Multimedia Tools and Applications (2024) 83:8063–8076

https://doi.org/10.1007/s11042-023-15923-8

Machine learning in detection and classification of leukemia

using C‑NMC_Leukemia

Fatma M. Talaat1 · Samah A. Gamel2

Received: 6 May 2022 / Revised: 4 March 2023 / Accepted: 29 May 2023 /

Published online: 13 June 2023
© The Author(s) 2023

Abstract
A significant issue in the field of illness diagnostics is the early detection and diagnosis
of leukemia, that is, the accurate distinction of malignant leukocytes with minimal costs
in the early stages of the disease. Flow cytometer equipment is few, and the methods
used at laboratory diagnostic centers are laborious despite the high prevalence of leu-
kemia. The present systematic review was carried out to review the works intending to
identify and categories leukemia by utilizing machine learning. It was motivated by the
potential of machine learning (machine learning (ML)) in disease diagnosis. Leukemia is
a blood-forming tissues cancer that affects the bone marrow and lymphatic system. It can
be treated more effectively if it is detected early. This work developed a new classifica-
tion model for blood microscopic pictures that distinguishes between leukemia-free and
leukemia-affected images. The general proposed method in this paper consists of three
main steps which are: (i) Image_Preprocessing, (ii) Feature Extraction, and (iii) Classifi-
cation. An optimized CNN (OCNN) is used for classification. OCNN is utilized to detect
and classify the photo as "normal" or "abnormal". Fuzzy optimization is used to opti-
mize the hyperparameters of CNN. It is a quite beneficial to use fuzzy logic in the opti-
mization of CNN. As illustrated from results it is shown that, with the using of OCNN
classifier and after the optimization of the hyperparameters of the CNN, it achieved the
best results due to the enhancement of the performance of the CNN. The OCNN has
achieved 99.99% accuracy with C-NMC_Leukemia dataset.

Keywords Machine learning · Leukemia detection · C-NMC_Leukemia · Optimized CNN

(OCNN)

* Fatma M. Talaat
fatma.nada@ai.kfs.edu.eg
Samah A. Gamel
sgamel@horus.edu.eg
1
Faculty of Artificial Intelligence, Kafrelsheikh University, Kafrelsheikh, Egypt
2
Faculty of Engineering, Horus University, Damietta, Egypt

13
Vol.:(0123456789)
8064 Multimedia Tools and Applications (2024) 83:8063–8076

1 Introduction
Leukemia is a type of blood cancer that starts in the bone marrow and results in a large
number of abnormal blood cells. These blood cells, often known as blasts or leukemia cells,
are not fully matured. Bleeding, bruising, bone discomfort, weariness, fever, and a higher
risk of infection are all possible symptoms. A shortage of regular blood cells causes these
symptoms. Blood tests or a bone marrow biopsy are usually used to make the diagnosis. The
actual causes of leukemia are unknown to scientists. It appears to be the result of a mix of
environmental and genetic factors. A fundamental concern in the realm of illness diagnostics
is the precise differentiation of malignant leukocytes with low expense in the early stages of
the disease, which is a major difficulty. Flow cytometry equipment is in restricted supply,
and the procedures accessible in laboratory diagnostic centers are time-consuming. Leuke-
mia is the most frequent type of blood cancer in people of all ages, especially youngsters.
Excessive proliferation and immature growth of blood cells cause this abnormal phenom-
enon, which can harm red blood cells, bone marrow, and the immune system. With over
60,000 new cases reported in 2018, leukemia accounts for more than 3.5 percent of all new
cancer cases in the United States. Malignant white blood cells, also known as lymphoblasts,
travel through the bloodstream to other organs such as the spleen, brain, liver, and kidneys,
where they metastasize to vital bodily tissues [2, 13, 23]. On the basis of microscopic pic-
tures, hematologists in cell transplant facilities can differentiate and diagnose various kinds
of leukemia. Some forms of leukemia are easier to identify and distinguish than others if the
slide is properly stained, but determining underlying leukemia requires more technology.
The stained slides of the most frequent kinds of leukemia as illustrated in Fig. 1. Leukemia
is divided into four categories, acute myeloid leukemia (AML) Fig. 1a, acute lymphoblastic
leukemia (ALL) Fig. 1b, chronic myeloid leukemia (CML) Fig. 1c, and chronic lymphocytic
leukemia (CLL) Fig. 1d, as well as a few less frequent kinds.
In general, leukemia research is divided into two categories: clinical or translational
research and basic research. Clinical/translational research focuses on studying disease
in a specific and generally immediately applicable manner, such as testing a new drug in
humans. Basic science research, on the other hand, looks at the disease process from afar,
such as whether a suspected carcinogen can cause leukemic changes in isolated cells in the
laboratory or how the DNA changes inside leukemia cells as the disease progresses. The
findings of basic research studies are generally less immediately useful to people suffering
from the disease. But the speed of detection of the disease leads to avoiding deteriora-
tion of the patient’s health. Researchers, clinicians, and hematologists have traditionally
struggled to make an early diagnosis of leukemia. Lymphocytic enlargement, pallor, fever,
and weight loss are all indications of leukemia, but they can also be signs of other ill-
nesses. Because the symptoms of leukemia are so minor in the early stages, diagnosing it
can be challenging. Although microscopic examination of PBS is the most commonly used
method for a leukemia diagnosis, obtaining and analyzing bone marrow samples is the gold
standard for leukemia diagnosis [3, 23, 29, 42]. Several research studies have used machine
learning (ML) and computer-aided diagnostic approaches for laboratory image analysis in
the last two decades in the hopes of overcoming the limits of a late leukemia diagnosis and
determining its subgroups. In this research, blood smear pictures were evaluated for diag-
nosing, distinguishing, and counting cells in distinct kinds of leukemia [41, 43].
Traditional approaches couldn’t evaluate or uncover patterns in such a massive amount
of data. It has been demonstrated that machine learning is ideally adapted to dealing with
vast amounts of complex data and could prove to be a useful tool in understanding and
combating disease. Traditionally, experienced practitioners assess diagnostic tests and

