Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Author: search.filters.author.Niyas, S.Has files: No

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Depth Data based Chroma Keying using Grab-cut Segmentation
    (Institute of Electrical and Electronics Engineers Inc., 2018) Lestari, P.; Niyas, S.; Krisnandi, D.
    The research presents a depth-image based automatic object segmentation for chroma key editing in multimedia applications. Depth data taken from advanced depth data capturing devices like Microsoft Kinect has a key role in this research. The proposed approach uses both color and depth data and this hybrid segmentation generates results with clear foreground object boundaries. The present system is exclusively designed for segmenting human subjects from a chroma keyed scene. An Aggregate Channel Feature (ACF) based human detection is also employed here to eliminate false detection due to other foreground objects. The depth data results in dark regions may create small errors over edge pixel segmentation. So, the whole process is carried through a sequence of image processing techniques. The pixels nearby the head portion first restored using K-means clustering and then a coarse level segmentation of the human subjects is obtained using Fuzzy C means segmentation. A color characteristic-based segmentation us used here to eliminate most of the background pixels from the foreground subjects. After this coarse level segmentation, an adaptive Tri-map generation is employed and the ultimate fine level segmentation is achieved using Grab-cut Segmentation to generate the foreground human subjects with accurate edge boundaries for matting chroma keyed images/frames. Experimental results validate the segmentation results and its ability for an error free automated segmentation in editing chroma keyed images or video. © 2018 IEEE.
  • No Thumbnail Available
    Item
    Automated Molecular Subtyping of Breast Cancer Through Immunohistochemistry Image Analysis
    (Springer Science and Business Media Deutschland GmbH, 2023) Niyas, S.; Priya, S.; Oswal, R.; Mathew, T.; Kini, J.R.; Rajan, J.
    Molecular subtyping has a significant role in cancer prognosis and targeted therapy. However, the prevalent manual procedure for this has disadvantages, such as deficit of medical experts, inter-observer variability, and high time consumption. This paper suggests a novel approach to automate molecular subtyping of breast cancer using an end-to-end deep learning model. Immunohistochemistry (IHC) images of the tumor tissues are analyzed using a three-stage system to determine the subtype. A modified Res-UNet CNN architecture is used in the first stage to segregate the biomarker responses. This is followed by using a CNN classifier to determine the status of the four biomarkers. Finally, the biomarker statuses are combined to determine the specific subtype of breast cancer. For each IHC biomarker, the performance of segmentation models is analyzed qualitatively and quantitatively. In addition, the patient-level biomarker prediction results are also assessed. The findings of the suggested technique demonstrate the potential of computer-aided techniques to diagnose the subtypes of breast cancer. The proposed automated molecular subtyping approach can accelerate pathology procedures, considerably reduce pathologists’ workload, and minimize the overall cost and time required for diagnosis and treatment planning. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.