Модели и методы искусственного интеллекта в обработке медицинских изображений высокого качества / Models and Methods of Artificial Intelligence of High-Quality Medical Images Processing тема диссертации и автореферата по ВАК РФ 00.00.00, кандидат наук Ал-Аззави Зобеда Хатиф Наджи

• Ал-Аззави Зобеда Хатиф Наджи
• кандидат наук
• 2025

Оглавление диссертации кандидат наук Ал-Аззави Зобеда Хатиф Наджи

CONTENTS

Introduction

Chapter 1. Methods and analysis of Image processing

1.1. Digital image

1.2. Synthesis, and technical sciences of Image Noise

1.3. Methods of type filter

1.3.1. Gaussian filter:

1.3.2. Mean Filter

1.3.3. Median filter

1.4. Analysis Steps and Process in Image Processing

1.4.1. Image Acquisition

1.4.2. Image Restoration

1.4.3. Feature and, Pattern Extraction

1.4.4. Image Enhancement

1.5. Reliability and safety of Image Enhancement

1.6. Theory of reliability of the use of Image segmentation

1.7. Theory of reliability of the use of Image compression

1.8. Theory of reliability of the use of Image Recognition and Classification

1.9. Motivation example of image processing

1.10. Conclusion

Chapter 2. Models of Artificial intelligence and Deep learning

2.1. Introduction

2.2. Algorithm and models of Deep learning

2.3. Fundamental Terminology related to deep learning

2.4. Models of Neural Network NN

2.4.1. Perceptron NN

2.4.2. Feed forward NN

2.4.3. Radial basis function NN

2.4.4 Kohonen Self Organizing NN

2.4.5. Recurrent NN

2.4.6. Convolutional NN

2.5. Standard Structure of convolution neural network Models

2.5.1. Convolution layers

2.5.2. Pooling layer

2.5.3. A fully connected layer:

2.6. Models of CNN

2.6.1. LeNet

2.6.2. AlexNet

2.6.3. ZFNet

2.6.4. Google Net

2.6.5. Inception V2V3

2.6.6. Inception V4

2.6.7. VGGNet

2.6.8. ResNet

2.6.9. Channel boosted

2.6.10. MobileNet

2.7. Analysis of optimizer methods in deep learning model

2.8. The most common optimizer methods for deep learning:

2.9. Function of optimizer in deep learning

2.10. Challenge of Deep Learning

2.11. Result and analysis

2.12. Conclusion

Chapter 3. Solve problem classification-based algorithm in deep learning

3.1. Introduction

3.2. Related work

3.3. Methodology and Information theory

3.3.1. Dataset

3.3.2. Pe-processing

3.3.3. pre-trained models

3.3.4. Transfer learning

3.3.5. Training Vgg19

3.3.6. Training Inceptionv3

3.3.7. Training Resnet101

3.3.8. Stacking Ensemble Learning

3.4. Results

Chapter 4 -Methods and algorithm for medical image segmentation

4.1. Introduction

4.2. Related work

4.3. Models of information process segmentation approach

4.3.1. V-net model

4.3.2. U-net model

4.3.3. Attention 2D-layer

4.4. Proposed Algorithm

4.5 Evaluation criteria

4.6. Descriptive for Data Set

4.7. Result and discussion

4.8. Conclusion:

Conclusion

Notation

Refences

List of figures

List of tables

Appendix

Введение диссертации (часть автореферата) на тему «Модели и методы искусственного интеллекта в обработке медицинских изображений высокого качества / Models and Methods of Artificial Intelligence of High-Quality Medical Images Processing»

Introduction

Relevance of the research topic. Developments in deep learning technologies, which are mostly focused on image data sets for segmentation and classification, have significantly affected research medicine and model training infrastructure globally. In the interests of medical image for subsequent work in hospitals and research enterprises that determine the development of the diagnosis of disease and progress in general in all countries of the world, the rapid growth of deep learning technologies creates new conditions and imposes new requirements for the informatization and automation of learning and research processes.

The field of medical imaging has been greatly revolutionized by deep learning, providing novel approaches for the diagnosis, monitoring, and treatment of many disorders. Deep learning algorithms, namely convolutional neural networks (CNNs), are employed in medical imaging to evaluate and interpret intricate medical pictures such as X-rays, MRIs, CT scans, and ultrasounds.

