Article: Habitual detection and measurement of human blood cells on hyperspectral imagery for convolutional neural network Journal: International Journal of Biomedical Engineering and Technology (IJBET) 2024 Vol.45 No.1 pp.1 - 12 Abstract: This paper's proposed strategy has taken a stab at a different dataset of smear pictures, where it has performed acceptably. With the accuracy and the disclosure execution of the proposed system, it will in general be said that the procedure has the potential to ease off the manual platelet recognising verification and counting measure. A complete platelet incorporate is a critical test in clinical end to survey by and large disease. Usually, platelets are counted actually using haemocytometer close by other examination office equipment's and substance blends, which is a dreary and monotonous commission. In this effort, the makers nearby an AI advance for modified ID and together with of three kinds of platelets using 'you simply look once' (YOLO) object acknowledgment and gathering computation. YOLO structure has been set up with a changed arrangement BCCD Dataset of blood smear picture to normally recognise and check red platelets, white platelets, and platelets. Experimental results demonstrate that the proposed model can achieve better classification performance than traditional CNNs and widely used support vector machine (SVM) approaches, especially as training small-sample-size situations. Inderscience Publishers - linking academia, business and industry through research

Title: Habitual detection and measurement of human blood cells on hyperspectral imagery for convolutional neural network

Authors: T. Arumuga Maria Devi; PaulRaj Thangaselvi

Addresses: Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Abishekapatti, Tirunelveli – 627012, Tamil Nadu, India ' Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Abishekapatti, Tirunelveli – 627012, Tamil Nadu, India

Abstract: This paper's proposed strategy has taken a stab at a different dataset of smear pictures, where it has performed acceptably. With the accuracy and the disclosure execution of the proposed system, it will in general be said that the procedure has the potential to ease off the manual platelet recognising verification and counting measure. A complete platelet incorporate is a critical test in clinical end to survey by and large disease. Usually, platelets are counted actually using haemocytometer close by other examination office equipment's and substance blends, which is a dreary and monotonous commission. In this effort, the makers nearby an AI advance for modified ID and together with of three kinds of platelets using 'you simply look once' (YOLO) object acknowledgment and gathering computation. YOLO structure has been set up with a changed arrangement BCCD Dataset of blood smear picture to normally recognise and check red platelets, white platelets, and platelets. Experimental results demonstrate that the proposed model can achieve better classification performance than traditional CNNs and widely used support vector machine (SVM) approaches, especially as training small-sample-size situations.

Keywords: classification; detection; feature extraction; red blood cell; RBC; segmentation; sickle cell disease; support vector machine; SVM.

DOI: 10.1504/IJBET.2024.138620

International Journal of Biomedical Engineering and Technology, 2024 Vol.45 No.1, pp.1 - 12

Received: 29 Sep 2022
Accepted: 29 Mar 2023
Published online: 16 May 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Habitual detection and measurement of human blood cells on hyperspectral imagery for convolutional neural network

Keep up-to-date