13
Multimedia Tools and Applications (2024) 83:8063–8076 8065

Fig. 1  The four main types of

leukemia

a: AM
ML b: AL
LL

c: CML
L d:
d CLL
C L

patient data based on years of medical study and training. However, in a number of tasks,
including initial diagnosis, prognosis estimation, and prediction of treatment problems, as
well as relapse tracking in hematologic malignancies, machine learning algorithms have
recently been demonstrated to be on par with professionals.
Two decades ago, the first studies of ML methods in the diagnosis of hematologic
malignancies were carried out. They began by using flow cytometry to identify leukemic
cells in blood samples and analyzing genetic data to provide the framework for machine
learning methods in the investigation of hematologic malignancies. It is a well-known
branch of artificial intelligence that consists of algorithms and mathematical relationships,
and it has been swiftly used in clinical research. ML allows computers to be programmed
without explicit experience and then learn from it. The results of incorporating these tech-
nologies into medical data processing have been astounding, and they have proven to be
quite effective in illness diagnosis [6, 10, 24]. According to research, ML approaches con-
siderably enhance complex medical decision-making processes in medical image process-
ing by extracting and then assessing the properties of these images [25, 30, 45].
The main contributions of this paper are:

1. Proposing a Proposed Leukemia Classification Technique (PLCT)

2. PLCT consists of three main phases as shown in Fig. 2 which are: (i) Image Preprocess-
ing Phase (IPP), (ii) Feature Extraction Phase (FEP), and (iii) Classification Phase (CP).
3. Finding best values for hyperparameters using a proposed hyperparameter tuning algorithm.
4. Comparing OCNN with the state-of-the-art used algorithms.

13
8066 Multimedia Tools and Applications (2024) 83:8063–8076

Image Feature
Classificaon
Preprocessing Extracon
Phase (CP)
Phase (IPP) Phase (FEP) Normal

- Convert to RGB - Opmized CNN

- CNN (Alexnet)
- Resize (OCNN)
- Augmentaon
Abnormal

Fig. 2  Framework of the Proposed Leukemia Classification Technique (PLCT)

The remaining work is organized as follows. In Section 2, some of the recent related
work in the detection and classification of Leukemia is presented. In Section 3, the pro-
posed method is presented. Experimental evaluation is provided in Section 4. And in Sec-
tion 5, we conclude this work.