These algorithms have exceptional proficiency in identifying patterns, rendering them exceptionally efficient for tasks such as tumor detection, organ segmentation, and disease classification. Deep learning models can be trained to accurately identify malignant tissues in brain, achieving a level of accuracy that is equivalent to or even beyond that of expert radiologists.

Generally, the application of medical image with deep learning techniques, models based on artificial intelligence, and algorithms in medical technology is widespread. In practical terms, a study must be conducted to specifically examine a particular type of disease in order to identify and define the characteristics of the medical image dataset that can be utilized to classify diseases using Artificial Intelligence algorithms.

Question of classification and segmentation approaches and models has been addressed to different extents by various authors and scientists, such as Zadeh Shirazi, Amin and his team, A.E Abulwafa. This research mainly describes the approaches and techniques of deep learning models.

The scope to which various authors and scientists have addressed the implementation of diagnoses disease approaches and deep learning models has varied, Artificial intelligence CNNs modeling for classification and segmentation task are reflected in the works of Russian and foreign scientists: Sergey Pavlov, and his team, Anna Kuzina and her team, Marina Pominova and her team, Naga Venkata Yaswanth Lankadasu with his team, Ekaterina Brui and her team, ZIJIAN WANG and his team, Zhang, Wenbo, and his team. On the other hand, other scientists have provided some other classification architectural model for the medical images to build strong frame worked of computer aided system to diagnose different disease with different medical data set, such as Visuha, Lara, and his team. Hammam, Ahmed A., and his team. Iram Shahzadi and her team.

The tasks of developing information classification models and procedures and assessing whether services could be rendered there are still relevant.

In comparison with the deep learning models developed by the scientists and researchers mentioned above, the modern technology used in this work, which developed and modified deep learning artificial intelligence models, has demonstrated its greater resilience to noise, durability, and error-correction capabilities; however, their approaches necessitate fewer computational resources than ours.

From the above, the relevance of the dissertation follows.

The aim of the work is to increase the efficiency of planning, designing and developing intelligent models have ability to recognize case patient with take in consideration accuracy, performance of models with different types of medical image. The development and optimization of automatic segmentation frameworks to separate anatomical landmarks, such as the tumor-related brain region.

The scientific task, that is solved in the dissertation of deep learning in medical imaging encompass the creation and implementation of sophisticated algorithms to address particular difficulties in the interpretation, analysis of medical images. The tasks encompass image segmentation relates to the accurate and detailed delineation of anatomical structures, this process is used to identify and outline malignancies in MRI, and Image classification relates to the process of categorizing medical photographs into predetermined groupings. This involves discriminating between benign and malignant lesions.

The object of research of is to study, analysis medical image processes, tuning different state-of art, develop stacking ensemble DL, modification segmentation DL models with MRI.

The subject of the research is the theoretical and practical foundations of analytical of image processing and AI algorithmic models for analyzing, classification, and segmentation, methods for identifying the case of tumor brain in MRI, provide insight developed models of CNN and methodologies, as well as highlighting the optimization methods used in the field of deep learning and clarifying which is better.

In order to accomplish this and resolve a scientific issue, a series of connected private scientific issues must be resolved first:

• Analysis and description of image processing domain of augmentation and generalization throughout diverse MRI data set.

• Methodological solutions, and analysis terminology information related to develop deep learning models and assess their performance, and optimizer methods.

• Models of Stacking ensemble deep learning.

• Models of Transfer learning.

• Investigation into architectures capable of handling and synthesizing data from various data types and modalities, such as CNN models.

• Examine techniques such as interpretable neural networks, and attention mechanisms that draw attention to the most important aspects of medical images.

• Models of segmentation medical image

Research methods. In the course of the research, image processing theory, operation of image processing, analysis of artificial intelligence technique, mathematical optimizer, classification deep learning models, theory segmentation and models of segmentation.

The scientific novelty of the results of the dissertation work is as follows

1. Describe different mathematical optimizer method which need it through enhancement training CNN models. Illustration different theory and information image processing like data augmentation technique and their impact on performance models.

2. An innovative approach utilizing multiscale sequential deep learning is suggested, constructed, and effectively applied for the simultaneous detection of brain tumors in MRI images. This is accomplished by extracting intricate data representations from MRI scans without the necessity of human oversight.

