Exploring CNNs for Blood Subtype Recognition: Binary and Multi-Class Image Classification

Blood cell subtype classification accuracy is essential for the diagnosis of many illnesses and ailments. This paper introduces a novel methodology for the classification of multi and binary subtypes of blood cells through the utilization of Convolutional Neural Networks (CNNs). The principal aim of this study is to provide a resilient and effective framework for the automated categorization of blood cell subtypes, with the intention of aiding healthcare practitioners in delivering prompt and precise diagnoses. In order to tackle the issue of multi-class classification, a CNN architecture is devised to proficiently acquire distinguishing characteristics from the unfiltered visual data and subsequently associate them with distinct subtypes of blood cells. The classification scenario of binary subtypes, emphasizing crucial distinctions has also been undertaken. For this, the model is being modified to the CNN structure and engaged in fine-tuning the model to augment its capacity to discern variations in the features of blood cells. The study introduces a resilient and adaptable CNN architecture for the categorization of both multi and binary subtypes of blood cells. The technique exhibits a notable level of precision and holds significant promise for practical medical applications, rendering it a useful asset within the realm of medical imaging and diagnostics. of many metrics, including accuracy, precision, recall, and F1-score, which serve to demonstrate its capacity to deliver dependable and consistent outcomes.