2 Related work

This section discusses a number of studies that have been conducted by researchers. The
literature discusses some of the previous research between (2012-2023). De Oliveira et al.
[22] suggest minor changes to typical neural network topologies to obtain good perfor-
mance in the categorization of malignant leukocytes. The architectures of VGG16, VGG19,
and Xception were the ones that were put to the test. To balance the training and validation
sets, data augmentation was used. Mirroring, rotation, blurring, shearing, and the addition
of salt and pepper noise were among the transformations used. Using a ResNeXt convo-
lutional neural network with Squeeze-and-Excitation modules, Jonas Prellberg et al. [28]
provides a simple yet successful classification method. The technique received a weighted
F1 score of 88.91 percent on the test set in the C-NMC online challenge. The source code
can be found at https://​github.​com/​jprel​lberg/​isbi2​019ca​ncer.
Authors in [26] propose the neighborhood-correction algorithm (NCA), which consists
of three major steps: (i) fine-tuning a pre-trained residual network with training data and
producing initial labels and feature maps for test data, (ii) building a Fisher vector for each
cell image based on its feature maps, and (iii) correcting the initial label of each test cell
image using weighted majority voting based on its most similar neighborhood. Supardi
et al. [1] proposed a classification system for acute leukemia that distinguishes between
two types: acute myelogenous leukemia (AML) and acute lymphocytic leukemia (ALL)
(ALL). From image samples, twelve features were manually retrieved. Finally, a K-NN
classifier was employed to categorize the data. Experiments on a database of 1500 photos
yielded an accuracy of 86%.
Kumar et al. [17] presented an automated acute leukemia detection system. The
technique began with noise and blurring in microscopic digital images being pre-
processed. Color, geometric, textural, and statistical characteristics were retrieved
and classified as benign or malignant. The k-nearest neighbor (K-NN) and nave Bayes

13
Multimedia Tools and Applications (2024) 83:8063–8076 8067

classification models were utilized Experimental experiments on a dataset of 60 blood

samples found that the K-NN classifier was superior, with a classification accuracy of
92.8 percent.
Madhukar et al. [20] created a classification approach based on AML that improved
image contrast and identified five features. The classification was done using an SVM
classifier. Experiments on a dataset of 50 photos yielded a classification accuracy of
93.5 percent. Setiawan et al. [31] proposed a classification scheme for cells of AML
subtypes M4, M5, and M7. The cells were first segmented using a color k-means algo-
rithm. Six statistical features were then retrieved and used to train a multi-class SVM
classifier. The results produced 87 percent segmentation accuracy and 92.9 percent
classification accuracy in the best-case scenario.
With feature extraction, coding, and classification, Faivdullah et al. [9] suggested a
three-layered system. The goal of this framework was to determine whether a patient
had leukemia and what form of leukemia they had based on a blood smear image.
In order to extract features, a dense scale-invariant feature transform was applied.
The retrieved feature vectors’ dimensionality was then decreased in the coding layer.
Finally, the categorization was done using a multi-class SVM classifier. Experiments
on a dataset of 400 samples yielded a classification accuracy of 79.37 percent.
Laosai and Chamnongthai [19] proposed an AML classification system based on
k-means and contour signature approaches to segment nuclei. Then, using morphol-
ogy, features such as cell size, cell color, and so on were extracted. The SVM classifier
had an accuracy of up to 92 percent in experiments on a dataset of 100 photos. Patel
and Mishra [27] established a microscopic image-based automated leukemia method.
During preprocessing, the system began by removing noise and blurring. After that,
k-means and Zack algorithms were used to segment the WBCs. After that, color, sta-
tistical, geometric, and textural aspects were retrieved. In [21], an SVM classifier was
used to differentiate between normal and abnormal photos. Experiments on a dataset
of 27 photos yielded a 93.57 percent accuracy. Table 1 summarizes the most common
recent studies in Leukemia classification that uses C-NMC 2019 Dataset. Finally, in
[4] the authors proposed a model for the categorization of acute leukemia images, they
suggested a CNN model based on the Tversky loss function includes six convolution
layers, four dense layers, and a Softmax activation function. The proposed approach
was 99% accurate in diagnosing acute leukemia types such as ALL and AML. Table 2
provides the pros and cons of the various algorithms discussed.

3 The proposed leukemia classification technique (PLCT)

The Proposed Leukemia Classification Technique (PLCT) consists of three main

phases as shown in Fig. 2 which are: (i) Image Preprocessing Phase (IPP), (ii) Feature
Extraction Phase (FEP), and (iii) Classification Phase (CP).

3.1 Image preprocessing phase (IPP)

The input blood microscopic images are first transformed into an RGB color model, and
then a number of processes are applied to them at this stage. After that, their dimensions
are set to 227 x 227. Finally, data augmentation is used to compensate for the lack of a