3. Develop of new frame work by merging various deep learning models, building an ensemble deep learning model that outperforms individual models and improves the performance of structure models. In this process multi-layer stacking is a more complex version of model stacking, involving multiple tiers of ensembles. In this approach, the result of one stacked model is used as the input for the next model. The hierarchical approach has the ability to capture intricate links and interactions between models, resulting in exceptional performance, especially in tasks that demand profound reasoning like tumor brain classification.

4. Development of a new segmentation method based on the most recent CNN architectures for the tumor brain MRI images. This is achieved by adapting V-net models by incorporating Attention mechanism, which is crucial for enabling the model to selectively focus on the relevant regions of the input, such as the area containing the tumor. Further these methods can assist in managing complexity by selecting processing the most crucial feature, hence lowering the computing burden and enhancing efficiency.

The reliability of the results, accuracy of the output of mathematical dependencies for calculating model parameters, the use of appropriate develop methods, the consistency between the results obtained through analytical calculations and modeling, and the alignment with known data published in domestic and foreign publications ensure the reliability of conclusions and recommendations.

Theoretical of the work. The theoretical point of the work centers around the establishment of a complete framework for image processing tools, encompassing the description and development of diverse deep learning models. This framework additionally encompasses the of a technique and algorithm for categorizing and dividing medical images at a sophisticated level.

The practical significance of the work, lies in the way the research's findings may be applied to planning, designing, and developing AI model procedures, which will greatly enhance the accuracy of MRI segmentation and classification. The findings support the advancement of synthesis model theory and application for early disease detection. The fact that the outcomes highlight the work's practical value enables the development of system-technical proposals for their implementation in medical sectors, like stacking ensemble and transfer learning, as well as the acquisition of interesting medical information with high accuracy.

The dissertation's findings can be applied to future studies and the creation of computer-aided systems for MRI classification and segmentation, and can be applied to research and development in the medical field and engineering solutions for the classification and segmentation of MRI. Additionally, they can be utilized in the creation of models for various medical image operations.

The feasibility study conducted in the dissertation helps in selecting the appropriate principles for structural and architectural models of deep learning. Furthermore, deep learning models have the capability to improve the resolution of MRI images, enabling clearer visualization of intricate details. This is crucial for precise diagnosis, particularly in the detection of small abnormalities.

Use and implementation of the work result. The developed methodology has worked as an engineer in the AL-Farah Clinic Centre. Baghdad. Iraq.

Presentations and validation of the research results. The main findings and contributions of the dissertation were presented and discussed at one conference:

• International Conference Engineering and Telecommunication (En&T), Dolgoprudny, Russian Federation, November 24-25, 2023.

Publication result

7 publications which were published, also reflects the work contribution.

7 articles were published on the topic of the dissertation. Of these, 3 articles were indexed in Scopus, and 2 papers were published in international journal, 2 articles were published in conference.

Personal contribution of the author.

The applicant personally receives any scientific results that are submitted. The author actively participated in the planning, execution, analysis, and discussion of the work, as well as in the preparation of publications.

The applicant play an integral role in co-authored articles by initiating the problem, devising a methodology for its resolution, executing a computing experiment, and reporting the achieved outcomes.

Provisions submitted for defense.

• Developed stacking ensemble by merging three CNN models to improve performance for classification MRI.

• Analysis different mathematical optimizer methods.

• Enhance Transfer learning models by performance-based adjustments to additional hyperparameters and the number of fine-tuned layers.

• Modification and develop pre-train segmentation models by adding attention layer to increase performance models, decrease cost computation and enhance accuracy with ahigh resolution of medical image, this can be led to more powerful models.

Work Structure

The dissertation consists of an introduction, 4 chapters, analytical results, a conclusion, a list of figures, a list of references, and tables. The main part is described in 124pages.

Chapter 1. This chapter's goals provide a thorough understanding of images, assess and create a number of image-enhancing strategies that can be used in a range of medical contexts.

Chapter 2. In this chapter analysis and explain terminology of DL models and described different optimizer mathematic methods and CNN models.

Chapter3.In this chapter explain and design stacking ensemble and tuning Transfer learning.

Chapter4. In this chapter, information and analysis various deep learning techniques for medical image segmentation, including the V-net, U-net models and attention methods

Conclusion. Summarizes the key findings that bring the work in this thesis to a close and highly gets potential future research avenues that might be investigated considering this work.

