• Tuberculosis and Respiratory Diseases
• /
• v.63 no.1
• /
• pp.17-23
• /
• 2007

Background: Many diagnostic approaches for defining the definitive cause of pleurisy should be included due to the large variety of diseases resulting in pleural effusion. Although ADA is a useful diagnostic tool for making a differential diagnosis of pleural effusion, particularly for tuberculous pleural effusion, a definitive diagnostic cut-off value remains problematic in Korea. It was hypothesized that ADA multiplied by the Lymphocyte/Neutrophil ratio(L/N ratio) might be more powerful for making a differential diagnosis of pleural effusion. Methods: One hundred and ninety patients, who underwent thoracentesis and treatment in Chung-Ang University Hospital from January, 2005 through to February 2006, were evaluated. The clinical characteristics, radiologic data and the examination of the pleural effusion were analyzed retrospectively. Results: 1. Among the 190 patients, 59 patients (31.1%) were diagnosed with tuberculous pleurisy, 45 patients(23.7%) with parapneumonic effusion, 42 patients(22.1%) with malignant effusions, 36 patients(18.9%) with transudate, and 8 patients(4.2%) with empyema. One hundred and twenty one patients were found to have an ADA activity of 1 to 39 IU/L(63.7%). Twenty-nine were found to have an ADA activity of 40 to 75 IU/L(15.3%) and 40 were found to have an ADA activity of 75 IU/L or greater(21.0%). 2. Among the patients with tuberculous pleurisy, 5(8%), 18(30%) and 36 patients(60%) had an ADA activity ranging from 1 to 39 IU/L, 40 to 75 IU/L, and 75 IU/L or greater, respectively. In those with an ADA activitiy 40 to 75 IU/L, 18 patients(62%) had tuberculous pleurisy, 9(31%) had parapneumonic effusion and empyema, and 1(3.4%) had a malignant effusion. 3. In those with an ADA activity of 40 to 75 IU/L, there was no significant difference between tuberculous pleurisy and non-tuberculous pleural effusion(tuberculous pleurisy : 61.3 ${\pm}$ 9.2 IU/L, non-tuberculous pleural effusion : 53.3${\pm}$10.5 IU/L). 4. The mean L/N ratio of those with tuberculous pleurisy was 39.1 ${\pm}$ 44.6, which was significantly higher than nontuberculous pleural effusion patients (p<0.05). The mean ADA x L/N ratio of the tuberculous pleurisy patients was 2,445.7 ${\pm}$ 2,818.5, which was significantly higher than the non-tuberculous pleural effusion patients (level p<0.05). 5. ROC analysis showed that the ADA x L/N ratio had a higher diagnostic value than the ADA alone in the group with an ADA between 40-75 IU/L. Conclusion: The ADA multiplied by the lymphocyte-to-neutrophil ratio might provide a more definitive diagnosis of tuberculous pleurisy.

• The Korean Journal of Nuclear Medicine
• /
• v.37 no.4
• /
• pp.219-228
• /
• 2003

Purpose: The amount of salvaged myocardium is an important prognostic factor in patients with acute myocardial infarction (MI). We investigated if early Tl-201 SPECT imaging could be used to predict the salvaged myocardium and functional recovery in acute MI after primary PTCA. Materials and Methods: In 36 patients with first acute MI treated with primary PTCA, serial echocardiography and Tl-201 SPECT imaging ($5.8{\pm}2.1$ days after PTDA) were performed. Regional wall motion and perfusion were quantified with on 16-segment myocardial model with 5-point and 4-point scaling system, respectively. Results: Wall motion was improved in 78 of the 212 dyssynergic segments on 1 month follow-up echocardiography and 97 on 7 months follow-up echocardiography, which were proved to be salvaged myocardium. The areas under receiver operating characteristic curves of Tl-201 perfusion score for detecting salvaged myocardial segments were 0.79 for 1 month follow-up and 0.83 for 7 months follow-up. The sensitivity and specificity of Tl-201 redistribution images with optimum cutoff of 40% of peak thallium activity for detecting salvaged myocardium were 84.6% and 55.2% for 1 month follow-up, and 87.6% and 64.3% for 7 months follow-up, respectively. There was a linear relationship between the percentage of peak thallium activity on early redistribution imaging and the likelihood of segmental functional improvement 7 months after reperfusion. Conclusion: Tl-201 myocardial perfusion SPECT imaging performed early within 10 days after reperfusion can be used to predict the salvaged myocardium and functional recovery with high sensitivity during the 7 months following primary PTCA in patients with acute MI.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.141-154
• /
• 2019

