OBJECTIVES: Complete Barrett's eradication endoscopic mucosal resection (CBE-EMR) is the removal of all epithelium with curative intent eliminating high-grade dysplasia (HGD)/intramucosal carcinoma (IMC) and reducing risk metachronous lesion development. We report our single tertiary referral center's long-term clinical experience using this modality in HGD/IMC management. METHODS: In study, we retrospectively reviewed patients who had CBE-EMR for esophagus (BE) been entered into...

Background Prenatal and perinatal factors have been linked to affective disorders. We therefore undertook an exploratory study determine whether prenatal exposure severe famine was associated with increased risk of Method Monthly birth cohorts that were exposed unexposed the Dutch Hunger Winter 1944–45 identified. The cumulative incidences psychoses neurotic depression (ICD–9 criteria) compared between during each trimester gestation. Results relative (RR) psychosis (broad restricted...

Based on binary particle swarm optimisation (BPSO) and information theory, this paper proposes two new filter feature selection methods for classification problems. The first algorithm is based BPSO the mutual of each pair features, which determines relevance redundancy selected subset. second entropy group evaluates Different weights in fitness functions proposed algorithms are used to further improve their performance terms number features accuracy. In experiments, a decision tree (DT)...

Feature selection has the two main objectives of minimising classification error rate and number features. Based on binary particle swarm optimisation (BPSO), we develop novel multi-objective feature frameworks for classification, which are PSO using idea non-dominated sorting (NSBPSO) ideas crowding, mutation dominance (CMDBPSO). Four methods then developed by applying mutual information entropy as different filter evaluation criteria in each proposed frameworks. The algorithms examined...

The authors tested whether early prenatal exposure to famine was associated with schizophrenia spectrum personality disorder in addition being per se as shown a previous study.The risk of schizoid disorder, defined by ICD-6 ICD-9, men at age 18 years compared birth cohorts that were conceived the height Dutch Hunger Winter and unexposed 1944-1946 region Holland.The exposed cohort had significantly greater (relative = 2.01) disorder.Prenatal nutritional deficiency years.

This study relates violent behavior of schizophrenic inpatients to demographic, historical, EEG, neurological, and neuropsychological variables. Patients were classified into high (N = 28), low 27), or no 34) violence groups. There significant differences among the groups on demographic historical variables, except for prevalence crime, which was higher in both than nonviolent patients. Neurological abnormalities differentiated groups, with group evidencing more other two area integrative...

Feature selection is a multi-objective problem with the two main conflicting objectives of minimising number features and maximising classification performance. However, most existing feature algorithms are single objective do not appropriately reflect actual need. There small algorithms, which wrapper based accordingly computationally expensive less general than filter algorithms. Evolutionary computation techniques particularly suitable for optimisation because they use population...

Unbalanced classification is an essential machine learning task, which has attracted widespread attention from both the academic and industrial communities due mainly to its broad applications. Evolutionary computation (EC) contributed greatly unbalanced classification. However, best of our knowledge, there have not been any comprehensive investigations on strengths weaknesses alternative EC methods in addressing various challenging problems This article reviews literature utilize techniques...

This study examined the relationship between childhood experience and homelessness in psychiatric patients.Three large diverse samples of homeless patients (N = 512) were compared with a sample who had never been 271), respect to foster care, group home placement, running away. One drawn from admitted state mental hospital. In this hospital population, risk ratios for lifetime prevalence could be derived.In three samples, over 15% history 10% 20% These figures 2%, 1%, 5%, respectively,...

OBJECTIVES: Cost-effective programs are needed to assist homeless persons with severe mental illness in their transition from shelters community living. The authors investigated the cost-effectiveness of critical time intervention program, a time-limited adaptation intensive case management, which has been shown significantly reduce recurrent homelessness among men illness. METHODS: Ninety-six study participants recruited psychiatric program men's public shelter 1991 1993 were randomly...

A sample of 87 psychiatric inpatients known to have been assaultive while in the hospital was contrasted with a matched group nonviolent patients identify personal risk factors that distinguished two groups. Data were collected using history interview, neurological examination, and an electroencephalogram. The four identified--neurological abnormality, violent crime, suicide attempts, deviant family environment childhood--were used develop statistical model predicting which subjects newly...

The authors compared the risk of schizophrenia in Dutch birth cohorts that were or not exposed during second trimester gestation to 1957 A2 influenza epidemic. Exposed did have a higher schizophrenia. These findings suggest that, some populations, there is no relation between prenatal exposure and

Feature selection is a multi-objective problem, where the two main objectives are to maximize classification accuracy and minimize number of features. However, most existing algorithms belong single objective, wrapper approaches. In this work, we investigate use binary particle swarm optimization (BPSO) probabilistic rough set (PRS) for feature selection. We PRS propose new measure features based on which filter objective algorithm (PSOPRSE) developed. Then filter-based (MORSE) proposed,...