13
 Table 1  Recent studies in Leukemia classification that uses C-NMC 2019 Dataset
8068

References Used Methodology Description Limitation Results

13
De Oliveira et al. (2021) [22] Data Augmentation and Convo- This work proposes simple modifications to High memory In this study, the best model achieved
lutional Neural Networks standard neural network architectures to usage, High an F1-score of 92.60%, precision of
achieve high performance in the malignant computa- 91.14%, sensitivity of 94.10%, and
leukocyte classification problem. The tional time specificity of 90.86% for the malig-
tested architectures were VGG16, VGG19, nant class.
and Xception. Data augmentation was
employed to balance the training and
validation sets. Transformations such as
mirroring, rotation, blurring, shearing,
and the addition of salt and pepper noise
were used.
Jonas Prellberg ResNeXt convolutional neural To obtain good performance in the malig- Low efficiency On the test set, the technique was tested
et al. (2020) [28] network with Squeeze-and- nant leukocyte classification challenge, and received a weighted F1-score of
Excitation modules. this paper suggests modest modifications 88.91 percent.
to typical neural network topologies.
Mirroring, rotation, blurring, shearing, and
adding salt and pepper noise were among
the transformations employed.
Yongsheng Pan et al. (2019) [26] Neighborhood-Correction Algo- Authors propose the neighborhood-cor- Complexity, Experiments show that in early test-
rithm (NCA) rection algorithm (NCA), which consists High compu- ing, the suggested NCA achieves a
of three major steps: (i) fine-tuning a tational Cost weighted F1-score of 92.50 percent
pre-trained residual network with train- and a balanced accuracy of 91.73 per-
ing data and producing initial labels and cent, and in final testing, it achieves a
feature maps for test data, (ii) building a weighted F1-score of 91.04 percent,
Fisher vector for each cell image based on ranking first in C-NMC.
its feature maps, and (iii) correcting the
initial label of each test cell image using
weighted majority voting based on its most
similar neighborhood.
Multimedia Tools and Applications (2024) 83:8063–8076
 Table 2  Pros and cons of the various algorithms
Algorithm Pros. Cons.

Neural Networks • Effective for both large and small datasets need only less statistical training • High computational cost
[5] • Capable of detecting complex nonlinear relationships between dependent • Network behavior that isn’t described causes issues
and independent variables • Works with numerical data
• Network duration is unknown
• Prone to overfitting
k-NN • There is no need for training • Incapable of handling large datasets
[18] • New data can be added without difficulty • Incapable of handling high dimensions
• It is simple to deploy • Expensive computation
• Requires feature scaling
Multimedia Tools and Applications (2024) 83:8063–8076

SVM • High accuracy • Consumes a lot of memory

[16] • No need for a linearly separable feature space • only capable of binary classification
• Works well with unstructured and semi-structured data • Doesn’t scale to large datasets
• Scales well to high-dimensional data • Takes a long time to learn
Deep Learning • Can be applied to a variety of applications and data types • Training is extremely expensive because of complicated data models
[15] • Does not require feature engineering; features are automatically deduced • Expensive GPUs are required
• What is learned is difficult to comprehend
Naïve Bayes • Quick Convergence • Assumes that all qualities are linearly independent; however, this is
[14] • Requires minimal training data not the case in reality.
• Simple Classifier • There’s a chance you’ll lose precision
• Effective for both large and small datasets • You won’t be able to change dependencies
• Each feature is independent of the others • Assumes that numeric properties are distributed appropriately.
8069

13
8070 Multimedia Tools and Applications (2024) 83:8063–8076

large dataset, which is necessary for deep neural networks to complete their training and
testing phases. Translation, reflection, and rotation are the three operations that make up
data augmentation. The images are shifted along the X-and Y-axes in translation, with
selected values randomly bound by the interval [1, 9, 17, 19–22, 26–28, 31]. The images
are mirrored along the vertical axis during the reflection process. Finally, the photos are
rotated right or left with a random rotation angle of values bounded by the interval [21–25]
with a step equal to five during the rotation process.

3.2 Feature extraction phase (FEP)

CNN is one of the most common network designs used in machine-learning applications.
The capacity of CNNs to complete tasks regardless of tilting, translation, or scaling is the
major reason for their success [44]. Convolutional, pooling, and fully connected layers are
the three primary types of layers in the CNN architecture, as depicted in Fig. 3. Convolu-
tional layers compute the output of neurons by adding the bias to the weighted sum and
using a rectified linear unit as an activation function (ReLu).

3.3 Classification phase (CP)

Image classification could be useful in robotics, human-machine interface, surveillance,

retrieval, transportation, and other domains. It is one of the most active study areas in com-
puter vision. It entails the categorization of photographs. It looks for regions in an image
that might contain a specific object, then extracts and classifies each of these regions with
the use of an image classification model. An optimized CNN (OCNN) is used for classifica-
tion. OCNN is utilized to detect and classify the photo as "normal" or "abnormal." Fuzzy
optimization is used to optimize the CNN hyperparameters. The use of fuzzy logic in optimi-
zation is a quite beneficial. When faced with a challenging task, the fuzzy logic system can
deliver the most efficient solution. It’s simple to tweak the performance to improve or change
it. CNN has a significant advantage over its predecessors in that it can detect crucial charac-
teristics without the requirement for human interaction. CNN’s performance is improved by
optimizing its hyperparameters. The overall steps of the OCNN are shown in Algorithm 1.

