BACKGROUND: Since the human genome project was completed in 2003, there have been numerous reports on cancer and related markers. This study was aimed to develop a system to extract automatically information regarding the relationship between cancer and tumor markers from biomedical literatures. METHODS: Named entities of tumor markers were recognized by both a dictionary-based method and machine learning technology of the support vector machine. Named entities of cancers were recognized by the MeSH dictionary. RESULTS: Relational and filtering keywords were selected after annotating 160 abstracts from PubMed. Relational information was extracted only when one of the relational keywords was in an appropriate position along the parse tree of a sentence with both tumor marker and disease entities. The performance of the system developed in this study was evaluated with another set of 77 abstracts. With the relational and filtering keyword used in the system, precision was 94.38% and recall was 66.14%, while without the expert knowledge precision was 49.16% and recall was 69.29%. CONCLUSIONS: We developed a system that can extract relational information between a tumor and its markers by incorporating expert knowledge into the system. The system exploiting expert knowledge would serve as a reference when developing another information extraction system in various medical fields.
Abstracting and Indexing as Topic ; Algorithms ; Database Management Systems ; Humans ; *Medical Informatics Computing ; Neoplasms/metabolism ; Programming Languages ; *PubMed ; Software ; *Tumor Markers, Biological

BACKGROUND: Panel reactive antibody (PRA) is to screen and identify HLA antibody. Majority of antibody specificities in high-PRA are directed against cross reactive group (CREG). Thus, this study was to know the advantage of identifying CREG specificity and whether antibody specificities are changed according to CREG classification. METHODS: HLA class I antibodies were identified from 159 sera from 108 patients in Asan Medical Center, who had shown more than 5% PRA by anti-human globulin (AHG)-complement-dependent cytotoxicity (CDC). Tail analysis-based computer program was developed to identify specificities, applying both Rodey (R-ABC) and Takemoto (T-ABC) classification. The results were also compared with those obtained when without CREG application (ABC). RESULTS: Among 151 cases in which HLA specificities was identified, the frequency of CREG specificity was 22.5% in R-ABC and 27.2% in T-ABC. Eleven cases showed CREG specificities only in one classification. However, the individual antigen specificities in one hand were all included in the CREG identified in the other hand. CREG specificities in samples with PRA >50% (60%) were more frequently identified than those in samples with PRA < or =50% (9%) (in R-ABC, P<0.0001). Without applying CREG to interpretation, specificity was not identified in 9 cases. CONCLUSIONS: Application of CREG enhanced the rate of antibody identification. Antibody specificities of those cases where CREG specificities were different between Rodey and Takemoto classifications were almost the same when compared at the individual antigen level. Therefore, it was thought that it makes no difference to use any one of these two classifications in interpreting PRA.
Alleles ; Antibodies/blood ; *Antibody Specificity ; Cross Reactions ; HLA Antigens/genetics/*immunology ; Histocompatibility Antigens Class I/immunology ; Histocompatibility Testing ; Humans ; Kidney Transplantation ; Reproducibility of Results ; Retrospective Studies

BACKGROUND: Hepatitis B virus (HBV) DNA quantification is necessary for starting and monitoring of antiviral therapy in patients with chronic hepatitis B. This study was intended to assess the clinical performance of Abbott RealTime HBV Quantification kit (Abbott Laboratories, USA). METHODS: The performance was evaluated in terms of precision, linearity, detection sensitivity, cross-reactivity, and carry-over. A correlation with the Real-Q HBV Quantification kit (BioSewoom Inc., Korea) was also examined using serum samples from 64 patients diagnosed with chronic hepatitis B and underwent lamivudine therapy in Asan Medical Center. We verified the trueness of the system by comparing the outputs with the assigned values of the BBI panel (BBI Diagnostics, USA). RESULTS: Within-run and between-run coefficients of variation (CV) were 3.56-4.71% and 3.03-4.98%, respectively. Linearity was manifested ranging from 53 to 10(9) copies/mL and the detection sensitivity was verified to be 51 copies/mL. None of hepatitis C virus showed cross-reactivity. No cross-contamination occurred when negative and positive samples were alternatively placed in a row. It showed a good correlation with the Real-Q HBV (r2=0.9609) and the test results for the BBI panel were also well agreed to the assigned values (r2=0.9933). CONCLUSIONS: The performance of Abbott RealTime HBV Quantification kit was excellent; thus, it should be widely used in starting and monitoring of antiviral therapy in Korean patients with chronic hepatitis B.
Computer Systems ; DNA, Viral/*blood ; Hepatitis B virus/genetics/*isolation & purification ; Hepatitis B, Chronic/*virology ; Humans ; Polymerase Chain Reaction/*methods ; Reagent Kits, Diagnostic ; Viral Load/*methods