I give thanks to Allah for guiding me on this trip, as well as the process of carrying out my research and finishing my dissertation.

I would want to express my gratitude to my supervisor, Nazarov Alexey Nikolaevich, for giving me such a great topic to work. Thank you for being available and for your positive outlook.

Lastly, I want to express my gratitude to my amazing family and friends for their help over the years. I couldn't have made it through all of the challenges without my mother, father, husband, kids and my friends.

Заключение диссертации по теме «Другие cпециальности», Ал-Аззави Зобеда Хатиф Наджи

Conclusion

The main conclusion of the dissertation research was as follows.

1. Analysis image processing includes methods for enhancing visual quality, enabling certain functions like object recognition, and extracting relevant information from images. It is essential to several fields, including as digital photography, computer vision, surveillance, and medical imaging. Image processing is a key component of modern technology because of the tremendous improvements in efficiency and accuracy that have been made possible by advances in algorithms and computational capacity. It is anticipated that as the discipline develops, its applications will grow even more, spurring innovation in both theoretical and applied research.

2. The proposed analytical relations that enable the classification of medical images are based on the generated models. The Stacking ensemble combining the advantages of several machine learning models to create a robust method, stacking ensemble models have shown to be extremely effective in the classification of brain tumors from MRIs. The ensemble model may detect and classify tumors more accurately and reliably by stacking different classifiers, such as convolutional neural network (CNN) models, to capture distinct patterns and nuances within the MRI data. In the domain of medical diagnostics, where accuracy can have a direct impact on patient outcomes, this technique reduces the shortcomings of individual models, leading to improved generalization and more accurate predictions.

3. Transfer learning has emerged as a key strategy for medical image classification, providing a response to the problem of sparsely available labeled data in the medical domain. Transfer learning makes it possible to apply information from general domains to particular medical challenges by utilizing pre-trained models on big datasets. This method greatly increases the efficiency and accuracy of classifying medical images, allowing for quicker and more accurate diagnosis.

4. A scientific problem statement is developed, and the dissertation addresses its solution. Through the application of an attention mechanism with V-net model, the model is able to selectively focus on the areas of the image that are most important, enhancing segmentation accuracy by highlighting important structures and reducing background information that is not significant. More accurate and dependable segmentations are produced by this combination, which is essential for medical diagnosis and therapy planning. The use of attention layers enhances the model's resilience to fluctuations in the data, hence augmenting its suitability for diverse imaging modalities and patient demographics.

Список литературы диссертационного исследования кандидат наук Ал-Аззави Зобеда Хатиф Наджи, 2025 год

Refences

[1] Sakhare, V. C., S. V. Kamble, and J. A. Alase. "Noise models in digital image processing." International Advanced Research Journal in Science, Engineering and Technology 7.12 (2020): 10-17148.

[2] Naji, Zobeda Hatif, and A. N. Nazarov. "Medical image enhancement based AI techniques: a review." Synchroinfo Journal 8.2 (2022): 19-23.

[3] Young, Ian T., Jan J. Gerbrands, and Lucas J. Van Vliet. Fundamentals of image processing. Vol. 841. Delft: Delft University of Technology, 1998.

[4] Zobeda Hatif Naji Al-Azzwi, Alexey.N.N. "Medical image Classification using Transfer Learning: Convolutional Neural Network Models approach. " J. Electr. Syst., vol. 20, no. 6s, pp. 2561-2569, 2024, doi: 10.52783/jes.3243.

[5] Sugimori, Hiroyuki. "Classification of computed tomography images in different slice positions using deep learning." Journal of healthcare engineering 2018.1 (2018): 1753480.

[6] Al-Azzwi, Zobeda Hatif Naji, and A. N. Nazarov. "Brain tumor classification based on improved stacked ensemble deep learning methods." Asian Pacific journal of cancer prevention: APJCP 24.6 (2023): 2141.

[7] Magud, Ognjen, E. V. A. Tuba, and Nebojsa Bacanin. "Medical ultrasound image speckle noise reduction by adaptive median filter." Wseas Trans. Biol. Biomed 14 (2017): 38-46.

[8] Liu, Juan, et al. "Artificial intelligence-based image enhancement in PET imaging: noise reduction and resolution enhancement." PET clinics 16.4 (2021): 553.

