Deep Categorization of Blood Cells using Depthwise Convolutions
T Sudarshan Rao1, N Rohan Sai2, D Koteswara Rao3
1T Sudarshan Rao*, Department of Computer Science and Engineering, GITAM (Deemed to be University), Hyderabad. (Telangana), India.
2N Rohan Sai, Department of Computer Science and Engineering, GITAM (Deemed to be University), Visakhapatnam. (Andhra Pradesh), India.
3D Koteswara Rao, Department of Computer Science and Engineering, GITAM (Deemed to be University), Hyderabad. (Telangana), India.
Manuscript received on October 20, 2021. | Revised Manuscript received on October 27, 2021. | Manuscript published on October 30, 2021. | PP: 76-79 | Retrieval Number: 100.1/ijitee.L957410101221 | DOI: 10.35940/ijitee.L9574.10101221
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Modern-day computation has become indispensable in the healthcare industry. From medical image processing to cost reduction, Artificial Intelligence has proved its significance in solving complex healthcare problems. One of the primary areas in which it can be of greater use in hematology. Categorization of white-blood cells is imperative to pre-identify abnormalities. Through this paper, we collected image samples for 4 major White Blood cell groups, which are Neutrophils, Lymphocytes, Monocytes, and Eosinophils. The aim of this research is to put forward an intelligent system that efficiently alleviates the stringent requirement of a cytological study. The proposed system classifies 4 white-blood-cell types based on their morphological variation. With the experimental modulations that we chose to integrate, the presented model attained an accuracy of 97%.
Keywords: Artificial Intelligence, Cytology, Hematology.
Scope of the Article: Artificial Intelligence