WBC-KICNet: knowledge-infused convolutional neural network for white blood cell classification

WBC-KICNet: knowledge-infused convolutional neural network for white blood cell classification Jeneessha P http://orcid.org/0000-0001-8522-0480 Vinoth Kumar Balasubramanian http://orcid.org/0000-0001-8564-5120 M Murugappan http://orcid.org/0000-0002-5839-4589 Abstract

White blood cells (WBCs) are useful for diagnosing infectious diseases and infections. Machine learning and deep learning have been used to classify WBCs from blood smear images. Despite advances in machine learning, there has been little research on applying medical domain knowledge to convolutional neural networks (CNNs) to improve WBC classification. The existing models are often inaccurate, rely on manual input, and fail to incorporate external medical knowledge into decision-making. This study used the blood cell count and detection dataset which contains images of monocytes, lymphocytes, neutrophils, and eosinophils for WBC classification. In this paper, we propose a CNN model for WBC classification called WBC-KICNet (knowledge-infused convolutional neural network). The present work uses two CNN models: the first model generates the knowledge vector from input images and the domain expert (hematologist); the second model extracts deep features from the input image. A feature fusion mechanism is then used to combine these two features to classify the WBCs. Several metrics have been used to evaluate the performance of the WBC-KICNet model. These measures yielded impressive results. Accuracy, precision, recall, specificity, and F1-score were rated 99.22%, 99.25%, 99%, 99.77%, and 99.25%, respectively. In each of the WBC classes, accuracy rates are: 98.7% for eosinophils, 99.83% for lymphocytes, 100% for monocytes, and 98.32% for neutrophils. As a result, the proposed WBC-KICNet classifies WBCs accurately and without much misclassification, and the results have been confirmed by a statistical hypothesis test ( t -test).
10 01 2024 09 01 2024 035086 10.1088/crossmark-policy iopscience.iop.org WBC-KICNet: knowledge-infused convolutional neural network for white blood cell classification Machine Learning: Science and Technology paper © 2024 The Author(s). Published by IOP Publishing Ltd 2024-06-16 2024-09-12 2024-10-01 https://creativecommons.org/licenses/by/4.0/ https://iopscience.iop.org/info/page/text-and-data-mining 10.1088/2632-2153/ad7a4e https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e/pdf https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e/pdf https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e/pdf https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e/pdf https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e/pdf https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e https://iopscience.iop.org/article/10.1088/2632-2153/ad7a4e/pdf Cognit. Syst. Res. Tiwari 10.1016/j.cogsys.2018.08.022 52 1036 2018 Detection of subtype blood cells using deep learning Diagnostics Cheuque 10.3390/diagnostics12020248 12 248 2022 An efficient multi-level convolutional neural network approach for white blood cells classification SN Appl. Sci. Cinar 10.1007/s42452-021-04485-9 3 1 2021 Classification of lymphocytes, monocytes, eosinophils, and neutrophils on white blood cells using hybrid Alexnet-GoogleNet-SVM BMC Bioinform. Chen 10.1186/s12859-022-04824-6 23 282 2022 Accurate classification of white blood cells by coupling pre-trained ResNet and DenseNet with SCAM mechanism Neural Comput. Appl. Fırat 10.1007/s00521-023-09158-9 36 1599 2024 Classification of microscopic peripheral blood cell images using multi- branch light weight CNN- based model Appl. Soft Comput. Togacar 10.1016/j.asoc.2020.106810 97 2020 Classification of white blood cells using deep features obtained from convolutional neural network models based on the combination of feature selection methods Expert Syst. Appl. Banik 10.1016/j.eswa.2020.113211 149 2020 An automatic nucleus segmentation and CNN model-based classification method of white blood cell J. Ambient Intell. Humaniz. Comput. Dong 10.1007/s12652-021-03642-7 14 9839 2023 A novel feature fusion based deep learning framework for white blood cells classification PLoS One Wang 10.1371/journal.pone.0261848 17 2022 WBC-AMNet: automatic classification of WBC images using deep feature fusion network based on focalized attention mechanism Opt. Laser Technol. Wang 10.1016/j.optlastec.2021.106931 139 2021 A 3D attention networks for classification of white blood cells from microscopy hyperspectral images Biomed. Signal Process. Control Ha 10.1016/j.bspc.2022.103611 75 2022 Fine-grained interactive attention learning for semi-supervised white blood cell classification Shah 1 2022 Detection of different types of blood cells: a comparative analysis Lei 10.1109/SCET.2012.6341960 2012 Multiclass classification of microarray data samples with flexible neural tree J. Biomed. Inf. Wang 10.1016/j.jbi.2008.12.010 42 654 2009 A neural network-based biomarker association information extraction approach for cancer classification Biomed. Signal Process. Control Girdhar 10.1016/j.bspc.2021.103156 71 2022 Classification of white blood cell using convolution neural network Neural Comput. Appl. Balasubramanian 10.1007/s00521-022-07279-1 34 16089 2022 An approach to classify white blood cells using convolutional neural network optimized by particle swarm optimization algorithm Comput. Intell. Neurosci. Sharma 10.1155/2022/7384131 2022 2022 Deep learning model for the automatic classification of white blood cells IEEE Trans. Med. Imaging Shin 10.1109/TMI.2016.2528162 35 1285 2016 Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning Artif. Cells, Nanomed. Biotechnol. Yao 10.1080/21691401.2021.1879823 49 147 2021 Classification of white blood cells using weighted optimize deformable convolutional neural networks Zhu 10.1109/ICCV.2017.244 2223 2017 Unpaired image-to-image translation using cycle-consistent adversarial networks Med. Image Anal. Xie 10.1016/j.media.2021.101985 69 2021 A survey on incorporating domain knowledge into deep learning for medical image analysis J. Inf. Technol. Constr. Tian 10.36680/j.itcon.2024.013 29 264 2024 Integrating domain knowledge with deep learning model for automated worker activity classification in mobile work zone IEEE Trans. Med. Imaging Xie 10.1109/TMI.2018.2876510 3894 991 2019 Knowledge based collaborative deep learning for benign—malignant lung nodule classification on chest CT Med. Image Anal. Liu 10.1016/j.media.2019.101555 58 2019 Automated detection and classification of thyroid nodules in ultrasound images using clinical knowledge-guided convolutional neural networks Kaggle BCCD Dataset Adv. Comput. Data Sci. Joshi 10.1007/978-981-13-9942-8_27 1046 282 2019 Issues in training a convolutional neural network model for image classification Chen 11624 Learning active contours for medical image segmentation Soil Sci. Soc. Am. J. Fan 10.2136/sssaj2017.01.0009 18 1139 2017 Measurement of soil color: a comparison between smartphone camera and the munsell color charts Expert Syst. Appl. Yang 10.1016/j.eswa.2021.116436 194 2022 Efficient active contour model for medical image segmentation and correction based on edge and region information Int. J. Autom. Technol. Gong 10.1007/s11633-018-1117-z 15 2 2018 An overview of contour detection approaches