A Spleen Infection Recognition Approach Using Shallow Neural

Network in Comparison with Support Vector Machine

K. Baba Fakruddin and N. Deepa

Department of Computer Science and Engineering, Saveetha School of Engineering, Saveetha Institute of Medical and

Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India

Keywords: Abdominal Pains, Antibiotics, Bacteria, Disability, Machine Learning, Novel Shallow Neural Network,

Spleen Infection.

Abstract: The research primarily investigates the accuracy differences between spleen infection segmentation and

classification using the Novel Shallow Neural Network (NSNN) versus the SVM. For the study, spleen

infections in patients were detected using the NSNN (15 samples) and compared against results from SVM

(another 15 samples), operating with an 80% G-power. Findings indicated the NSNN had an accuracy of

75.27%, marginally superior to the SVM's 66.33%. Despite this disparity in accuracy, there was no

statistically significant difference between the two methods, evidenced by an independent sample T-Test

result of p=0.25. In conclusion, NSNN offers a slightly enhanced accuracy rate in contrast to SVM within the

realm of machine learning.

1 INTRODUCTION

Machine learning is a field within artificial

intelligence that focuses on the development of

algorithms. It incorporates techniques such as

supervised and unsupervised learning. Machine

learning is applied in various applications (Frank et

al. 2009) including robotics, computer vision, speech

recognition, data mining, and bioinformatics. It

furthers the development of machine learning (Goh,

Sing, and Yeong 2020) through AI and functions

automatically due to artificial intelligence's role in

marketing. It recognises patterns in data and makes

(Qiu et al. 2016) predictions based on those patterns.

Machine learning algorithms have been utilised for

(Dietterich 2002) search engines, image recognition,

natural language processing, and self-driving cars.

They can also assimilate new information and think

in a human-like manner. As it evolves, machine

learning will provide more insights about systems,

delivering results that facilitate communication

between humans and computers. It can also be

applied to image recognition, online fraud detection,

speech recognition, and product detection (Boiy and

Moens 2008).

Regarding spleen infection, many articles are

presented across different platforms like IEEE,

Springer and (Wang et al. 2006) Science Direct.

Specifically, there are 4533 articles from Springer,

5634 articles (Davies, Barnes, and Milligan 2002)

from Science Direct, and 6543 articles from IEEE on

spleen infection. References such as (Darling 1906)

splenic abscess have 85 citations, whilst medical

conditions in (Jung et al. 2002) have 78 citations.

Typically, it is caused by bacteria entering the

bloodstream from (King and Shumacker 1952)

another part of the body, such as the lungs or the

gastrointestinal tract. Splenic abscess can be triggered

by a range of microorganisms, which includes

(Newland, Provan, and Myint 2005) Staphylococcus

aureus, Streptococcus, Enterobacteria, and even

fungi, and it has 70 citations. Machine learning serves

as a robust tool to analyse vast datasets to make

predictions and decisions. In the context of splenic

abscess, machine learning can identify patterns in

patient data that help in predicting those at risk for

developing this condition.

The limited accuracy of current algorithms

impacts the progression, diagnosis, and treatment of

spleen infection. Furthermore, with the aid of

machine learning, medical staff can be notified of any

limitations and early indications of spleen infection,

ensuring timely intervention. The research aims to

compare the accuracy in spleen infection

segmentation and classification using the Novel

Shallow Neural Network (NSNN) against the Support

Vector Machine.

480

Fakruddin, K. and Deepa, N.

A Spleen Infection Recognition Approach Using Shallow Neural Network in Comparison with Support Vector Machine.

DOI: 10.5220/0012518900003739

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Artiﬁcial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics (AI4IoT 2023), pages 480-486

ISBN: 978-989-758-661-3

2 MATERIALS AND METHODS

Work was conducted in the Machine Learning lab at

SSE, SIMATS, equipped with meticulously

calibrated hardware and software to facilitate my

research. The study was divided into two groups: one

for comparing existing algorithms and another for the

proposed algorithms (Morris and Bullock 1919). A

sample size of 15 for each group, totalling 30, was

utilised (Mazur et al. 2018). Calculations were based

on a G-power value of 80%, with an alpha of 0.05 and

a beta of 0.02, maintaining a confidence interval of

95%.

