Data mining differs primarily from traditional data analysis on an important dimension, namely the scale of the data. That is the reason why not only statistical but also computer science principles are needed to extract information from large data sets. In this paper we briefly review data mining, its characteristics, typical data mining algorithms, and potential and ongoing applications of data mining at biopharmaceutical industries. The distinguishing characteristics of data mining lie in its understandability, scalability, its problem driven nature, and its analysis of retrospective or observational data in contrast to experimentally designed data. At a high level one can identify three types of problems for which data mining is useful: description, prediction and search. Brief review of data mining algorithms include decision trees and rules, nonlinear classification methods, memory-based methods, model-based clustering, and graphical dependency models. Application areas covered are discovery compound libraries, clinical trial and disease management data, genomics and proteomics, structural databases for candidate drug compounds, and other applications of pharmaceutical relevance.
Classification ; Data Mining* ; Dataset ; Decision Trees ; Disease Management ; Drug Discovery* ; Genomics ; Proteomics ; Retrospective Studies ; Statistics as Topic

Cortical Dysplasia(CD) is increasingly recognized as a cause of intractable epilepsy since it can be easily diagnosed with high resolution magnetic resonance imaging(MRI). We analyzed 24 consecutive cases with cortical dysplasia who underwent resective surgery at our institute between September, 1992 and December, 1995. MRI was demonstrated to be a decisive tool to identify CD in 16 cases(66.7%). However, the remaining 8 cases(33.3%) were confirmed to have CD based on histological examination. Location of the lesion included temporal(n=9), central(n=5), multilobar(n=5), and frontal(n=5). Intracranial EEG recording was performed to delineate the epileptogenic zone in 16 cases:subdural grid(10 cases) and depth electrode(6 cases). Resection was performed in temporal(n=9), frontal(n=7), central(n=6), parietal(n=1), and multilobar(n=1). Complete resection was possible in 15(62.5%) and partial resection in 9(37.5%). Histological examination revealed dyslamination of cortical layers only(n=9), additional dysplastic neurons(n=7), and additional balloon cells(n=8). Surgical outcome was graded as seizure free(n=14:58.3%), rare seizures(n=3:12.5%), <50% reduction of seizure frequency(n=2:8%), and no change(n=5:20.8%) after mena follow-up of 12.7 months. The most influencing factor on the surgical outcome was the degree of completeness of resection(p<0.05). However, detectability of CD on MRI, severity of histological findings, age of seizure onset, and duration of seizure were not significantly related with the surgical outcome. High suspicion is required to detect the CD among the patients with intractable epilepsy and surgical strategies should be well-planned to improve the surgical outcome in the patients with CD.
Electroencephalography ; Epilepsy ; Follow-Up Studies ; Humans ; Magnetic Resonance Imaging ; Malformations of Cortical Development* ; Seizures

With rapid population aging, the socioeconomic burden caused by dementia care is snowballing. Although a few community-based studies of Alzheimer's disease (AD) have been performed in Korea, there has never been a nationwide hospital-based study thereof. We aimed to identify the demographics and clinical characteristics of mild-to-moderate AD patients from the Clinical Research Center for Dementia of Korea (CREDOS) registry. A total of 1,786 patients were consecutively included from September 2005 to June 2010. Each patient underwent comprehensive neurological examination, interview for caregivers, laboratory investigations, neuropsychological tests, and brain MRI. The mean age was 74.0 yr and the female percentage 67.0%. The mean period of education was 7.1 yr and the frequency of early-onset AD (< 65 yr old) was 18.8%. Among the vascular risk factors, hypertension (48.9%) and diabetes mellitus (22.3%) were the most frequent. The mean score of the Korean version of Mini-Mental State Examination (K-MMSE) was 19.2 and the mean sum of box scores of Clinical Dementia Rating (CDR-SB) 5.1. Based on the well-structured, nationwide, and hospital-based registry, this study provides the unique clinical characteristics of AD and emphasizes the importance of vascular factors in AD in Korea.
Aged ; Aged, 80 and over ; Alzheimer Disease/complications/*diagnosis ; Brain/radionuclide imaging ; Caregivers ; Dementia/diagnosis ; Demography ; Diabetes Mellitus, Type 2/etiology ; Female ; Hospitals ; Humans ; Hypertension/etiology ; Interviews as Topic ; Magnetic Resonance Imaging ; Male ; Middle Aged ; Questionnaires ; *Registries ; Republic of Korea ; Risk Factors

