Carcinoembryonic Antigens Segmentation and Quantitative Analysis from Fluorescent Images using Principal Component Analysis and Adaptive K-means Clustering

• Published in: 2023 International Conference on Energy, Power, Environment, Control, and Computing (ICEPECC) pp. 1 - 6
• Main Authors: Aslam, Muhammad Aqeel, Mahnoor, Shahzadi, Munir, Muhammad Asif, Cheema, Saman, Hassan, Khawaja Humble, Sajid, Abdullah
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.03.2023
• Subjects: Adaptive K-means Clustering; Antigens; Clustering algorithms; Dimensionality reduction; Fluorescence; Fluorescent Images; Image segmentation; Machine learning algorithms; Principal Component Analysis (PCA); Segmentation; Simple Linear Iterative Clustering (SLIC) Algorithm; Strips; Superpixels
• DOI: 10.1109/ICEPECC57281.2023.10209525

Now fluorescent scan images are extensively used for the detection of antigens. The identification and treatment of the tumor is done with the help of these images The speed of detection is the major part for such systems. Although, bulk of research has already been done, segmentation of images is the area still need improvement. Characterization of the images is difficult task due to the diverse nature of the input images. This paper presents a novel method for the segmentation. The segmentation is done using superpixels. In the proposed algorithm the super pixels are studied on the basis of their average value. This value is computed with the help of Principal component analysis and then PCA system is utilized to compute a feature vector corresponding to the each superpixel. The stated method was implemented in MATLAB 2017. Our system integrates a series of algorithms. These algorithms are used for quantitative image analysis.