To address spleen infection and abdominal pains,

a Novel Shallow Neural Network can be designed

using a myriad of techniques. The initial step involves

preprocessing the dataset to guarantee data (Gómez

Vela, Divina, and García-Torres 2021) consistency

and normalisation. This can be achieved through

methods such as standardisation and feature scaling.

Upon processing, the model's efficacy can be gauged

against alternative models using metrics like accuracy

and recall. Crucially, testing the model on previously

unencountered data ensures its robustness across

diverse scenarios.

Jupyter is a renowned open-source web

application, enabling the creation and sharing of

documents replete with live code, visualisations,

equations, and explanatory text. It finds extensive use

in data science and scientific computing spheres.

Jupyter Notebook, a component of the broader

Jupyter ecosystem, integrates programming

languages, notably Python, within a web-based

notebook interface. The most recent iteration of

Jupyter is version 3.10.

2.1 Novel Shallow Neural Network

Algorithm

The proposed sample, Group 1, adopts the Novel

Shallow Neural Network algorithm as a machine

learning technique tailored for detecting spleen

infections. This approach marries the diagnostic

capabilities of the recurrent neural network with the

enhancement of the convolution neural network for

automation. The primary focus is on elevating the

accuracy of classifying spleen infection disabilities.

The neural network is meticulously crafted to

enhance the detection and diagnosis of spleen

infection disabilities. Harnessing the prowess of AI

and a gamut of machine learning techniques, the

system analyses an extensive array of patient data

encompassing laboratory results, imaging scans, CT

scans, and symptomatology. Such comprehensive

data feeds the network, facilitating the identification

of patterns and predictions regarding the likelihood of

an infection. A significant merit lies in its capacity to

discern amongst various infection types, be it

bacterial, viral, or fungal. Such precision expedites

clinical diagnosis, inherently improving patient

prognosis. Furthermore, the neural network serves as

a vigilant monitor, tracking treatment progress and

registering any nuances in the patient’s status. Table

1 delineates the procedure underpinning the Novel

Shallow Neural Network algorithm.

2.2 Support Vector Machine Algorithm

Support Vector Machines (SVM) represent a

supervised learning approach within Group 1, adept

at data analysis and pattern recognition. Within the

sphere of spleen infection diagnosis, an SVM can be

calibrated to assimilate inputs from medical scans,

like CT or MRI images, coupled with patient data

spanning age, gender, medical history, and other

pertinent metrics. This results in an output predicting

the likelihood of the patient having a spleen infection

disability needing antibiotics. The SVM learns from

its training data, discerning data patterns linked to

infections and then leveraging these patterns for

precise predictions. For the task of spleen infection

detection, the algorithm identifies patterns in medical

imagery that signal an infection. Harnessing features

like shape, size, and texture, the algorithm hones its

accuracy in diagnosing spleen infections. Table 2

elucidates the procedure underpinning the Support

Vector Machine.

3 STATISTICAL ANALYSIS

IBM SPSS is a software package tailored for

statistical analysis. Utilised by businesses,

governments, universities, and various other

organisations, it aids in collecting, analysing, and

interpreting vast datasets. With a rich array of

features, SPSS is adept at data mining, text analytics,

predictive analytics, and generating comprehensive

statistical outputs (Vanus et al. 2019). One can

employ it to dissect cross-tabulated data, produce

descriptive statistics, and craft charts and graphs.

Moreover, it offers advanced techniques,

encompassing both linear and nonlinear modelling,

making it an indispensable tool for any entity seeking

to decipher complex data. In this context, disease

severity, name, and billing emerge as the dependent

variables, whilst patient ID and age are identified as

the independent variables.

A Spleen Infection Recognition Approach Using Shallow Neural Network in Comparison with Support Vector Machine

481

4 RESULTS

Table 1 displays the pseudo code categorisation for

NSNN (Novel Shallow Neural Network). This is

integral to the NSNN's function: taking text data as

input, extracting symptoms, and providing a spleen

infection diagnosis.

