Determining the dynamics of pathways associated with cancer progression is critical for understanding the etiology of diseases on genomic data. However analysis of the multi-omic data for ovarian cancer acquired by pretreatment and post-treatment is also still challenging due to complex characteristics such as appearance, boundaries and density of the tumor. In this paper, we devise anovarian component analysis, termed as novel technique which encompass the Deep Neural Network for prediction, Principle Component Analysis (PCA)for classification and dimensionality reduction on feature extracted. The feature selection and extraction on the ovarian image is carried out usingScale Invariant Feature Transform mechanism (SIFT),to select the discriminate features with prognostic value. Two initial feature pools were built using image features computed from pre-therapy CT images only and image feature difference computed from both pre- and post-therapy. A feature selection method was applied to select optimal features, and an equal-weighted fusion method was used to generate a new quantitative imaging marker from each pool to predict 6-month progression-free survival. The proposed model evaluate the progress of the tumor after applying the clinical trialsfor the identification of stage-specific and dynamic modules information .These process is applied on segment metastatic tumor tracked by radiologists on MRI images. The experimental result demonstrates that the stage-specific modules significantly improve the accuracy of cancer stage prediction. The prediction accuracy between quantitative imaging markers and the Response Evaluation Criteria in Solid Tumors (RECIST) criteria was also compared.
Volume 12 | 02-Special Issue
Pages: 434-441
DOI: 10.5373/JARDCS/V12SP2/SP20201089