Fig. 3  General architecture of convolution neural networks (CNNs) [40]

13
Multimedia Tools and Applications (2024) 83:8063–8076 8071

Input:
o HPT containing initial values for the hyperparameters of CNN.
Output:
o The optimal values for the hyperparameters.
Steps:
1: Initialize gbest and lr (gbest=Vi and lr=lr0)
2: Collect data from HPT
3: Add a loop between values in HPT
4: calculate the Fitness Value (FVi) using Fuzzy Logic via the three parameters stored in HPT.
5: compare the value of FVi if its greater than gbest it will perform gbest= FVi and lr=lri
6: Update the HPT
7: Assign the new values to the HPT

Algorithm 1  OCNN algorithm

4 Implementation and experiments

The following data identify the primary components of the computer system used for the
simulation tests. Although the actual hardware may alter the overall completion time for
a simulation, it should have little, if any, impact on the simulation outcomes. In contrast,
for completeness and to allow this study to be duplicated for future comparison, the fol-
lowing is a summary of the most important components. Intel FX6300 9-core processor
at 6.5GHz, Memory: 64 GB DDR3 RAM divided into two 32GB banks4 terabyte sata3
physical drive, 64-megabyte cache capacity. Windows 10 64-bit with Service Pack 1 is the
operating system that we used. Nota bene: Windows 11 was accessible as "upgrades" dur-
ing the research period. They were denied in order to ensure compatibility and consistency
of operation across the research. The following section describes the implementation of our
model, the experiments conducted, and the used dataset.

4.1 Dataset

The C-NMC 2019 dataset (https://​www.​kaggle.​com/​gaura​vrajp​al/​leuke​mia-​class​ifica​tion-​

v1-3-​incep​tionv3-​65-​29/​data) (Mourya et al. 2019) contains 15114 lymphocyte images
collected from 118 subjects and divided into three folders with names such as "C-NMC
training data" which contains 10661 cells, 7272 malignant cells from 47 subjects, and 3389
healthy cells from 26 subjects; "CNMC test preliminary phase data", which contains 1867

Fig. 4  Sample of the dataset

13
8072 Multimedia Tools and Applications (2024) 83:8063–8076

Table 3  Number of samples in Before After

training, validation and test sets Data Augmentation Data Augmentation
Malignant Healthy Malignant Healthy

Training 4364 2034 10000 10000

Validation 2181 1016 5000 5000
Test 727 339 N/A N/A

cells, 1219 malignant cells from 13 subjects, and 648 healthy cells from 15 subjects, and
“C-NMC test final phase data” containing 2586 unlabeled cells from 17 subjects. Single-
cell photos of malignant and benign lymphocytes already identified by experienced oncolo-
gists can be found in these folders. A sample of the data set is presented in Fig. 4.

4.2 Data augmentation

The original dataset was unbalanced and, for that reason, data augmentation was employed
to balance the Training and Validation sets. This technique was not applied to the Test
set. Standard image transformation techniques were used, such as mirroring, rotation, and
Gaussian blurring, to produce the augmented images.
We also attempted different augmentation techniques, and it was noticed that shearing
and addition of salt and pepper noise resulted in a better training performance and model
accuracy. An example of these techniques applied to a random image from the dataset can
be seen in Fig. 4. The augmented Training set had 20,000 samples, and the Validation set
has 10,000 samples, as shown in Table 3.

4.3 Implementation and experiments

Python code was used to implement the proposed method including OCNN to evaluate
their performance in terms of precision, recall, accuracy, and specificity, which are defined
as shown.
Precision = TP∕ (TP + FP) (1)

Table 4  The performance of the Method Performance Metrics

proposed method (OCNN) vs.
previous classifiers Precision Recall Accuracy Specificity

OCNN 99.97% 100.00% 99.99% 99.98%

CNN 99.65% 100.00% 99.82% 99.65%
SVM-Linear 99.93% 98.72% 99.33% 99.93%
SVM-Gaussian 99.93% 99.43% 99.68% 99.93%
SVM-Cubic 99.93% 99.65% 99.79% 99.93%
K-NN 99.64% 98.44% 99.04% 99.65%
LD 99.64% 97.38% 98.51% 99.65%
DT 95.69% 95.96% 95.82% 95.67%