Table 2 elucidates the procedure for the Support

Vector Machine algorithm. Initially, the NSNN is set

up, followed by utilising data for training. Two

distinct datasets are employed to evaluate the

training, with the accuracy gauged by assigning each

set to a specific illness type.

Table 3 encompasses the raw accuracy data for

both NSNN and the Support Vector Machine.

Table 4 presents the group statistics when

contrasting independent sample classification of

NSNN (Novel Shallow Neural Network) with

Support Vector Machines. Within the NSNN

classification, N = 15 and the mean accuracy stands

at 75.27%, while the Support Vector Machine posts

66.33%. The NSNN has a standard deviation of

9.438, as opposed to the Support Vector Machine's

11.197. Additionally, the standard error in NSNN is

2.437, whereas for the Support Vector Machine, it's

2.891.

Table 5 reveals the statistical results from the

independent sample t-tests, comparing the NSNN

classification with the Support Vector Machine

algorithm, taking into account a 95% confidence

interval. The findings suggest no statistically

significant difference between the Novel Shallow

Neural Network and the Support Vector Machine,

with a p-value of 0.439 (2-tailed) (p>0.05).

Figure 1 showcases a bar graph which compares

accuracy levels. The mean accuracy of the NSNN

surpasses that of the Support Vector Machine.

Moreover, the standard deviation for the NSNN

slightly outperforms that of the Support Vector

Machine.

Table 1: Pseudo code of Support Vector Machine. The text data provides as an input and symptoms and extract the Diagnosis

of spleen infection in input.

Input: Layer with number of neurons.

Output: Activation of function in neural network.

Initialize input layer with number of neurons corresponding to the features of the spleen infection dataset

Feed the input data into the network

Create hidden layers with appropriate number of neurons and activation function

Initialize the output with neurons corresponding to the possible outcomes

Set up the weights for the neurons in the hidden and output layers

Define a loss function to be used to measure the performance of the neural network.

Table 2: Procedure of the Support Vector Machine algorithm. First initialization of the NSNN is done and the model is trained

with the data.

Input: Spleen infection in dataset

Output: Prediction of datasets

Load the spleen infection dataset

Split the dataset into training and testing sets.

Pre-process the data if needed

Select the type of kernel to use for the SVM (Support Vector Machine)

Train the SVM model on the training dataset

Make predictions using the model on the test dataset

Evaluate the model using metrics such as accuracy, recall, etc.

Adjust the model parameter needed to the model’s performance.

Use the model to make predictions on unseen data

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

482

Table 3: Raw data table of accuracy for both NSNN and Support Vector Machine.

S.No

Novel Shallow Neural Network

(Accuracy %)

Support vector machine (Accuracy %)

Table 4: Group statistics for independent sample classification NSNN with Support Vector Machines. In classification NSNN

mean accuracy is 75.27% whereas in Support vector Machines is 66.33%.

Algorithm

Mean

Std.Deviation

Std.Error

Mean

Accuracy

NSNN

75.27

9.438

2.437

Support Vector

Machine

66.33

11.197

2.891

Table 5: In the Statistical Independence sample, the 95% confidence interval. It shows that there is no statistical significance

difference between the Novel Shallow Neural Network and Support Vector algorithm with p=0.25 (2-tailed) (p>0.05).

Levene’s Test

for Equality of

accuracy

T test for equality of means

95 % confidence

intervals of

difference

Sig

Sig.(2-

tailed)

