEG) was investigated in fifteen mixed-breed dogs anesthetized with ketamine, propofol and isoflurane. Animals were randomly allocated to three groups (n = 5) by anesthetic agents; group 1 (ketamine 5 mg/kg, IV), group 2 (propofol $156\;{\mu}g$/kg/min, IV) and group 3 (isoflurane 2.0% end-tidal concentration). Medetomidine ($40\;{\mu}g$/kg, IM) was used in all dogs as a preanesthetic agent. Recording electrode for EEG was positioned at CZ. EEG, heart rate, systolic/diastolic blood pressure, $pCO_2$, $pO_2$ and blood pH were measured before anesthesia, after anesthesia and after every bleedings. Three bleedings were accomplished by drawing blood through the femoral artery catheter at a rate of 7 ml/kg (10% of total blood volume) for 10 minutes. In the course of hemorrhage, a systolic/diastolic pressure continuously decreased in all groups. The $pCO_2$ values and heart rates were increased in all groups. The $pO_2$ values were most significantly increased in group 1 compared with those in other groups. The pH values were not significantly changed. On statistical analysis of EEG, there was no significant changes in group 1 and 3. But in group 2, band 3, 4 and 7 were significantly altered after 2nd and 3rd bleeding. Power alterations of band 3, 4 and 7 were thought to be related with hemorrhage over 20% of total blood volume in group 2. In conclusion, the regulation of infusion rate would be considered when a dog, anesthetized with propofol, bleed over 20% of total blood volume.

• Journal of Korean Society of Forest Science
• /
• v.103 no.1
• /
• pp.129-136
• /
• 2014

This study aims to investigate the effect of frequency of using forest environment on workers' stress. Workers in medical and counseling industry, which is recently known to require heavy emotional burden and thus cause a lot of stress and burnout syndrome, participated in the study. We classified 370 subjects, who were working at medical and counseling service institutions in Seoul metropolitan area, into two groups according to frequency of using forest environment by conducting a screening survey. 69 subjects submitted written consent and were recruited for either forest therapy program or control-group test; the "high" group (n=27) at the high frequency of using forest environment and the "low" group (n=42) at the low level. We measured the level of stress by using psychosocial indicators such as Worker's Stress Response Inventory (WSRI), Recovery Experience Questionnaire (REQ), and Maslach Burnout Inventory-General Survey (MBI-GS) and biological ones such as heart rate variability (HRV), cortisol, and Natural Killer cell (NK cell). The results suggested that the level of stress differed by frequency. The "high" group showed significantly low scores in the depression and work sub-scale and the total score of WSRI, emotional exhaustion, professional efficacy and total score of MBI-GS, and high scores in the total score of REQ. Similar tendencies were observed in the most of other indicators of psychosocial measures. Regarding the biological indicators, the "high" group showed the highest SDNN, RMSSD and TP measures of HRV and NK Cell activity and the low cortisol, although the statistical power did not reach the significant level. Our results suggest that the subjects who use forest environment frequently show favorable stress level both psychologically and biologically.

• Journal of Chest Surgery
• /
• v.34 no.6
• /
• pp.454-464
• /
• 2001

Background: It has been recognized that systemic inflammatory reaction and oxygen free radical formed by activated leukocyte in the procedure of cardiopulmonary bypass(CPB) frequently produce postoperative cardiac and pulmonary dysfunction. The purpose of this study was to evaluate the efficacy of leukocyte-depleting filters in the cardiopulmonary bypass circuit for patients undergoing open heart surgery(OHS). Material and method: The study involved 15 patients who underwent OHS with a Leukoguard-6 leukocyte filter placed in the arterial limbs of the bypass circuit(filter group, n=15) and 15 patients who did not have the filter(control group, n=15). We analyzed the differences between the groups in intraoperative changes of peripheral blood leukocyte and platelet counts, pre- and postbypass changes of malondialdehyde(MDA), troponin-T(TnT), 5'-nucleotidase(5'-NT) in coronary sinus blood, spontaneous recovery rate of heart beat after CPB, pre-and postoperative cardiac index(Cl) and pulmonary vascular resistance(PVR), and the amounts of postoperative bleeding and sternal wound complication. Result: During CPB, total leukocyte count of the filter group(9,567$\pm$ 842/㎣) was significantly less than that of the control group(13,573+1,167/㎣) (p<0.01), but there was no significant difference in platelet count between the groups. Postoperative levels of MDA(3.78+0.32 $\mu$mol/L vs 5.86+0.65 $\mu$mo1/L, p<0.01), TnT(0.40$\pm$0.04 ng/mL vs 0.59$\pm$0.08 ng/mL, p<0.05) and 5'-NT(3.88$\pm$0.61 U/L vs 5.80$\pm$0.90 U/L, p<0.05) were all significantly lower in the filter group than the control group. Postoperative Cl was higher in the filter group than the control group(3.26$\pm$0.18 L/$m^2$min vs 2.75$\pm$0.17 L/$m^2$/min, p=0.05). PVR of the filter group was lower than that of the control group(65.87$\pm$7.59 dyne/sec/cm$^{5}$ vs 110.80+12.22 dyne/sec/cm$^{5}$ , p<0.01). Spontaneous recovery rate of heart beat in the filter group was higher than that in the control group(12 patients vs 8 patients, p<0.05). Postoperative wound infection occurred in one case in the filter group and 4 case in the control group(p<0.05). Postoperative 24 hour blood loss of the filter group was more than that of the control group (614$\pm$107 mL vs 380+71 mL, p=0.05).

• Journal of Chest Surgery
• /
• v.29 no.6
• /
• pp.606-613
• /
• 1996

Cardiac transplantation has been the treatment of patients with end-stage heart disease since it was first performed in 1967. In Korea the first case was performed in 1992 and 42 patients underwent heart trans- plantation so far. The purpose of this article is to report short-term result of cardiac transplantation at our center. Between April 1994 and September 1995, 14 patients had undergone orthotopic heart transplantations. There was 12 male and 2 female patients. Mea recipient age was 34 years(range 11 to 54 years) and mean donor age was 28.4 years(16 to 50 years). Mean graft ischemic time was 120.7minutes(80 to 280 minutes). The follow-up period after transplantation was 11 months(3 to 17 months). Recipient diagnosis included dilated cardiomyopathy in 10, ischemic cardiomyopathy in 2, valvular cardiomyopathy in 1, congenital complex heart disease in 1 patient. The preoperative status of the recipients were state I (50%) and ll (50%) by UNOS classification and class 111 (5 patients) and class IV (9) by NYHA functional class. All patients were treated with triple-drug immunosuppression (cyclosporine, azathioprine, steroid) and induction with RATG. The rejection episodes were 5 times in 3 patients during the follow-up. Causes of infection were aspergillosis (2), and hepes zoster (1), CMV pneumonitis (1). Permanent pace- maker was inserted in 1 patient. Currently 9 patients are alive with seven patients in WYHA functional class I and two in class l . The ejection fraction increased from preoperative value of 19.9 $\pm$ 3.4% to postoperative value of 69.0 $\pm$ 5.6%. The causes of death were cellular rejection (1),chronic graft failure due to size-mismatching (1),respirat- oxy insufficiency due to asthma attack (1), subarachnoid hemorrhage (1), and RIO humoral rejection (1).