Rapid growth of internet technology and social media is progressing. Data mining technology has evolved to enable unstructured document representations in a variety of applications. Sentiment analysis is an important technology that can distinguish poor or high-quality content through text data of products, and it has proliferated during text mining. Sentiment analysis mainly analyzes people's opinions in text data by assigning predefined data categories as positive and negative. This has been studied in various directions in terms of accuracy from simple rule-based to dictionary-based approaches using predefined labels. In fact, sentiment analysis is one of the most active researches in natural language processing and is widely studied in text mining. When real online reviews aren't available for others, it's not only easy to openly collect information, but it also affects your business. In marketing, real-world information from customers is gathered on websites, not surveys. Depending on whether the website's posts are positive or negative, the customer response is reflected in the sales and tries to identify the information. However, many reviews on a website are not always good, and difficult to identify. The earlier studies in this research area used the reviews data of the Amazon.com shopping mal, but the research data used in the recent studies uses the data for stock market trends, blogs, news articles, weather forecasts, IMDB, and facebook etc. However, the lack of accuracy is recognized because sentiment calculations are changed according to the subject, paragraph, sentiment lexicon direction, and sentence strength. This study aims to classify the polarity analysis of sentiment analysis into positive and negative categories and increase the prediction accuracy of the polarity analysis using the pretrained IMDB review data set. First, the text classification algorithm related to sentiment analysis adopts the popular machine learning algorithms such as NB (naive bayes), SVM (support vector machines), XGboost, RF (random forests), and Gradient Boost as comparative models. Second, deep learning has demonstrated discriminative features that can extract complex features of data. Representative algorithms are CNN (convolution neural networks), RNN (recurrent neural networks), LSTM (long-short term memory). CNN can be used similarly to BoW when processing a sentence in vector format, but does not consider sequential data attributes. RNN can handle well in order because it takes into account the time information of the data, but there is a long-term dependency on memory. To solve the problem of long-term dependence, LSTM is used. For the comparison, CNN and LSTM were chosen as simple deep learning models. In addition to classical machine learning algorithms, CNN, LSTM, and the integrated models were analyzed. Although there are many parameters for the algorithms, we examined the relationship between numerical value and precision to find the optimal combination. And, we tried to figure out how the models work well for sentiment analysis and how these models work. This study proposes integrated CNN and LSTM algorithms to extract the positive and negative features of text analysis. The reasons for mixing these two algorithms are as follows. CNN can extract features for the classification automatically by applying convolution layer and massively parallel processing. LSTM is not capable of highly parallel processing. Like faucets, the LSTM has input, output, and forget gates that can be moved and controlled at a desired time. These gates have the advantage of placing memory blocks on hidden nodes. The memory block of the LSTM may not store all the data, but it can solve the CNN's long-term dependency problem. Furthermore, when LSTM is used in CNN's pooling layer, it has an end-to-end structure, so that spatial and temporal features can be designed simultaneously. In combination with CNN-LSTM, 90.33% accuracy was measured. This is slower than CNN, but faster than LSTM. The presented model was more accurate than other models. In addition, each word embedding layer can be improved when training the kernel step by step. CNN-LSTM can improve the weakness of each model, and there is an advantage of improving the learning by layer using the end-to-end structure of LSTM. Based on these reasons, this study tries to enhance the classification accuracy of movie reviews using the integrated CNN-LSTM model.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• v.5 no.1
• /
• pp.51-61
• /
• 2002