[9] Rani, Neetu. "Image processing techniques: a review." Journal on Today's Ideas-Tomorrow's Technologies 5.1 (2017): 40-49.

[10] Guzman-Cabrera, Rafael, et al. "Digital image processing technique for breast cancer

detection." International Journal of Thermophysics 34 (2013): 1519-1531.

[11] Caridade, C. M. R. "Applications (ideas) in linear algebra with digital image processing. Can we do, teach, motivate and evaluate." Journal of Information Systems Engineering & Management 4.4 (2019): 1-10.

[12] Tyagi, Vipin. Understanding digital image processing. CRC Press, 2018.

[13] Owotogbe, J. S., T. S. Ibiyemi, and B. A. Adu. "A comprehensive review on various types of noise in image processing." International Journal of Scientific & Engineering Research 10.11 (2019): 388-393.

[14] Hussein, Taha, Hoger K. Omar, and Kamal Jihad. "A study on image noise and various image denoising techniques." ResearchJet Journal of Analysis and Inventions 2.11 (2021): 27-44.

[15] Verma, Rohit, and Jahid Ali. "A comparative study of various types of image noise and efficient noise removal techniques." International Journal of advanced research in computer science and software engineering 3.10 (2013).

[16] Hien, Nguyen Ngoc, et al. "Image noise removal method based on thresholding and regularization techniques." IEEE Access 10 (2022): 71584-71597.

[17] Ishitani, Tohru, and Mitsugu Sato. "Contrast-to-gradient method for the evaluation of image resolution taking account of random noise in scanning electron microscopy." Microscopy 53.3 (2004): 245-255.

[18] Kumar, Suresh, et al. "Performance comparison of median and wiener filter in image de-noising." International Journal of Computer Applications 12.4 (2010): 27-31.

[19] Chithra, K., and T. Santhanam. "A novel denoising technique for mixed noise removal from grayscale and color images." Journal of Theoretical & Applied Information Technology 96.3 (2018).

[20] Mishra, Vikas Kumar, Shobhit Kumar, and Neeraj Shukla. "Image acquisition and techniques to perform image acquisition." SAMRIDDHI: A Journal of Physical

Sciences, Engineering and Technology 9.01 (2017): 21-24.

[21] Gonzalez, Rafael C. Digital image processing. Pearson education india, 2009.

[22] Rani, Shilpa, Sonika Jindal, and Bhavneet Kaur. "A brief review on image restoration techniques." International Journal of Computer Applications 150.12 (2016): 30-33.

[23] Dara, Suresh, et al. "Feature extraction in medical images by using deep learning approach." International Journal of Pure and Applied Mathematics 120.6 (2018): 305-312.

[24] Li, Yinhao, Yutaro Iwamoto, and Yen-Wei Chen. "Medical Image Enhancement

Using Deep Learning." Deep Learning in Healthcare: Paradigms and Applications. Cham: Springer International Publishing, 2019. 53-76.

[25] Zhou, Abel, Qi Tan, and Rob Davidson. "Image enhancement using convolutional neural network." 2020 International Conference on Image, Video Processing and Artificial Intelligence. Vol. 11584. SPIE, 2020.

[26] Lore, Kin Gwn, Adedotun Akintayo, and Soumik Sarkar. "LLNet: A deep autoencoder approach to natural low-light image enhancement." Pattern Recognition 61 (2017): 650-662.

[27] Z. H. N. Al-azzwi and A. N. Nazarov."MRI Semantic Segmentation based on Optimize V-net with 2D Attention." Adv. Sci. Technol. Eng. Syst. J., vol. 8, no. 4, pp. 73-80, 2023.

[28] IBRAHEEM M.K.I., AL-KHAFAJI I.M., AL-AZZAWI Z.H. "A GENETIC ALGORITHIM-BASED INTRA CODING ALGORITHM FOR H.266/VVC", https://www.elibrary.ru/item.asp,id=65640761,2023.

[29] Samra, H. S. "Image compression techniques." International Journal of Computers &

Technology 2.2 (2012).

[30] Ibraheem, Murooj Khalid Ibraheem, Al-khafaji Israa M. Abdalameer, and Al-azzawi Zobeda Hatif Naji. "A genetic approach-based intra coding algorithm for H.

266/VVC." Информатика и автоматизация 23.3 (2024): 801-830.

