BackgroundIn the application of microarray data, how to select a small number of informative genes from thousands of genes that may contribute to the occurrence of cancers is an important issue. Many researchers use various computational intelligence methods to analyzed gene expression data.ResultsTo achieve efficient gene selection from thousands of candidate genes that can contribute in identifying cancers, this study aims at developing a novel method utilizing particle swarm optimization combined with a decision tree as the classifier. This study also compares the performance of our proposed method with other well-known benchmark classification methods (support vector machine, self-organizing map, back propagation neural network, C4.5 decision tree, Naive Bayes, CART decision tree, and artificial immune recognition system) and conducts experiments on 11 gene expression cancer datasets.ConclusionBased on statistical analysis, our proposed method outperforms other popular classifiers for all test datasets, and is compatible to SVM for certain specific datasets. Further, the housekeeping genes with various expression patterns and tissue-specific genes are identified. These genes provide a high discrimination power on cancer classification.
The effect of comorbidity on lung cancer patients' survival has been widely reported. The aim of this study was to investigate the effects of comorbidity on the establishment of the diagnosis of lung cancer and survival in lung cancer patients in Taiwan by using a nationwide population-based study design. This study collected various comorbidity patients and analyzed data regarding the lung cancer diagnosis and survival during a 16-year follow-up period (1995–2010). In total, 101,776 lung cancer patients were included, comprising 44,770 with and 57,006 without comorbidity. The Kaplan–Meier analyses were used to compare overall survival between lung cancer patients with and without comorbidity. In our cohort, chronic bronchitis patients who developed lung cancer had the lowest overall survival in one (45%), five (28.6%), and ten years (26.2%) since lung cancer diagnosis. Among lung cancer patients with nonpulmonary comorbidities, patients with hypertension had the lowest overall survival in one (47.9%), five (30.5%), and ten (28.2%) years since lung cancer diagnosis. In 2010, patients with and without comorbidity had 14.86 and 9.31 clinical visits, respectively. Lung cancer patients with preexisting comorbidity had higher frequency of physician visits. The presence of comorbid conditions was associated with early diagnosis of lung cancer.
Decision tree (DT) analysis was applied in this cross-sectional study to investigate caries experience in children by using clinical and microbiological data obtained from parent–child pairs. Thirty pairs of parents and children were recruited from periodontal and pediatric dental clinics. All participants were clinically examined for caries and periodontitis by a calibrated examiner. Cariogenic and periodontopathic bacteria examinations were conducted. The Kendall rank correlation coefficient was used to measure the association between data variables obtained through clinical and microbiological examinations. A classificatory inductive decision tree was generated using the C4.5 algorithm with the top-down approach. The C4.5 DT analysis was applied to classify major influential factors for children dental caries experience. The DT identified parents’ periodontal health classification, decayed, missing, filled permanent teeth (DMFT) index, periodontopathic test (PerioCheck) result, and periodontal pocket depth as the classification factors for children caries experience. 13.3% of children were identified with a low decayed, missing, filled primary teeth (dmft) index (dmft < 3) whose parents had a periodontal pocket depth ≤3.7, PerioCheck score >1, DMFT index <13.5, and periodontal classification >2. The DT model for this study sample had an accuracy of 93.33%. Here, parental periodontal status and parents’ DMFT were the factors forming the DT for children’s caries experience.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.