BACKGROUND: Human leukocyte antigen (HLA) typing based on polymerase chain reaction (PCR) is rapidly replacing the conventional serological method. This study was intended to evaluate Bio- SewoomTM HLA-A, -B, -C PCR/SSP kit (BioSewoom SSP) which had recently been developed in Korea. METHODS: A total of 158 samples from domestic (21) and international (137) HLA proficiency testing (PT) were genotyped with BioSewoom SSP, and its results were compared to consensus results. For comparison with INNO-LiPA HLA-A, -B, -C Typing Kit (INNO-LiPA, Innogenetics, Belgium), 20 samples of Koreans were genotyped with both kits for each HLA-A, -B, -C locus. RESULTS: Among the 21 samples of domestic PT, BioSewoom SSP showed ambiguities as follows: 4 samples (19.0%) in HLA-A, 6 (28.6%) in HLA-B, and 1 (4.8%) in HLA-C. The ambiguities could be resolved by considering the allele distribution of Koreans. Among the 137 samples from international PT, BioSewoom SSP also showed ambiguities as follows: 12 samples (8.8%) in HLA-A, 26 (19.0%) in HLA-B and 6 (4.4%) in HLA-C. Considering the allele distribution of Koreans, the serologic equivalents obtained from BioSewoom SSP showed a full agreement with those of INNO-LiPA in all the loci tested. Twelve (0.007%) among 1,760 PCR reactions from the 21 samples of domestic PT and 20 patient samples produced faint nonspecific bands, but it was negligible. PCR failure of internal control just barely occurred (15 PCR reactions, 0.009%), but it had no bearing on allele assignment. CONCLUSIONS: The performance of BioSewoom SSP was comparable to that of INNO-LiPA. All the ambiguities could be resolved by considering the allele distribution of Koreans. It is concluded that BioSewoom SSP has good performance to be used in routine HLA laboratories.
Alleles ; Genotype ; HLA-A Antigens/*genetics ; HLA-B Antigens/*genetics ; HLA-C Antigens/*genetics ; Histocompatibility Testing/*methods ; Humans ; Polymerase Chain Reaction/*methods ; Reagent Kits, Diagnostic

BACKGROUND: Panel reactive antibody (PRA) test is used to determine whether a patient awaiting transplantation is previously sensitized. Tail analysis algorithm is widely used to identify antibody specificities, but it is very difficult to perform manually. METHODS: To develop a web-based program, PHP (5.1.2), Apache (2.0.55), and MySQL (5.0.22) were used. Tail analysis algorithm was applied to identify specificities, which analyzed statistically 2 x 2 tables representing reactivities to broad antigens, splits and cross reactive groups (CREG). Exploiting two CREG classifications of Rodey (R) and Takemoto (T), antibody specificities were identified by 3 methods (ABC, R-ABC, T-ABC) simultaneously. Performance of the system was evaluated using 159 samples that showed > or =6 PRA% by a lymphocytotoxicity assay. RESULTS: A web-based system that can identify HLA antibody specificities was implemented on www.koreanhla.com. Among 159 samples tested, antibody specificities were identified in 151 (95.0 %), but not in 8 samples with PRA >97%. Among the 151 samples, 110 showed broad or split specificities and 41 CREG specificities. CONCLUSIONS: We developed a web-based computer program for the identification of HLA antibody specificities. Accessible to everyone on the internet, this program should be of help in sharing PRA results among laboratories.
Algorithms ; *Antibody Specificity/genetics ; HLA Antigens/genetics/*immunology ; Histocompatibility Testing ; Humans ; Internet ; *Software

