Classification of White Blood Cell Abnormalities for Early Detection of Myeloproliferative Neoplasms Syndrome Using Backpropagation

Fitri, Zilvanhisna Emka; Imron, Arizal Mujibtamala Nanda

doi:10.1007/978-981-33-6926-9_43

Cited by 5 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The process is carried out to reduce the computational load [9]. After the cropping process is carried out, the next process is to split the RGB color space components, this is done because the RGB color image has a large value so it is difficult to segment [10]. Generally, RGB images are converted to other color spaces such as grayscale, HSV, or CIELab, but the focus of researchers is only on solving the RGB color space to determine which image components best represent the shape of the red dragon fruit among the red, green and blue components.…”

Section: Image Preprocessingmentioning

confidence: 99%

Comparison of Classification for Grading Red Dragon Fruit (<i>Hylocereus Costaricensis</i>)

Fitri

Baskara

Madjid

et al. 2022

JNTE

Self Cite

View full text Add to dashboard Cite

Pitaya is another name for dragon fruit which is currently a popular fruit, especially in Indonesia. One of the problems related to determining the quality of dragon fruit is the postharvest sorting and grading process. In general, farmers determine the grading system by measuring the weight or just looking at the size of the fruit, of course, this raises differences in grading perceptions so that it is not by SNI. This research is a development of previous research, but we changed the type of dragon fruit from white dragon fruit (Hylocereus undatus) to red dragon fruit (Hylocereus costaricensis). We also adapted the image processing and classification methods in previous studies and then compared them with other classification methods. The number of images in the training data is 216, and the number of images in the testing data is 75. The comparison of the accuracy of the three classification methods is 84% for the KNN method, 85.33% for the Naive Bayes method, and 86.67% for the Backpropagation method. So that the backpropagation method is the best classification method in classifying the quality grading of red dragon fruit. The network architecture used is 4, 8, 3 with a learning rate of 0.3 so that the training accuracy is 98.61% and the testing accuracy is 86.67%.

show abstract

Section: Image Preprocessingmentioning

confidence: 99%

Comparison of Classification for Grading Red Dragon Fruit (<i>Hylocereus Costaricensis</i>)

Fitri

Baskara

Madjid

et al. 2022

JNTE

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this research, two segmentation processes were carried out, namely grayscale thresholding and channel area thresholding. Grayscale thresholding is segmentation based on the threshold value of the gray level of the image, the process uses the formula equation [6] :…”

Section: Segmentationmentioning

confidence: 99%

“…Generally, if abnormal cell counts are found, then a peripheral blood smear is performed [4]. On examination of the peripheral blood smear, megathrombocytes or giant platelets were found [3] (Figure 1) but basophil cells were not found in myeloproliferative disorders [5], [6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Detection of Essential Thrombocythemia based on Platelet Count using Channel Area Thresholding

Destarianto

Noviana

Fitri

et al. 2022

J. RESTI (Rekayasa Sist. Teknol. Inf.)

Self Cite

View full text Add to dashboard Cite

Essential Thrombocythemia is one of the Myeloproliferative Neoplasms Syndrome where the mutation of the JAK2V617F gene causes the bone marrow to produce excessive platelets. For early detection of Essential Thrombocythemia disease using a full blood count and peripheral blood smear examination. The main characteristic is that giant platelets are found as large as young lymphocytes with a number of more than 21 cells in one field of view. The purpose of this research is to detect Essential Thrombocythemia by counting the number of platelets in the peripheral blood smear image. This research utilizes computer vision technique where the research stages consist of peripheral blood smear image, color conversion, image enhancement, segmentation, labeling process, feature extraction and K-Nearest Neighbor classification. There are three features used, namely the number of platelet cells, area and perimeter. The K-Nearest Neighbor method is able to classify 215 training data with an accuracy of 98.13% and classify 40 testing data with an accuracy of 100% based on the value of K = 3.

show abstract

“…In more recent study, autoencoder neural network was applied to identify a feature set of megakaryocytes cytomorphology to distinguish MPN subtype with outcome of area under the curve metrics shows the result of 0.95 [11]. In the following year, study by [12] was implemented morphological feature extraction to discriminate the size of blood cells in pre-processing stage for early detection of MPN. The backpropagation network architecture was used as a classifier with a learning rate variation to produce the best accuracy rate of 91.82%.…”

Section: Introductionmentioning

confidence: 99%

Hyperparameter Tuning in Deep Learning Approach for Classification of Classical Myeloproliferative Neoplasm

Yusof

Mashohor

Hanafi

et al. 2022

Malaysian J. Sci. Adv. Tech.

View full text Add to dashboard Cite

Histopathology images are an essential resource for defining biological compositions or examining the composition of cells and tissues. The analysis of histopathology images is also crucial in supporting different class of disease including for rare disease like Myeloproliferative Neoplasms (MPN). Despite technological advancement in diagnostic tools to boost procedure in classification of MPN, morphological assessment from histopathology images acquired by bone marrow trephine (BMT) is remained critical to confirm MPN subtypes. However, the outcome of assessment at a present is profoundly challenging due to subjective, poorly reproducible criteria and highly dependent on pathologist where it caused interobserver variability in the interpretation. To address, this study developed a classification of classical MPN namely polycythemia vera (PV), essential thrombocythemia (ET) and primary myelofibrosis (MF) using deep learning approach. Data collection was undergoing several image augmentations processes to increase features variability and expand the dataset. The augmented images were then fed into CNN classifier followed by implementation of cross validation method. Finally, the best classification model was performed 95.3% of accuracy by using Adamax optimizer. High accuracy and best output given by proposed model shows significant potential in the deployment of the classification of MPN and hence facilitates the interpretation and monitoring of samples beyond conventional approaches.

show abstract

Classification of White Blood Cell Abnormalities for Early Detection of Myeloproliferative Neoplasms Syndrome Using Backpropagation

Cited by 5 publications

References 4 publications

Comparison of Classification for Grading Red Dragon Fruit (<i>Hylocereus Costaricensis</i>)

Comparison of Classification for Grading Red Dragon Fruit (<i>Hylocereus Costaricensis</i>)

Detection of Essential Thrombocythemia based on Platelet Count using Channel Area Thresholding

Hyperparameter Tuning in Deep Learning Approach for Classification of Classical Myeloproliferative Neoplasm

Contact Info

Product

Resources

About