[31] Menze, Bjoern H., et al. "The multimodal brain tumor image segmentation benchmark (BRATS)." IEEE transactions on medical imaging 34.10 (2014): 19932024.

[32] Ballestar, Laura Mora, and Veronica Vilaplana. "MRI brain tumor segmentation and uncertainty estimation using 3D-UNet architectures." International MICCAI Brainlesion Workshop. Cham: Springer International Publishing, 2020.

[33] Ronneberger, Olaf, Philipp Fischer, and Thomas Brox. "U-net: Convolutional networks for biomedical image segmentation." Medical image computing and computer-assisted intervention-MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18. Springer international publishing, 2015.

[34] Kumar, Vinod, and Kanwal Garg. "Neural Network Based Approach for Detection of Abnormal Regions of Lung Cancer in X-Ray Image." International Journal of Engineering Research & Technology, ISSN (2012): 2278-0181.

[35] Latif, Ghazanfar. "DeepTumor: framework for Brain MR image classification, Segmentation and Tumor Detection." Diagnostics 12.11 (2022): 2888.

[36] Almalki, Yassir Edrees, et al. "Isolated convolutional-neural-network-based deep-feature extraction for brain tumor classification using shallow classifier." Diagnostics 12.8 (2022): 1793.

[37] Zadeh Shirazi, Amin, et al. "The application of deep convolutional neural networks to brain cancer images: a survey." Journal of personalized medicine 10.4 (2020): 224.

[38] Atiyah, Assalah Zaki, and Khawla Hussein Ali. "Brain mri images segmentation based on u-net architecture." IJEEE Journal 18.1 (2021): 21-27.

[39] Deshayes, Michel, and Pierre Maurel. "Remote sensing imagery and its contents." Options Méditerranéennes (1990): 21-30.

[40] Joutou, Taichi, and Keiji Yanai. "A food image recognition system with multiple

kernel learning." 2009 16th IEEE International Conference on Image Processing (ICIP). IEEE, 2009.

[41] Guan, Ling, ed. Multimedia image and video processing. CRC press, 2017.

[42] Ciora, Radu Adrian, and Carmen Mihaela Simion. "Industrial applications of image processing." Acta Universitatis Cibiniensis. Technical Series 64.1 (2014): 17-21.

[43] Lin, Chenjie, et al. "Estimation of cultivated land quality based on soil hyperspectral data." Agriculture 12.1 (2022): 93.

[44] AL-AZZAWI Z.H.N, NAZAROV A.N, IBRAHEEM M. K. I. "Description of the Optimization Method and Challenge of Deep Learning." 2023. https://www.elibrary.ru/item.asp?id=65640795&pff=1

[45] Torma, Markus. "Kohonen self-organizing feature map and its use in clustering." ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision. Vol. 2357. SPIE, 1994.

[46] Abulwafa, Arwa E. "A survey of deep learning algorithms and its applications." Nile Journal of Communication and Computer Science 3.1 (2022): 28-49.

[47] Deekshith Shetty, H. C., et al. "Diving deep into deep learning: history, evolution, types and applications." International Journal of Innovative Technology and Exploring Engineering 9.3 (2020): 2278-3075.

[48] Riad, Rachid, et al. "Learning strides in convolutional neural networks." arXiv preprint arXiv:2202.01653 (2022).

[49] Mouton, Coenraad, Johannes C. Myburgh, and Marelie H. Davel. "Stride and translation invariance in CNNs." Southern African Conference for Artificial Intelligence Research. Cham: Springer International Publishing, 2020.

[50] Agarap, Abien Fred. "Deep learning using rectified linear units (relu)." arXiv preprint arXiv:1803.08375 (2018).

[51] Zhang, Xin, et al. "Research on scene classification method of high-resolution remote sensing images based on RFPNet." Applied Sciences 9.10 (2019): 2028.

[52] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks. " Advances in neural information processing systems 25 (2012).

[53] Jabbar, H., and Rafiqul Zaman Khan. "Methods to avoid over-fitting and under-fitting in supervised machine learning (comparative study)." Computer science, communication and instrumentation devices 70.10.3850 (2015): 978-981.

[54] Daday, Mark Jovic A., Arnel C. Fajardo, and Ruji P. Medina. "Enhancing FeedForward Neural Network in Image Classification." Proceedings of the 2nd International Conference on Computing and Big Data. 2019.