BACKGROUND: Twin-to-twin transfusion syndrome (TTTS) is a serious complication of monochorionic twin pregnancies. It results from disproportionate blood supply to each fetus caused by abnormal vascular anastomosis within the placenta. Amniotic fluid (AF) is an indicator reflecting the various conditions of the fetus, and an imbalance in AF volume is essential for the antenatal diagnosis of TTTS by ultrasound. In this study, two different mass spectrometry quantitative approaches were performed to identify differentially expressed proteins (DEPs) within matched pairs of AF samples.METHODS: We characterized the AF proteome in pooled AF samples collected from donor and recipient twin pairs (n = 5 each) with TTTS by a global proteomics profiling approach and then preformed the statistical analysis to determine the DEPs between the two groups. Next, we carried out a targeted proteomic approach (multiple reaction monitoring) with DEPs to achieve high-confident TTTS-associated AF proteins.RESULTS: A total of 103 AF proteins that were significantly altered in their abundances between donor and recipient fetuses. The majority of upregulated proteins identified in the recipient twins (including carbonic anhydrase 1, fibrinogen alpha chain, aminopeptidase N, alpha-fetoprotein, fibrinogen gamma chain, and basement membrane-specific heparan sulfate proteoglycan core protein) have been associated with cardiac or dermatologic disease, which is often seen in recipient twins as a result of volume overload. In contrast, proteins significantly upregulated in AF collected from donor twins (including IgGFc-binding protein, apolipoprotein C-I, complement C1q subcomponent subunit B, apolipoprotein C-III, apolipoprotein A-II, decorin, alpha-2-macroglobulin, apolipoprotein A-I, and fibronectin) were those previously shown to be associated with inflammation, ischemic cardiovascular complications or renal disease.CONCLUSION: In this study, we identified proteomic biomarkers in AF collected from donor and recipient twins in pregnancies complicated by TTTS that appear to reflect underlying functional and pathophysiological challenges faced by each of the fetuses.

BACKGROUND: Twin-to-twin transfusion syndrome (TTTS) is a serious complication of monochorionic twin pregnancies. It results from disproportionate blood supply to each fetus caused by abnormal vascular anastomosis within the placenta. Amniotic fluid (AF) is an indicator reflecting the various conditions of the fetus, and an imbalance in AF volume is essential for the antenatal diagnosis of TTTS by ultrasound. In this study, two different mass spectrometry quantitative approaches were performed to identify differentially expressed proteins (DEPs) within matched pairs of AF samples. METHODS: We characterized the AF proteome in pooled AF samples collected from donor and recipient twin pairs (n = 5 each) with TTTS by a global proteomics profiling approach and then preformed the statistical analysis to determine the DEPs between the two groups. Next, we carried out a targeted proteomic approach (multiple reaction monitoring) with DEPs to achieve high-confident TTTS-associated AF proteins. RESULTS: A total of 103 AF proteins that were significantly altered in their abundances between donor and recipient fetuses. The majority of upregulated proteins identified in the recipient twins (including carbonic anhydrase 1, fibrinogen alpha chain, aminopeptidase N, alpha-fetoprotein, fibrinogen gamma chain, and basement membrane-specific heparan sulfate proteoglycan core protein) have been associated with cardiac or dermatologic disease, which is often seen in recipient twins as a result of volume overload. In contrast, proteins significantly upregulated in AF collected from donor twins (including IgGFc-binding protein, apolipoprotein C-I, complement C1q subcomponent subunit B, apolipoprotein C-III, apolipoprotein A-II, decorin, alpha-2-macroglobulin, apolipoprotein A-I, and fibronectin) were those previously shown to be associated with inflammation, ischemic cardiovascular complications or renal disease. CONCLUSION In this study, we identified proteomic biomarkers in AF collected from donor and recipient twins in pregnancies complicated by TTTS that appear to reflect underlying functional and pathophysiological challenges faced by each of the fetuses.

The definition of the high-risk group for gestational diabetes mellitus (GDM) defined by the American College of Obstetricians and Gynecologists was changed from the criteria composed of five historic/demographic factors (old criteria) to the criteria consisting of 11 factors (new criteria) in 2017. To compare the predictive performances between these two sets of criteria.