2026-10-12T15:47:52Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33951064 2026-09-21T08:03:10Z category_24178 category_24508 item_type_3 month_year_09_2026

BCDD-SUST: An Annotated Peripheral Blood Cell Dataset for AI-Based Dengue Diagnosis Mutashim Ahammed Rafi (23838786) Shakhinur Islam Mondal (23864148) G. M. Nurnabi Azad Jewel (25073694) Molie Rahman (25073703) Arzuba Akter (25073705) Biochemistry and cell biology not elsewhere classified Infectious diseases Blood Cell Detection Medical Imaging Dengue Medical AI Deep Learning <p dir="ltr">BCDD-SUST (Blood Cell Dataset of Dengue - SUST) is an expert-validated peripheral blood smear image dataset developed for research in blood-cell detection, classification, digital hematology, computer vision, and AI-based dengue research. The dataset contains 5,692 peripheral blood smear images collected from 50 patient samples, including 25 dengue NS1-positive and 25 dengue NS1-negative samples.A total of 50,966 expert-validated bounding-box annotations are provided across seven cell classes: Band Neutrophil, Giant Platelet, Monocyte, Normal Lymphocyte, Platelet, Reactive Lymphocyte, and Segmented Neutrophil. The images were acquired at 100× oil-immersion magnification and standardized to 2560 × 1920 pixels. Annotations are provided in YOLO format and organized into training, validation, and test sets. Four expert hematologists participated in annotation review and quality control, with discrepancies resolved through consensus. The dataset is intended for research and benchmarking in peripheral blood smear analysis and AI-based blood-cell detection and classification.</p><p><br></p> 2026-09-21T08:03:10Z Dataset Dataset 10.6084/m9.figshare.33951064.v1 https://figshare.com/articles/dataset/BCDD-SUST_An_Annotated_Peripheral_Blood_Cell_Dataset_for_AI-Based_Dengue_Diagnosis/33951064 CC BY 4.0