[55] C. A. S. Farias, A. M. G. Louren?o, T. C. Carneiro, and T. H. C. Nunes, "A Monthly Rainfall-Runoff Model using Radial Basis Function Neural Networks for Pianco River Basin , Brazil," April, 2019.

[56] Mann, James, and Sheldon Gilbert. "An analog self-organizing neural network chip." Advances in neural information processing systems 1 (1988).

[57] LeCun, Yann, et al. "Gradient-based learning applied to document recognition." Proceedings of the IEEE 86.11 (1998): 2278-2324.

[58] Zeiler, Matthew D., and Rob Fergus. "Visualizing and understanding convolutional networks." Computer Vision-ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part I 13. Springer International Publishing, 2014.

[59] Szegedy, Christian, et al. "Going deeper with convolutions." Proceedings of the IEEE conference on computer vision and pattern recognition. 2015.

[60] Szegedy, Christian, et al. "Rethinking the inception architecture for computer vision."

Proceedings of the IEEE conference on computer vision and pattern recognition.

2016.

[61] Szegedy, Christian, et al. "Inception-v4, inception-resnet and the impact of residual connections on learning." Proceedings of the AAAI conference on artificial intelligence. Vol. 31. No. 1. 2017.

[62] Simonyan, Karen, and Andrew Zisserman. "Very deep convolutional networks for large-scale image recognition." arXiv preprint arXiv:1409.1556 (2014).

[63] Wu, Songtao, Shenghua Zhong, and Yan Liu. "Deep residual learning for image steganalysis." Multimedia tools and applications 77 (2018): 10437-10453.

[64] Tan, Mingxing, and Quoc Le. "Efficientnet: Rethinking model scaling for convolutional neural networks." International conference on machine learning. PMLR, 2019.

[65] Khan, Asifullah, Anabia Sohail, and Amna Ali. "A new channel boosted convolutional neural network using transfer learning." arXiv preprint arXiv:1804.08528 (2018).

[66] Shen, Hua, et al. "L-Net: A lightweight convolutional neural network for devices with low computing power." Information Sciences 660 (2024): 120131.

[67] Wei, Ruoqi, and Ausif Mahmood. "Optimizing few-shot learning based on variational autoencoders." Entropy 23.11 (2021): 1390.

[68] Phong, Nguyen Huu, and Bernardete Ribeiro. "Rethinking recurrent neural networks and other improvements for image classification." arXiv preprint arXiv:2007.15161 (2020).

[69] He, Ran, Jianxing Wang, and Zhengning Zhang. "A Novel Aircraft Refueling Behavior Detection Model based on Deep Learning." Journal of Physics: Conference Series. Vol. 1732. No. 1. IOP Publishing, 2021.

[70] Zhao, Zhenhao. "Skin cancer classification based on convolutional neural networks and vision transformers." Journal of Physics: Conference Series. Vol. 2405. No. 1.

IOP Publishing, 2022.

[71] Yang, Zihan. "Classification of picture art style based on VGGNET." Journal of Physics: Conference Series. Vol. 1774. No. 1. IOP Publishing, 2021.

[72] Anand, R., et al. "Modified Vgg deep learning architecture for Covid-19 classification using bio-medical images." IOP conference series: materials science and engineering. Vol. 1084. No. 1. IOP Publishing, 2021.

[73] Mateen, Muhammad, et al. "Fundus image classification using VGG-19 architecture with PCA and SVD." Symmetry 11.1 (2018): 1.

[74] Islam, Shazzadul, et al. "Bird species classification from an image using VGG-16 network." Proceedings of the 7th international conference on computer and communications management. 2019.

[75] Rodrigues Filho, Marcelo Luis, and Omar Andres Carmona Cortes. "Efficient Breast Cancer Classification Using Histopathological Images and a Simple VGG." Revista de Informática Teórica e Aplicada 29.1 (2022): 102-114.

[76] Tammina, Srikanth. "Transfer learning using vgg-16 with deep convolutional neural network for classifying images." International Journal of Scientific and Research Publications (IJSRP) 9.10 (2019): 143-150.

[77] He, Kaiming, et al. "Deep residual learning for image recognition." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.

[78] Howard, Andrew G., et al. "Mobilenets: Efficient convolutional neural networks for mobile vision applications." arXiv preprint arXiv:1704.04861 (2017).