13
Multimedia Tools and Applications (2024) 83:8063–8076 8073

101.00%

100.00%
OCNN
99.00%
CNN
98.00% SVM-Linear

SVM-Gaussian
97.00%
SVM-Cubic
96.00% K-NN

LD
95.00%
DT
94.00%

93.00%
Precision Recall Accuracy Specificity

Fig. 5  Comparison of OCNN vs. previous algorithms

Recall = TP∕ (TP + FN) (2)

Accuracy = (TP + TN)∕ (TP + TN + FP + FN) (3)

Specif icity = TN∕ (TN + FP) (4)

where
TP is true positive; TN is true negative; FP is false positive; and FN is false negative.
The performance of the proposed method (OCNN) is compared with the previous com-
mon used classifiers as Decision Tree (DT), Linear Discriminant (LD), Support Vector
Machine (SVM), and K-Nearest Neighbor (K-NN) as shown in Table 4.

F1-score
95.00%
94.00%
93.00%
92.00%
91.00%
90.00%
89.00%
88.00%
87.00%
86.00%
OCNN CNN Oliveira et al. Prellberg et al. Pan et al. (2019)
(2021) [22] (2020) [28] [26]

Fig. 6  OCNN vs. CNN and previous methods

13
8074 Multimedia Tools and Applications (2024) 83:8063–8076

Table 5  The performance of the Method F1-score

proposed method (OCNN) vs.
CNN and previous methods
OCNN 94.07%
CNN 93.86%
De Oliveira et al. (2021)[22] 92.60%
Jonas Prellberg 88.91%
et al. (2020) [28]
Yongsheng Pan et al. (2019)[26] 92.50%

A graphical representation of the results is shown in Fig. 5. From Fig. 5, It is shown

that the OCNN as a classifier achieved the best results due to the enhancement of the per-
formance of the CNN after the optimization of the hyperparameters of the CNN. Figure 6
illustrates the performance of the proposed algorithm vs. previous algorithms.
The performance of the proposed method (OCNN) is compared with CNN and previous
methods using F1-score as shown in Table 5.
It is shown from Table 5, the OCNN outperforms CNN and other algorithms.

5 Conclusions

A significant issue in the field of illness diagnostics is the early detection and diagnosis of
leukemia, that is, the accurate distinction of malignant leukocytes with minimal costs in the
early stages of the disease. Flow cytometer equipment is few, and the methods used at labo-
ratory diagnostic centers are laborious despite the high prevalence of leukemia. Leukemia
can be treated more effectively if it is detected early. This work developed a new classifi-
cation model for blood microscopic pictures that distinguishes between leukemia-free and
leukemia-affected images. The general proposed method in this paper consists of three main
steps which are: (i) Image_Preprocessing, (ii) Feature Extraction, and (iii) Classification.
An optimized CNN (OCNN) is used for classification. OCNN is utilized to detect and clas-
sify the photo as "normal" or "abnormal." Fuzzy optimization is used to optimize the CNN
hyperparameters. The use of fuzzy logic in optimization is a quite beneficial. From results,
it is shown that the OCNN as a classifier achieved the best results due to the enhancement
of the performance of the CNN after the optimization of the hyperparameters of the CNN.
In the future work, we can use OCNN [35] to achieve better results as it achieved a good
performance in [7, 8, 11, 12, 32, 33, 36–39]. We can also use correlation methods like [34].

Funding Open access funding provided by The Science, Technology & Innovation Funding Authority
(STDF) in cooperation with The Egyptian Knowledge Bank (EKB).

Data availability https://​www.​kaggle.​com/​gaura​vrajp​al/​leuke​mia-​class​ifica​tion-​v1-3-​incep​tionv3-​65-​29/​data

Declarations
Conflict of interest We wish to confirm that there are no known conflicts of interest associated with this
publication.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com-
mons licence, and indicate if changes were made. The images or other third party material in this article

13
Multimedia Tools and Applications (2024) 83:8063–8076 8075

are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly
from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.