Purpose: The most common causes of neonatal cholestasis are neonatal hepatitis (NH) and extrahepatic biliary atresia (EHBA). Since neonatal cholestasis presents with variable expression of same pathologic process and has similar clinical, biochemical, and histologic features between EHBA and idiopathic neonatal hepatitis (NH), differential diagnosis is often difficult. We reviewed the differences of clinical characteristics and laboratory data to find out any correlation between the results of $Tc^{99m}$ DISIDA scan and presence of acholic stool. Methods: Between June 1993 and January 2001, total 29 infants younger than 4 month-old underwent $Tc^{99m}$ DISIDA scan. Their biochemical tests and clinical course were reviewed retrospectively. Results: Patients who had negative intestinal activity on $Tc^{99m}$ DISIDA scan showed acholic stool and revealed higher serum direct bilirubin and urine bilirubin level. 18.2% of patients with acholic stool showed intestinal activity on $Tc^{99m}$ DISIDA scan and 81.8% of them did not. All the patients without acholic stool showed positive intestinal activity on $Tc^{99m}$ DISIDA scan. The result of $Tc^{99m}$ DISIDA scan and the presence of acholic stool showed high negative correlation (r :-0.858). Patients with acholic stool and negative intestinal activity on $Tc^{99m}$ DISIDA scan showed higher serum total bilirubin level. Patients without acholic stool and positive intestinal activity on $Tc^{99m}$ DISIDA scan showed higher serum level of ALT. Conclusion: Patients with acholic stool and negative intestinal activity showed high correlation, but 18.2% of patients with acholic stool showed positive intestinal activity. So operative cholangiogram or transcutaneous liver biopsy should be performed for confirmation.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.4
• /
• pp.301-306
• /
• 2008

Purpose: Recently, a number of patients needed total thyroidectomy and high dose radioiodine therapy (HD-RAI) get increased more. The aim of this study is to evaluate whether pathological staging (PS) and serum thyroglobulin (sTG) level could replace the diagnostic I-123 scan for the determination of therapeutic dose of HD-RAI in patients with differentiated thyroid cancer. Materials and Methods: Fifty eight patients (M:F=13;45, age $44.5{\pm}11.5\;yrs$) who underwent total thyroidectomy and central or regional lymph node dissection due to differentiated thyroid cancer were enrolled. Diagnostic scan of I-123 and sTG assay were also performed on off state of thyroid hormone. The therapeutic doses of I-131 (TD) were determined by the extent of uptakes on diagnostic I-123 scan as a gold standard. PS was graded by the criteria recommended in 6th edition of AJCC cancer staging manual except consideration of age. For comparison of the determination of therapeutic doses, PS and sTG were compared with the results of I-123 scan. Results: All patients were underwent HD-RAI. Among them, five patients (8.6%) were treated with 100 mCi of I-131, fourty three (74.1%) with 150 mCi, six (10.3%) with 180 mCi, three (5.2%) with 200 mCi, and one (1.7%) with 250 mCi, respectively. On the assessment of PS, average TDs were $154{\pm}25\;mCi$ in stage I (n=9), $175{\pm}50\;mCi$ in stage II (n=4), $149{\pm}21\;mCi$ in stage III (n=38), and $161{\pm}20\;mCi$ in stage IV (n=7). The statistical significance was not shown between PS and TD (p=0.169). Among fifty two patients who had available sTG, 25 patients (48.1%) having below 2 ng/mL of sTG were treated with $149{\pm}26\;mCi$ of I-131, 9 patients (17.3%) having $2{\leq}\;sTG\;<5\;ng/mL$ with $156{\pm}17\;mCi$, 5 patients (9.6%) having $5{\leq}\;sTG\;<10\;ng/mL$ with $156{\pm}13\;mCi$, 7 patients (13.5%) having $10{\leq}sTG\;<50\;ng/mL$ with $147{\pm}24\;mCi$, and 6 patients (11.5%) having above 50 ng/mL with $175{\pm}42\;mCi$. The statistical significance between sTG level and TD (p=0.252) was not shown. Conclusion: In conclusion, PS and sTG could not replace the determination of TD using I-123 scan for first HD-RAI in patients with differentiated thyroid cancer.