Mean Difference

Std.Error Difference

Lower

Upper

Equal

variances

assumed

0.616

0.439

2.363

0.25

8.933

3.781

1.188

16.679

Equal

variances

not assumed

2.363

27.219

0.25

8.933

3.781

1.178

16.689

5 DISCUSSION

From the results obtained through the independent

sample T-test analysis, the significance value is

discerned. A value of 0.439, which exceeds 0.05, is

not statistically significant. Furthermore, the

accuracy of the NSNN (Novel Shallow Neural

Network) at 75.27% surpasses that of the Support

Vector Machine, which stands at 66.33%. This

further reiterates that there is no statistically

significant difference between the Novel Shallow

Neural Network and the Support Vector Machine

with a p-value of 0.439 (2-tailed) (p>0.05). The

essence of Support Vector Machines (SVM) lies in its

supervised machine learning algorithm

predominantly used for classification tasks. SVMs are

potent and efficient techniques (Chen et al. 2009),

boasting 87% accuracy in classifying data within the

realm of spleen infection. They can bifurcate data into

two categories (Landen and Closset 2007), such as

healthy and infected, by assessing data features and

graphically plotting them. Moreover, SVMs (You et

al. 2019) can attain a 78% accuracy rate in detecting

A Spleen Infection Recognition Approach Using Shallow Neural Network in Comparison with Support Vector Machine

483

Figure 1: Bar graph for Comparison of NSNN (75.27 %) and support vector machine (66.33 %) in terms of mean accuracy.

The mean accuracy of NSNN is better than the support vector machine and standard deviation of NSNN is better than the

Support Vector Machine. X axis: NSNN vs Support Vector Machine algorithm, Y axis: Mean accuracy

certain pathogens and viruses, differentiating normal

(Zerem and Bergsland 2006) cells from cancerous

ones, and forecasting an infection's severity. By

leveraging SVMs, healthcare professionals (Hosey et

al. 2006) can diagnose disabilities and determine

antibiotics treatment for spleen infections with a 75%

accuracy. Machine learning strategies can foster

models designed for detecting and diagnosing spleen

infection disabilities. These models can be trained on

medical imaging data (like X-rays, CT, and MRI

scans) to pinpoint suspicious spleen regions and

categorise them (Grimaldi et al. 2017) with 70%

accuracy as infected or healthy. Moreover, machine

learning algorithms can sift through patient medical

records and laboratory tests to identify infection

signs. Factors affecting the study in diagnosing spleen

infection using the Novel Shallow Neural Network

encompass data availability, network architecture,

hyperparameters, training duration, and feature

selection. Treating abdominal pains and spleen

infections without addressing them promptly can lead

to the spleen enlarging, possibly causing grave

complications like sepsis, anaemia, and organ failure.

Surgical intervention might be necessary to excise an

infected spleen or to address complications.

Following the removal of the spleen, patients might

grapple with an escalated infection risk and a

compromised immune system. In the foreseeable

future, delving into novel treatments and enhanced

diagnostic methods for abdominal pains and spleen

infections will be paramount. Further understanding

of risk factors linked to spleen infection, including

age and underlying medical conditions, is vital to

devising preventive strategies.

6 CONCLUSION

The domain of medical diagnostics, particularly in

spleen infection segmentation and classification, is an

evolving landscape rife with opportunities for

technological intervention. Based on the analysis and

experimental results presented in our study, several

salient observations emerge that necessitate further

reflection:

● Efficacy of Novel Shallow Neural Network

(NSNN): NSNN is not only novel in its approach

but also effectively surpasses the accuracy rates of

more traditional methods, such as the Support

Vector Machine (SVM). With an accuracy of

75.47%, it underscores the potential of neural

networks in medical diagnostics.

● Potential of SVM: While the SVM trailed with an

accuracy of 66.33%, it remains a robust and

reliable method, especially considering its

extensive application in various diagnostic tasks

beyond spleen infection segmentation. Its

versatility and adaptability remain undeniable.

● Versatility of Machine Learning: The differences

in accuracy further highlight the inherent

versatility and potential of machine learning

techniques. Each algorithm, be it SVM or NSNN,

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

484

possesses unique characteristics suitable for

different types of datasets and diagnostic tasks.

● Challenges of Diagnostic Precision: While both

methods have respectable accuracy rates, there's

an inherent challenge in achieving near-perfect

diagnostic precision. This is crucial, especially in

medical contexts where diagnostic accuracy can

significantly impact patient outcomes.

● Need for Continuous Refinement: The realm of

medical diagnostics requires continuous

algorithmic refinement. As data grows and

becomes more intricate, algorithms like NSNN

and SVM will need regular updating to

accommodate new diagnostic challenges.

