Publications (3)3.61 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes a novel Discrete Cosine Transform (DCT) spectrum based approach to enhance the performance of a face recognition (FR) system employing a unique astroid shaped feature selection from the DCT spectrum. Individual stages of the FR system are examined and an attempt is made to improve each stage. A Binary Particle Swarm Optimization (BPSO)-based feature selection algorithm is used to search the feature vector space for the optimal feature subset. Experimental results show the promising performance of astroid shaped DCT feature extraction for face recognition on ORL, UMIST, Extended Yale B and Color FERET databases.
    Proceedings of the CUBE International Information Technology Conference; 09/2012
  • Source

    Oral Oncology 02/2012; 48(6):e19. DOI:10.1016/j.oraloncology.2012.02.006 · 3.61 Impact Factor
  • Source
    Article: ORCAdb
    [Show abstract] [Hide abstract]
    ABSTRACT: OrCa-dB: A complete catalogue of molecular and clinical information in oral carcinogenesis Oral Squamous Cell Carcinomas (OSCC), are the most common head and neck neoplasms, that account for about 90% of all oral cancers and is much more common in worldwide especially Asian countries. 1 High throughput genomic technologies such as micro-array, proteomics, transcriptomics and gene sequencing develop-ment has resulted in creation of enormous data and exposure of hundreds of genes and microRNAs that are differentially expressed, thus providing researchers important resources to potentially ex-plore the molecular mechanisms and help in developing diagnostic methods or therapeutic strategies. We believe this integrated knowledge database will serve to provide researchers with a multi-tude of information related to oral cancer aiming to support researchers as well as clinicians at a molecular level. We have compiled molecular as well as clinical information on oral cancer in a freely accessible database, OrCa-dB, which is avail-able at OrCa-dB is a compendium of genes, proteins, miRNAs, pathway and tissue specific expression patterns involved in oral cancer development and progression. OrCa-dB was constructed after extensive manual curation of PubMed re-cords, dated from 1970 to December 2011, for genes impaired in molecular and genetic events during oral cancer progression. An important highlight of OrCa-dB is the inclusion of clinically relevant information such as role of socioeconomic background in early diag-nosis, benefits of screening interventions, chemotherapeutic drugs, use of targeted therapy, clinical trials in oral cancer, management of patients and their quality of life. Each entry in OrCa-dB is linked to external databases such as UniProt, 2 SwissProt, 3 KEGG pathway database, 4 MirBase, 5 EMBL 6 and PDB 7 as an added advantage for harnessing related biological insights. To the best of our knowledge none of the existing oral cancer databases would cover such a vast expanse of information narrowing down the gap between bench research and clinical evaluation of oral cancer. OrCa-dB will be a specialized, first of its kind value-added data-base that will enable effortless pursuit of relevant knowledge for all experimentally determined human oral-cancer-related infor-mation and their interactomics along with clinical facts, thus mak-ing it a unique resource in the area of oral cancer biology. We are working to amplify the quantity of data and to supply additional database function such as querying the database with keywords. Incorporation of additional search and retrieval features to enhance the usability of the web interface will also be consid-ered in the next version of OrCa-dB. Details on microarray and gene expression will also be included in the next update. Since many biological functions involve the formation of protein–protein complexes we also propose to construct a protein interaction map to depict the same.