References
1. (2020) Classification of Blasts inAcute Leukemia Blood samplesUsing k-NearestNeighbour—IEEEConfer-
ence Publication. Available online: https://​ieeex​plore.​ieee.​org/​abstr​act/​docum​ent/​61947​69/ (accessed on 3
February 2020).
2. (2018) Hematology TAS of facts-and-statistics. http://​www.​hemat​ology.​org/. Accessed 25 Oct 2022
3. (2020) Childhood leukemias. https://​doi.​org/​10.​1056/​NEJM1​99506​15332​2407
4. Ansari S et al (2023) A customized efficient deep learning model for the diagnosis of acute leukemia cells
based on lymphocyte and monocyte images. Electronics 12(2):322
5. Das NN et al (2022) Automated deep transfer learning-based approach for detection of COVID-19 infection
in chest X-rays.". Irbm 43(2):114–119
6. Ehrenstein V, Nielsen H, Pedersen AB, Johnsen SP, Pedersen L (2017) Clinical epidemiology in the era of
big data: new opportunities, familiar challenges. Clin Epidemiol 9:245–250
7. El-Rashidy N, ElSayed NE, El-Ghamry A, Talaat FM (2022) Prediction of gestational diabetes based
on explainable deep learning and fog computing Soft Comput 26(3):. https://​doi.​org/​10.​1007/​
s00500-​022-​07420-1
8. Nora El-Rashidy, Nesma Ebrahim, Amir el Ghamry, Fatma M. Talaat, (2022) Utilizing fog computing and
explainable deep learning techniques for gestational diabetes prediction. Neural Comput Applic. https://​
doi.​org/​10.​1007/​s00521-​022-​08007-5
9. Faivdullah L, Azahar F, Htike ZZ, Naing WN (2015) Leukemia detection from blood smears. J Med Bioeng
4:488–491
10. Ghaderzadeh M, Rebecca F, Standring A (2013) Comparing performance of different neural networks for
early detection of cancer from benign hyperplasia of prostate. Appl Med Inform 33:45–54
11 Hanaa S, Fatma BT (2022) Detection and Classification Using Deep Learning and Sine-Cosine Fitness
Grey Wolf Optimization. Bioengineering 10(1):18. https://​doi.​org/​10.​3390/​bioen​ginee​ring1​00100​18
12. Hassan E, El-Rashidy N, Talaat FM (2022) Review: Mask R-CNN Models. https://​doi.​org/​10.​21608/​njccs.​
2022.​280047.
13. Hegde RB, Prasad K, Hebbar H, Singh BMK, Sandhya I (2019) Automated decision support system for
detection of leukemia from peripheral blood smear images. J Digit Imaging 33:361–374
14. Hsu Chih-Yu, Wang Shuai, Qiao Yu (2021) Intrusion detection by machine learning for multimedia plat-
form. Multimedia Tools Applic 80(19):29643–29656
15. Kumar N et al. (2021) Efficient automated disease diagnosis using machine learning models. J Healthcare
Eng 2021:‫‏‬
16. Kumar N, Aggarwal D (2021) Learning-based focused WEB crawler." IETE J Res 1-9‫‏‬
17 Kumar S, Mishra S, Asthana P (2018) Automated detection of acute leukemia using k-mean clustering
algorithm. Advances in Computer and Computational Sciences. Springer, Berlin/Heidelberg, pp 655–670
18 Kumar N et al (2022) Automatic Diagnosis of Covid-19 Related Pneumonia from CXR and CT-Scan
Images. Eng Technol Appl Sci Res 12(1):7993–7997
19. Laosai J, Chamnongthai K (2014) Acute leukemia classification by using SVM and K-Means clustering. In
Proceedings of the 2014 IEEE International Electrical Engineering Congress (iEECON), Chonburi, pp. 1–4
20. Madhukar M, Agaian S, Chronopoulos AT (2012) Deterministic model for acute myelogenous leukemia
classification. In Proceedings of the 2012 IEEE International Conference on Systems, Man, and Cybernet-
ics (SMC), Seoul, pp. 433–438
21. Maria IJ, Devi T, Ravi D (2020) Machine learning algorithms for diagnosis of leukemia. Int J Sci Technol
Res 9(1):
22. Maurício de Oliveira J, Dantas D (2021) Classification of normal versus leukemic cells with data augmen-
tation and convolutional neural networks. In Proceedings of the 16th International Joint Conference on
Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) – vol.
4: VISAPP, pp. 685-692. https://​doi.​org/​10.​5220/​00102​57406​850692
23. Namayandeh SM, Khazaei Z, LariNajafi M, Goodarzi E, Moslem A (2020) GLOBAL Leukemia in chil-
dren 0–14 statistics 2018, incidence and mortality and human development index (HDI): GLOBOCAN
sources and methods. Asian Pac J Cancer Prevent 21(5):1487–1494
24. Obermeyer Z, Emanuel EJ (2016) Predicting the future—big data, machine learning, and clinical medi-
cine. N Engl J Med 375(13):1216–1219