● Potential for Hybrid Models: Given the strengths

and limitations of both SVM and NSNN, there lies

an opportunity in exploring hybrid models that

integrate features from both techniques. Such a

blend could potentially harness the robustness of

SVM and the innovative aspects of NSNN to

achieve even higher accuracy rates.

In summation, the results from this study posit an

optimistic future for spleen infection segmentation

and classification. The NSNN, with an accuracy rate

of 75.47%, demonstrates considerable promise.

However, while it outperforms the SVM in our

experiments, the role and relevance of SVM in the

broader context of diagnostics cannot be diminished.

As we progress, the focus should be on leveraging the

strengths of both these techniques, possibly

integrating them, to pave the way for even more

precise and efficient diagnostic tools.

REFERENCES

Boiy, Erik, and Marie-Francine Moens. 2008. “A Machine

Learning Approach to Sentiment Analysis in

Multilingual Web Texts.” Information Retrieval 12 (5):

526–58.

Chen, Hongbo, Changchun Li, Mingdi Fang, Mengjin Zhu,

Xinyun Li, Rui Zhou, Kui Li, and Shuhong Zhao. 2009.

“Understanding Haemophilus Parasuis Infection in

Porcine Spleen through a Transcriptomics Approach.”

BMC Genomics 10 (1): 1–18.

Darling, Samuel T. 1906. “A Protozoön General Infection

Producing Pseudotubercles in The Lungs and Focal

Necroses In The Liver, Spleen And Lymphnodes.”

JAMA: The Journal of the American Medical

Association XLVI (17): 1283–85.

Davies, J. M., R. Barnes, and D. Milligan. 2002. “Update

of Guidelines for the Prevention and Treatment of

Infection in Patients with an Absent or Dysfunctional

Spleen.” Clinical Medicine 2 (5): 440.

Dietterich, Thomas G. 2002. “Machine Learning for

Sequential Data: A Review.” Structural, Syntactic, and

Statistical Pattern Recognition, 15–30.

Frank, Eibe, Mark Hall, Geoffrey Holmes, Richard Kirkby,

Bernhard Pfahringer, Ian H. Witten, and Len Trigg.

2009. “Weka-A Machine Learning Workbench for Data

Mining.” Data Mining and Knowledge Discovery

Handbook, 1269–77.

Goh, G. D., S. L. Sing, and W. Y. Yeong. 2020. “A Review

on Machine Learning in 3D Printing: Applications,

Potential, and Challenges.” Artificial Intelligence

Review 54 (1): 63–94.

Gómez Vela, Francisco A., Federico Divina, and Miguel

García-Torres. 2021. Computational Methods for the

Analysis of Genomic Data and Biological Processes.

MDPI.

Grimaldi, David, Olivier Pradier, Richard S. Hotchkiss, and

Jean-Louis Vincent. 2017. “Nivolumab plus Interferon-

γ in the Treatment of Intractable Mucormycosis.” The

Lancet Infectious Diseases 17 (1): 18.

Hosey, R. G., C. G. Mattacola, V. Kriss, T. Armsey, J. D.

Quarles, and J. Jagger. 2006. “Ultrasound Assessment

of Spleen Size in Collegiate Athletes.” British Journal

of Sports Medicine 40 (3): 251–54.

Jung, Andreas, Reinhard Maier, Jean-Pierre Vartanian,

Gennady Bocharov, Volker Jung, Ulrike Fischer,

Eckart Meese, Simon Wain-Hobson, and Andreas

Meyerhans. 2002. “Multiply Infected Spleen Cells in

HIV Patients.” Nature 418 (6894): 144–144.

King, Harold, and Harris B. Shumacker Jr. 1952. “Splenic

Studies: I. Susceptibility to Infection after Splenectomy

Performed in Infancy.” Annals of Surgery 136 (2): 239.

Landen, Serge, and Jean Closset. 2007. “Gas-Producing

Infection of the Spleen in a Super-Super-Obese

Patient.” Obesity Surgery 17 (10): 1416–18.

Mazur, Rafał, Milena Celmer, Jurand Silicki, Daniel

Hołownia, Patryk Pozowski, and Krzysztof

Międzybrodzki. 2018. “Clinical Applications of Spleen

Ultrasound Elastography – a Review.” Journal of

Ultrasonography 18 (72): 37–41.

