Enhancing Hydatid Cyst Classification with Deep Learning and Convolutional Neural Networks Using CT Scans

Mohammad Nazir Akbari; Abed Azizi

doi:10.60141/AJID/V.2.I.1.2

Authors

Mohammad Nazir Akbari Kabul University https://orcid.org/0009-0009-7228-1749
Abed Azizi Kabul University https://orcid.org/0009-0009-2334-5793

DOI:

https://doi.org/10.60141/AJID/V.2.I.1.2

Keywords:

Artificial Intelligence, Deep learning, CT scan, Hydatid Cysts, CNN, Neural Networks

Abstract

Background: Hydatid cysts, caused by Echinococcus granulosis, are a serious health concern with potential complications. Traditional diagnostic methods, like clinical examination and imaging interpretation, can be subjective and error-prone. Artificial Intelligence and Deep Learning techniques can revolutionize healthcare by enhancing disease detection and diagnosis, with the study focusing on precise detection and classification.

Methods: A Convolutional Neural Network (CNN) model was developed, utilizing image preprocessing techniques to accurately classify hydatid cysts in Computed Tomography (CT) scans. Training relied on a curated dataset, enabling the model to learn and identify key patterns indicative of hydatid cyst presence and its stage detection in CT scan images.

Result: The AI model employed in this study achieved a 90% accuracy in classifying hydatid cyst stages using CT scan images. By providing essential information about the cyst stage, healthcare professionals can accurately inform patients based on CT scan analysis.

Conclusion: The study explores the use of AI and DL in hydatid cyst stage classification using a CNN model trained on CT scan images. The approach aims to reduce hydatid cyst growth rates by aiding in early detection, highlighting the significant transformation in the healthcare industry due to advancements in disease detection, diagnosis, and treatment.

Author Biographies

Mohammad Nazir Akbari, Kabul University

Department of Information Systems, Faculty of Computer Science, Kabul University, Kabul, Afghanistan

Abed Azizi, Kabul University

Department of Information Systems, Faculty of Computer Science, Kabul University, Kabul, Afghanistan.

References

Agudelo Higuita NI, Brunetti E, McCloskey C. Cystic echinococcosis. Journal of clinical microbiology. 2016;54(3):518-23.

Anwar SM, Majid M, Qayyum A, Awais M, Alnowami M, Khan MK. Medical image analysis using convolutional neural networks: a review. Journal of medical systems. 2018;42:1-13.

Kjossev KT, Losanoff JE. Classification of hydatid liver cysts. Journal of gastroenterology and hepatology. 2005;20(3):352-9.

F.Derbel. Hydatid Cysts of the Liver - Diagnosis, Complications and Treatment. InTech. Aug. 14, 2012.

Group WIW. International classification of ultrasound images in cystic echinococcosis for application in clinical and field epidemiological settings. Acta tropica. 2003;85(2):253-61.

LeCun Y, Bengio Y, Hinton G. Deep learning. nature. 2015;521(7553):436-44.

Hsieh J. Computed tomography: principles, design, artifacts, and recent advances. 2003.

wikipedia. [Available from: https://en.wikipedia.org/wiki/Kaggle.

Ahmad T, Aziz MN. Data preprocessing and feature selection for machine learning intrusion detection systems. ICIC Express Lett. 2019;13(2):93-101.

Saponara S, Elhanashi A, editors. Impact of Image Resizing on Deep Learning Detectors for Training Time and Model Performance2022; Cham: Springer International Publishing.

Solawetz J. roboflow [Internet]: roboflow. 2020. Available from: https://blog.roboflow.com/train-test-split/.

Chlap P, Min H, Vandenberg N, Dowling J, Holloway L, Haworth A. A review of medical image data augmentation techniques for deep learning applications. Journal of Medical Imaging and Radiation Oncology. 2021;65(5):545-63.

Belaid ON, Loudini M. Classification of brain tumor by combination of pre-trained vgg16 cnn. Journal of Information Technology Management. 2020;12(2):13-25.

ImagNet. [Available from: https://www.image-net.org/.

Jin J, Dundar A, Culurciello E. Flattened convolutional neural networks for feedforward acceleration. arXiv preprint arXiv:14125474. 2014.

Bjorck N, Gomes CP, Selman B, Weinberger KQ. Understanding batch normalization. Advances in neural information processing systems. 2018;31.

17. Garbin C, Zhu X, Marques O. Dropout vs. batch normalization: an empirical study of their impact to deep learning. Multimedia Tools and Applications. 2020;79:12777-815.

Wang M, Lu S, Zhu D, Lin J, Wang Z, editors. A high-speed and low-complexity architecture for softmax function in deep learning. 2018 IEEE asia pacific conference on circuits and systems (APCCAS); 2018: IEEE.

1Sagar R. What Does Freezing A Layer Mean And How Does It Help In Fine Tuning Neural Networks. analytics India Magazine. 2019 May 25, 2019.

Zhao H, Gallo O, Frosio I, Kautz J. Loss functions for neural networks for image processing. arXiv preprint arXiv:151108861. 2015.

Chauhan NS. KD nuggets [Internet]2020. Available from: https://www.kdnuggets.com/2020/05/model-evaluation-metrics-machine-learning.html.

Fawcett T. An introduction to ROC analysis. Pattern recognition letters. 2006;27(8):861-74.