BACKGROUND: Hepatitis B virus (HBV) DNA quantification is important for the management of HBV infection and identification of the development of resistance. The susceptibility to contamination and more variable reproducibility of results with the conventional HBV DNA quantification method have raised the need of a more simple and accurate method for HBV DNA quantification. Real-time quantitative PCR assays recently introduced in the laboratory can meet these needs. In this study, we evaluated the performance of the Real-Q HBV Quantification kit developed in Korea. METHODS: We evaluated the recovery of DNA extraction, the interference of internal control, an analytical sensitivity, specificity, and reproducibility, a clinical specificity, and a reportable range of the Real-Q HBV Quantification kit. The quantification result was also compared to that obtained by the Digene Hybrid-Capture II. RESULTS: The mean percent recovery was 108.6% and there was no interference with the internal control on DNA extraction. None of HIV, hepatitis C virus, or cytomegalovirus showed a cross-reactivity with HBV. This assay detected HBV DNA in a linear range from 10(2) to 10(10) copies/mL, with the detection limit of 56 copies/mL. The assay exhibited a low within-run CV (coefficient of variation) (8.7-11.9%), between-run CV (10.5-14.7%), and between-day CV (13.2-21.4%). No HBV DNA was detected in any of 100 samples without HBV, resulting in a clinical specificity of 100%. The levels of HBV DNA showed a good correlation with those determined with Digene Hybrid-Capture II (R2=0.9827). CONCLUSIONS: The Real-Q HBV Quantification kit showed a good analytical sensitivity, specificity, and high reliability with a broad reportable range. This assay should be clinically useful in managing patients with HBV infection.
Cytomegalovirus ; DNA* ; Hepacivirus ; Hepatitis B virus ; HIV ; Humans ; Korea ; Limit of Detection ; Polymerase Chain Reaction ; Real-Time Polymerase Chain Reaction* ; Reproducibility of Results ; Sensitivity and Specificity

PURPOSE: To assess the usefulness of cardiac MR imaging (MRI) in the diagnosis of acute myocardial infarction and in the assessment of myocardial viability in comparision with Tl-201 SPECT. MATERIALS AND METHODS: We retrospectively studied 17 patients who complained of chest pain and dyspnea with cardiac MRI . The patients were evaluated for the presence or absence of high signal intensity on T2-weighted image (T2WI), abnormal wall motion on 2D - FIESTA, perfusion defect on Gd-DTPA enhanced T1WI, and delayed myocardial enhancement on 15-minutes delay Gd-DTPA enhanced T1WI. The results were correlated with the images on Tl-201 SPECT, taken at rest and stress, through which reversibility of perfusion defect was assessed. RESULTS: Both cardiac MRI and Tl-201 SPECT proved to be useful methods for diagnosing acute myocardial infarction. In order of decreasing correspondence, T2WI, Tl-201 SPECT, delayed enhancement study, and wall motion images all showed significant statistical correlation with the clinical diagnosis of myocardial infarction. Perfusion MRI, on the other hand, showed no significant statistical difference was found between Tl-201 SPECT and cardiac MRI. The results on T2WI showed high accordance with those on Tl-201 SPECT, while delayed myocardial enhancement and wall motion studies showed no agreement with Tl-201 SPECT. CONCLUSION: Cardiac MRI is useful method for diagnosis of acute myocardiac infarction. With respect to the assessment of myocardial viability, the results obtained on cardiac MRI showed high agreement with those on Tl-201 SPECT. However, further study is necessary at this point for standardization and establishment of the methods for assessing myocardial viability on cardiac MRI.
Chest Pain ; Diagnosis ; Dyspnea ; Gadolinium DTPA ; Hand ; Humans ; Infarction ; Magnetic Resonance Imaging* ; Myocardial Infarction* ; Perfusion ; Retrospective Studies ; Tomography, Emission-Computed, Single-Photon*