Morris, Dudley H., and Frederick D. Bullock. 1919. “THE

IMPORTANCE OF THE SPLEEN IN RESISTANCE

TO INFECTION.” Annals of Surgery 70 (5): 513.

Newland, Adrian, Drew Provan, and Steven Myint. 2005.

“Preventing Severe Infection after Splenectomy.” BMJ

331 (7514): 417–18.

Qiu, Junfei, Qihui Wu, Guoru Ding, Yuhua Xu, and Shuo

Feng. 2016. “A Survey of Machine Learning for Big

Data Processing.” EURASIP Journal on Advances in

Signal Processing 2016 (1): 1–16.

Vanus, Jan, Jan Kubicek, Ojan M. Gorjani, and Jiri

Koziorek. 2019. “Using the IBM SPSS SW Tool with

Wavelet Transformation for CO2 Prediction within IoT

in Smart Home Care.” Sensors 19 (6): 1407.

Wang, Y. Q., L. Lü, S. P. Weng, J. N. Huang, S-M Chan,

and J. G. He. 2006. “Molecular Epidemiology and

Phylogenetic Analysis of a Marine Fish Infectious

Spleen and Kidney Necrosis Virus-like (ISKNV-Like)

Virus.” Archives of Virology 152 (4): 763–73.

A Spleen Infection Recognition Approach Using Shallow Neural Network in Comparison with Support Vector Machine

485

You, Zhen, Qinghe Zhang, Changjun Liu, Jiuzhou Song,

Ning Yang, and Ling Lian. 2019. “Integrated Analysis

of lncRNA and mRNA Repertoires in Marek’s Disease

Infected Spleens Identifies Genes Relevant to

Resistance.” BMC Genomics 20 (1): 1–15.

Zerem, Enver, and Jacob Bergsland. 2006. “Ultrasound

Guided Percutaneous Treatment for Splenic Abscesses:

The Significance in Treatment of Critically Ill

Patients.” World Journal of Gastroenterology: WJG 12

(45): 7341.

Boiy, Erik, and Marie-Francine Moens. 2008. “A Machine

Learning Approach to Sentiment Analysis in

Multilingual Web Texts.” Information Retrieval 12 (5):

526–58.

Chen, Hongbo, Changchun Li, Mingdi Fang, Mengjin Zhu,

Xinyun Li, Rui Zhou, Kui Li, and Shuhong Zhao. 2009.

“Understanding Haemophilus Parasuis Infection in

Porcine Spleen through a Transcriptomics Approach.”

BMC Genomics 10 (1): 1–18.

Darling, Samuel T. 1906. “A Protozoön General Infection

Producing Pseudotubercles in The Lungs and Focal

Necroses In The Liver, Spleen And Lymphnodes.”

JAMA: The Journal of the American Medical

Association XLVI (17): 1283–85.

Davies, J. M., R. Barnes, and D. Milligan. 2002. “Update

of Guidelines for the Prevention and Treatment of

Infection in Patients with an Absent or Dysfunctional

Spleen.” Clinical Medicine 2 (5): 440.

Dietterich, Thomas G. 2002. “Machine Learning for

Sequential Data: A Review.” Structural, Syntactic, and

Statistical Pattern Recognition, 15–30.

Frank, Eibe, Mark Hall, Geoffrey Holmes, Richard Kirkby,

Bernhard Pfahringer, Ian H. Witten, and Len Trigg.

2009. “Weka-A Machine Learning Workbench for Data

Mining.” Data Mining and Knowledge Discovery

Handbook, 1269–77.

Goh, G. D., S. L. Sing, and W. Y. Yeong. 2020. “A Review

on Machine Learning in 3D Printing: Applications,

Potential, and Challenges.” Artificial Intelligence

Review 54 (1): 63–94.

Gómez Vela, Francisco A., Federico Divina, and Miguel

García-Torres. 2021. Computational Methods for the

Analysis of Genomic Data and Biological Processes.

MDPI.