13
8076 Multimedia Tools and Applications (2024) 83:8063–8076

25. Oikonomidis I, Kyriazis N, Argyros AA (2011) Efficient model-based 3D tracking of hand articulations
using Kinect, In Proceedings of the 22nd British Machine Vision Conference, BMVC’2011, Dundee
26. Pan Y et al. (2019) Neighborhood-Correction algorithm for classification of normal and malignant cells”,
Lecture Notes Bioeng
27 Patel N, Mishra A (2015) Automated leukemia detection using microscopic images. Procedia Comput Sci
58:635–642
28. Prellberg J, Kramer O (2020) Acute lymphoblastic leukemia classificationfrom microscopic images using
convolutionalneural networks,arXiv:1906.09020v2 [cs.CV] 1 Apr 2020
29. Pui C-H (1995) Childhood leukemias. N Engl J Med 332(24):1618–1630
30. Reader TW, Flin R, Mearns K, Cuthbertson BH (2009) Developing a team performance framework for the
intensive care unit. Crit Care Med 37(5):1787–1793
31. Setiawan A, Harjoko A, Ratnaningsih T, Suryani E, Palgunadi S (2018) Classification of cell types in
Acute Myeloid Leukemia (AML) of M4, M5 and M7 subtypes with support vector machine classifier.
In Proceedings of the 2018 International Conference on Information and Communications Technology
(ICOIACT), Yogyakarta, pp. 45–49
32. Talaat FM (2022) Effective deep q-networks (edqn) strategy for resource allocation based on optimized
reinforcementlearning algorithm. Multimedia Tools Applic 1–17
33. Talaat FM (2022) Effective prediction and resource allocation method (EPRAM) in fog computing envi-
ronment for smart healthcare system. Multimed Tools Appl
34. Talaat FM, Gamel SA (2022) Predicting the impact of no. Of authors on no. Of citations of research pub-
lications based on neural networks. J Ambient Intell Humaniz Comput ahead-of-print. https://​doi.​org/​10.​
1007/​s12652-​022-​03882-1
35. Talaat FM, Gamel SA (2022) RL based hyper-parameters optimization algorithm (ROA) for convolutional
neural network. J Ambient Intell Human Comput. https://​doi.​org/​10.​1007/​s12652-​022-​03788-y
36. Talaat FM, Saraya MS, Saleh, AI, Ali HA, Ali SH, (2020) A load balancing and optimization strategy
(LBOS) using reinforcement learning in fog computing environment J Ambient Intell Human Comput
11(12):. https://​doi.​org/​10.​1007/​s12652-​020-​01768-8
37. Talaat FM, Ali SH, Saleh AI, Ali HA (2020) Effective cache replacement strategy (ECRS) for real-time
fog computing environment. Cluster Comput 23(16):. https://​doi.​org/​10.​1007/​s10586-​020-​03089-z
38. Talaat FM, Ali HA, Saraya MS et al (2022) Effective scheduling algorithm for load balancing in fog envi-
ronment using CNN and MPSO. Knowl Inf Syst 64:773–797. https://​doi.​org/​10.​1007/​s10115-​021-​01649-2
39. Talaat Fatma M, Alshathri Samah, Nasr Aida A (2022) A New Reliable System For Managing Virtual
Cloud Network”. Comput Mater Continua 73(3):5863–5885. https://​doi.​org/​10.​32604/​cmc.​2022.​02654
40. Vogado LHS, Veras RDMS, Andrade AR, De Araujo FHD, e Silva RRV, Aires KRT (2017) Diagnosing
leukemia in blood smear images using an ensemble of classifiers and pre-trained convolutional neural net-
works. In Proceedings of the 2017 IEEE 30th SIBGRAPI Conference on Graphics, Patterns and Images
(SIBGRAPI), Niteroi, pp. 367–373
41. Wen J, Xu Y, Li Z, Ma Z, Xu Y (2018) Inter-class sparsity based discriminative least square regression.
Neural Netw 102:36–47
42. Wolach O, Stone RM (2017) Mixed-phenotype acute leukemia. Curr Opin Hematol 24(2):139–145
43. Xing F, Yang L (2016) Robust nucleus/cell detection and segmentation in digital pathology and micros-
copy images: a comprehensive review. IEEE Rev Biomed Eng 9:234–263
44. Yu W, Chang J, Yang C, Zhang L, Shen H, Xia Y, Sha J (2017) Automatic classification of leukocytes
using deep neural network. Proceedings of the 2017 IEEE 12th International Conference on ASIC (ASI-
CON), Guiyang, China. IEEE, Piscataway, pp 1041–1044
45. Zhao J, Zhang M, Zhou Z, Chu J, Cao F (2017) Automatic detection and classification of leukocytes using
convolutional neural networks. Med Biol Eng Comput 55(8):1287–1301

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

13