The systematic genome analysis platform is presented to increase the automation and accuracy of chromosome-based detection in Down syndrome. This research aimed to overcome the significant problems in cytogenetic image processing, such as noise interference, boundary ambiguity and structural integrity loss during processing. The main goal is to create an end-to-end system that achieves high-fidelity denoising, sharp segmentation, more content feature extraction and strong classification. The proposed methodology combines four main modules, namely Entropy-Guided Denoise Network (EGD-Net) used for entropy-guided noise reduction, Genomic Structure Fragmenting Network (GSF-Net) used for multi-stage genomic segmentation, Hybrid Transform Feature Aggregator (HTF-A) used for Feature Extraction and Adaptive Weighted Pattern Recognition (AWPR-Net) used for Classification. The image of the chromosomes is gradually narrowed in each module and the key patterns of the genome are maintained. The Results illustrate significant improvements when compared to current methods. The EGD-Net proposed a PSNR of 41.27 dB and SSIM of 0.981, which validates the use of the method in noise reduction and preserving structure. The general classification phase reached an accuracy of 94.741% which has good reliability in differentiating normal and abnormal chromosomes. These measures prove that the suggested multi-stage design is solid. The conclusion proves that the integrated pipeline provides higher visual quality, better structural mapping and more powerful predictability. The framework applies a realistic resolution to genomic screening settings, which are based on robust, dependable and high-resolution interpretation of chromosomes.
Keywords: Chromosome Segmentation, Down Syndrome Detection, Feature Aggregation, Genomic Image Analysis, Pattern Classification, Prenatal Screening.
How to cite this article: Shivakumar BL, Priya U, A Hybrid Deep Feature Aggregation and Adaptive Weighted Classifier Model for Robust Cytogenetic Screening of Down Syndrome. Int J Drug Deliv Technol. 2026;16(2s): 247-265; DOI: 10.25258/ijddt.16.247-265