• Korean Journal of Animal Reproduction
• /
• v.24 no.1
• /
• pp.1-18
• /
• 2000

Eighty dairy farms in Palmers ton North area in New Zealand were surveyed on 1) general characteristics (10 Questions), 2) milk yield and feed supplementary (7 questions), 3) reproductive efficiencies (12 questions) and 4) reproductive disorders (12 questions) by mail questions from February to July, 1998. Among those 4 items from 38 dairy farms (47.5%), especially in items 1) and 2), overall dairy farming situation, supplementary feeding and milk yields were surveyed and analyzed for Korean dairy farmers (especially in Cheju island) to have better understanding or higher economical gains. The results were as follows. 1. In dairy experience, 21 (45%) among 38 dairy farms surveyed were answered that farming less than 15 years, 15~19 year, 20~25 years and over 26 years experience were 3 (7.9%), 7 (18.4%), 6 (15.8%) and 5 (13.2%) which generally showed longer experience compare to Korean dairy farming situation. In survey of labour input and business goal of dairy farming, self-managing farms, sharemilkers, unpaid family manpowering farms, manager running farms, farms with hired worker, farms with part time helper and other type was 21 (55.3%), 10 (26.3%), 2 (3.5%), 3 (5.3%), 18 (31.6%), 2 (3.5%), and 1 (1.8%), respectively. 2. Analyzing pasture and tillable land, pasture according to feeding scale (200, 300 and 400 heads) were 56, 90 and 165.3 ha, and tillable lands were 51, 78 and 165 ha which showed some differences among feeding scale. In recording methods in 38 farms replied, 36 (95%) dairy handbook and 23 (70%) dual methods taking farms were higher than that of 10 (26.3%) computer and 15(39.5%) well-recorder methods. 3. Dairy waste processing facilities in environmental field were almost perfect except of metropolitan area, and so no problem was developed in its control so far. Hence, 26 farm (68.4%) of pond system was higher rather than those in 8 (21.2%) of using as organic manure after storing feces of dairy cattle, 1(2.6%) bunker system and 3 (7.9%) other type farms. 4. In milking facilities, 33 farms (86.9%) of Harringbone types were higher than those in 3 (7.9%) of Walkthrough types, 1 (2.6%) of Rotary system and other types. Although the construction facilities was not enough, this system show the world-leveled dairy country to attempted to elevate economic gains using the advantage of climatic condition. 5. In milking day and yearly yield per head, average 275 milking days and 87 drying days were longer than that of 228 average milking days in New Zealand. Annual total milk yield per head and milk solid (ms) was 3,990 kg and approximately 319 kg. Dairy milk solid (ms) per head, milk yield, fat percentage was 1.2 kg, 15.5 kg and average 4.83% which was much higher than in other country, and milk protein was average 3.75%. 6. In coclusion, Palmerstone North has been a center of dairy farming in New Zealand for the last 21 years. Their dairy farming history is 6~9 year longer than ours and the average number of milking cows per farm is 355, which is much greater than that (35) of Korea. They do not have dairy barn, but only milking parlors. Cows are taken care of by family 0.5 persons), are on a planned calving schedule in spring (93%) and milked for 240~280 days a year, avoiding winter. Cows are dried according to milk yield and body condition score. This management system is quite different from that of Korean dairy farms. Cows are not fed concentrates, relying entirely on pasture forages and the average milk yield per cow is 3,500 kg, which is about 1/2 milk yield of Korean dairy farms. They were bred to produce high fat milk with an average of 4.5%. Their milk production cost is the lowest in the world and the country's economy relies heavily on milk production. We Korean farmers may try to increase farming size, decreasing labor and management costs.