Grimaldi, David, Olivier Pradier, Richard S. Hotchkiss, and

Jean-Louis Vincent. 2017. “Nivolumab plus Interferon-

γ in the Treatment of Intractable Mucormycosis.” The

Lancet Infectious Diseases 17 (1): 18.

G. Ramkumar, R. Thandaiah Prabu, Ngangbam Phalguni

Singh, U. Maheswaran, Experimental analysis of brain

tumor detection system using Machine learning

approach, Materials Today: Proceedings, 2021, ISSN

22147853, https://doi.org/10.1016/j.matpr.2021.01.246

Hosey, R. G., C. G. Mattacola, V. Kriss, T. Armsey, J. D.

Quarles, and J. Jagger. 2006. “Ultrasound Assessment

of Spleen Size in Collegiate Athletes.” British Journal

of Sports Medicine 40 (3): 251–54.

Jung, Andreas, Reinhard Maier, Jean-Pierre Vartanian,

Gennady Bocharov, Volker Jung, Ulrike Fischer,

Eckart Meese, Simon Wain-Hobson, and Andreas

Meyerhans. 2002. “Multiply Infected Spleen Cells in

HIV Patients.” Nature 418 (6894): 144–144.

King, Harold, and Harris B. Shumacker Jr. 1952. “Splenic

Studies: I. Susceptibility to Infection after Splenectomy

Performed in Infancy.” Annals of Surgery 136 (2): 239.

Kumar M, M., Sivakumar, V. L., Devi V, S.,

Nagabhooshanam, N., & Thanappan, S. (2022).

Investigation on Durability Behavior of Fiber

Reinforced Concrete with Steel Slag/Bacteria beneath

Diverse Exposure Conditions. Advances in Materials

Science and Engineering, 2022.

Landen, Serge, and Jean Closset. 2007. “Gas-Producing

Infection of the Spleen in a Super-Super-Obese

Patient.” Obesity Surgery 17 (10): 1416–18.

Morris, Dudley H., and Frederick D. Bullock. 1919. “The

Importance of The Spleen in Resistance to Infection.”

Annals of Surgery 70 (5): 513.

Newland, Adrian, Drew Provan, and Steven Myint. 2005.

“Preventing Severe Infection after Splenectomy.” BMJ

331 (7514): 417–18.

Qiu, Junfei, Qihui Wu, Guoru Ding, Yuhua Xu, and Shuo

Feng. 2016. “A Survey of Machine Learning for Big

Data Processing.” EURASIP Journal on Advances in

Signal Processing 2016 (1): 1–16.

S. G and R. G, "Automated Breast Cancer Classification

based on Modified Deep learning Convolutional Neural

Network following Dual Segmentation," 2022 3rd

International Conference on Electronics and

Sustainable Communication Systems (ICESC),

Coimbatore, India, 2022, pp. 1562-1569, doi:

10.1109/ICESC54411.2022.9885299.

Vanus, Jan, Jan Kubicek, Ojan M. Gorjani, and Jiri

Koziorek. 2019. “Using the IBM SPSS SW Tool with

Wavelet Transformation for CO2 Prediction within IoT

in Smart Home Care.” Sensors 19 (6): 1407.

Wang, Y. Q., L. Lü, S. P. Weng, J. N. Huang, S-M Chan,

and J. G. He. 2006. “Molecular Epidemiology and

Phylogenetic Analysis of a Marine Fish Infectious

Spleen and Kidney Necrosis Virus-like (ISKNV-Like)

Virus.” Archives of Virology 152 (4): 763–73.

You, Zhen, Qinghe Zhang, Changjun Liu, Jiuzhou Song,

Ning Yang, and Ling Lian. 2019. “Integrated Analysis

of lncRNA and mRNA Repertoires in Marek’s Disease

Infected Spleens Identifies Genes Relevant to

Resistance.” BMC Genomics 20 (1): 1–15.

Zerem, Enver, and Jacob Bergsland. 2006. “Ultrasound

Guided Percutaneous Treatment for Splenic Abscesses:

The Significance in Treatment of Critically Ill

Patients.” World Journal of Gastroenterology: WJG 12

(45): 7341

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

486