Indian Institute of Technology Bhubaneswar

About: Indian Institute of Technology Bhubaneswar is a education organization based out in Bhubaneswar, India. It is known for research contribution in the topics: Large Hadron Collider & Computer science. The organization has 1185 authors who have published 3132 publications receiving 48832 citations.

Hybrid approach using KPSO and RLS for RBFNN design for breast cancer detection

Abstract: Neural networks have been employed in many medical applications including breast cancer classification. Innovation in diagnostic features of tumors may play a central role in development of new treatment methods for earliest stage of breast cancer detection. This study proposes a new hybrid for breast cancer detection by extending the application of a variation of particle swarm optimization called K-particle swarm optimization (KPSO). In this paper, the centers and variances of radial basis functional neural network are initialized by KPSO and then updated using back propagation. The weights are updated using recursive least square instead of back propagation. The results are compared with some recently developed techniques. It is found that the proposed technique provides more accurate result and better classification as compared to some other techniques.

The role of land surface schemes in the regional climate model (regcm) for seasonal scale simulations over western himalaya

Abstract: Climate prediction over the Western Himalaya is a challenging task due to the highly variable altitude and orientation of orographic barriers. Surface characteristics also play a vital role in climate simulations and need appropriate representation in the models. In this study, two land surface parameterization schemes (LSPS), the Biosphere-Atmosphere Transfer Scheme (BATS) and the Common Land Model (CLM, version 3.5) in the regional climate model (RegCM, version 4) have been tested over the Himalayan region for nine distinct winter seasons in respect of seasonal precipitation (three years each for excess, normal and deficit). Reanalysis II data of the National Centers for Environmental Prediction (NCEP)/Department of Energy (DOE) have been used as initial and lateral boundary conditions for the RegCM model. In order to provide land surface boundary conditions in the RegCM model, geophysical parameters (10 min resolution) obtained from the United States Geophysical Survey were used. The performance of two LSPS (CLM and BATS) coupled with the RegCM is evaluated against gridded precipitation and surface temperature data sets from the India Meteorological Department (IMD). It is found that the simulated surface temperature and precipitation are better represented in the CLM scheme than in the BATS when compared with observations. Further, several statistical analysis such as bias, root mean square error (RMSE), spatial correlation coefficient (CC) and skill scores like the equitable threat score (ETS) and the probability of detection (POD) are estimated for evaluating RegCM simulations using both LSPS. Results indicate that the RMSE decreases and the CC increases with the use of the CLM compared to BATS. ETS and POD also indicate that the performance of the model is better with the CLM than with the BATS in simulating seasonal scale precipitation. Overall, results suggest that the performance of the RegCM coupled with the CLM scheme improves the model skill in predicting winter precipitation (by 15-25%) and temperature (by 10-20%) over the Western Himalaya.

Fluoride contamination of groundwater and its seasonal variability in parts of Purulia district, West Bengal, India

Abstract: The contamination of aquifers by fluoride and arsenic is a major cause of concern in several parts of India. A study has thus been conducted to evaluate the extent and severity of fluoride contamination and also its seasonal variability. Two blocks (Purulia-1 and Purulia-2) were considered for this purpose. Twenty groundwater samples (in each season) were collected from tube wells during the pre-monsoon and post-monsoon seasons. In addition to fluoride, groundwater samples were also analyzed for major cations, anions, and other trace elements. The concentration of fluoride shows significant seasonal variation and ranges between 0.94–2.52 and 0.25–1.43 mg/l during the pre-monsoon and post-monsoon seasons, respectively. In pre-monsoon season, more than 40% of the water samples show fluoride concentrations higher than the WHO limit. However, during the post-monsoon season, none of the groundwater sample shows fluoride concentrations higher than the WHO limit. Lesser concentration during the post-monsoon season is attributed to the dilution effect by the percolating rainwater, which has also been reflected in the form of a decrease in concentrations of other elements. The petrographic studies of the rock samples collected from the study area show that the rocks are mainly composed of plagioclase, orthoclase, and quartz with abundant biotite. The weathering and dissolution of biotite plays an important role in controlling the fluoride concentrations in the groundwater of the study area.