PURPOSE: A sentinel lymph node mapping with blue dye has been well accepted as a common procedure in breast cancer surgery. However, it is well known that blue dye absorbed into the circulation may interfere pulse oximetery reading. The aim of this study was to evaluate the change of pulse oximetery reading by isosulfan blue dye injection during sentinel lymph node mapping. METHODS: Thirteen breast cancer patients with normal preoperative cardiopulmonary functions were studied. Four ml of isosulfan blue dye was injected subdermally when the patient became stable after induction of general anesthesia. The pulse oximetery was monitored continuously. Multiple arterial blood gas analyses (ABGA) were performed before dye injection and 10, 30, 40 minutes after dye injection. The results of oxygen saturturation by oximetery (SpO2) and the results of arterial oxygen tension (SaO2) and arterial oxygen saturation (SaO2) by ABGA were compared. RESULTS: The value of both SaO2 and PaO2 measured by ABGA has not been altered by isosulfan dye injection. However SpO2 decreased by isosulfan dye injection. SpO2 decrease started 8.2+/-1.5 (2~0) minutes after dye injection and returned to preinjection level by 85.7+/-5.6 (60~126) minutes after injection. The lowest vaule of SpO2 was 95.6+/-1.2% (93~97). Mean duration of SpO2 decrease was 77.5+/-6.2 (40~117) minutes. The duration of SpO2 decrease was longer in the aged patients, but it was not statistically significant (p=0.3). There was no siginificant difference in duration of SpO2 decrease according to injection site, operation method, and body mass index (BMI). CONCLUSION: .Isosulfan dye injection using for sentinel lymph node mapping causes no change in true ABGA results but causes a mild reversible decrease in SpO2, It is important to look for other causes when SpO2 decrease is significant and persistent.
Anesthesia, General ; Blood Gas Analysis ; Body Mass Index ; Breast Neoplasms ; Humans ; Lymph Nodes ; Oxygen ; Reading*

PURPOSE: To assess the value of positional shifting to a gravity-dependent area, as revealed by HRCT, in differ-entiating pulmonary edema (PE) from other conditions. MATERIALS AND METHODS: Sixteen consecutive patients in whom plain radiographs suggested the presence of pulmonary edema but the clinical findings were indefinite underwent HRCT of the lung. For initial scanning they were in the supine position, and then in the prone position. Findings of ground-glass opacity, interlobular septal thickening and peribronchovascular interstitial thickening were analyzed in terms of the presence and degree of shifting to a gravity-dependent area, a grade of high, intermediate or low being assigned. RESULTS: PE was diagnosed in 8 of 16 cases, the remainder being designated as non-pulmonary edema (NPE). Ground-glass opacity was observed in all 16, while the degree of positional shifting was found to be high in ten (PE:NPE=6:4), intermediate in four (PE:NPE=2:2), and low in two (PE:NPE=0:2). There was no significant difference between the two groups (p > 0.05). Interlobular septal thickening was observed in all but two NPE cases; the degree of shifting was high in six (PE:NPE=6:0), intermediate in one (PE), and low in seven (PE:NPE=1:6). Shifting was significantly more prominent in PE than in NPE cases (p<0.05). Peribronchovas-cular interstitial thickening was positive in all PE cases and one NPE case, with no positional shifting. CONCLUSION: Positional shifting of interlobular septal thickening to a gravity-dependent area, as demonstrated by HRCT, is the most specific indicator of pulmonary edema.
Edema ; Humans ; Lung ; Prone Position ; Pulmonary Edema* ; Supine Position

In the evaluation of vascular lesions, MR can be used to distinguish slow- from high-flow lesions on the basis of the observed spin-echo MR signal characteristics. MR imaging can also represent features of the static tissues of the vascular lesions that are composed of fibrofatty components, as well as thromboses, phleboliths and muscle atrophy. This paper illustrates the MR findings of various vascular lesions, correlating them with the pathologic specimen and emphasizing on the static tissues.
Magnetic Resonance Imaging* ; Muscular Atrophy ; Thrombosis