[79] Yadav, Kaustubh. "A Comprehensive Study on Optimization Strategies for Gradient Descent In Deep Learning." arXiv preprint arXiv:2101.02397 (2021).

[80] Lu, Jun. "Gradient descent, stochastic optimization, and other tales." arXiv preprint arXiv:2205.00832 (2022).

[81] Mehmood, Faisal, Shabir Ahmad, and Taeg Keun Whangbo. "An efficient

optimization technique for training deep neural networks." Mathematics 11.6 (2023): 1360.

[82] Shulman, David. "Optimization methods in deep learning: A comprehensive overview." arXiv preprint arXiv:2302.09566 (2023).

[83] Zaheer, Raniah, and Humera Shaziya. "A study of the optimization algorithms in deep learning." 2019 third international conference on inventive systems and control (ICISC). IEEE, 2019.

[84] Soydaner, Derya. "A comparison of optimization algorithms for deep learning." International Journal of Pattern Recognition and Artificial Intelligence 34.13 (2020): 2052013.

[85] Hamedon, Syasya Nadhirah, Juliana Johari, and Fazlina Ahmat Ruslan. "Optimizing photovoltaic output performance prediction: a deep learning approach with LSTM neural networks and Adam optimizer." Journal of Electrical and Electronic Systems Research (JEESR) 25.1 (2024): 89-98.

[86] Sun, Shiliang, et al. "A survey of optimization methods from a machine learning perspective." IEEE transactions on cybernetics 50.8 (2019): 3668-3681.

[87] Rum, Siti Nurulain Mohd, and Fariz Az Zuhri Nawawi. "FishDeTec: a fish identification application using image recognition approach." International Journal of Advanced Computer Science and Applications 12.3 (2021).

[88] Akinyelu, Andronicus A., et al. "Brain tumor diagnosis using machine learning, convolutional neural networks, capsule neural networks and vision transformers, applied to MRI: a survey." Journal of imaging 8.8 (2022): 205.

[89] Futrega, Michal, et al. "Optimized U-Net for brain tumor segmentation." International MICCAI brainlesion workshop. Cham: Springer International Publishing, 2021.

[90] Nguyen, Dat Tien, Jiho Choi, and Kang Ryoung Park. "Thyroid nodule segmentation

in ultrasound image based on information fusion of suggestion and enhancement networks." Mathematics 10.19 (2022): 3484.

[91] Jaworek-Korjakowska, Joanna, Pawel Kleczek, and Marek Gorgon. "Melanoma thickness prediction based on convolutional neural network with VGG-19 model transfer learning." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2019.

[92] Filatov, Dmytro, and Ghulam Nabi Ahmad Hassan Yar. "Brain tumor diagnosis and classification via pre-trained convolutional neural networks." arXiv preprint arXiv:2208.00768 (2022).

[93] Khan, Muhammad Attique, et al. "Multimodal brain tumor classification using deep learning and robust feature selection: A machine learning application for radiologists." Diagnostics 10.8 (2020): 565.

[94] Pominova, Marina, et al. "Voxelwise 3d convolutional and recurrent neural networks for epilepsy and depression diagnostics from structural and functional mri

data." 2018 IEEE International Conference on Data Mining Workshops (ICDMW). IEEE, 2018.

[95] Muthukrishnan, Ranjan, and Miyilsamy Radha. "Edge detection techniques for image segmentation." International Journal of Computer Science & Information Technology 3.6 (2011): 259.

[96] Miki, Yuma, et al. "Classification of teeth in cone-beam CT using deep convolutional neural network." Computers in biology and medicine 80 (2017): 24-29.

[97] Choras, Ryszard S. "Feature extraction for CBIR and biometrics applications." 7th WSEAS International Conference on Applied Computer Science. Vol. 7. 2007.

[98] Pavlov, Sergey, et al. "Weakly supervised fine tuning approach for brain tumor segmentation problem." 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA). IEEE, 2019.

[99] Saraiva, Arata Andrade, et al. "Classification of Images of Childhood Pneumonia using Convolutional Neural Networks." Bioimaging. 2019.

[100] Biswas, Shreya, et al. "Prediction of COVID-19 from chest CT images using an ensemble of deep learning models." Applied Sciences 11.15 (2021): 7004.

[101] Saha, Oishy, et al. "A multi-model based ensembling approach to detect COVID-19 from chest X-ray images." 2020 IEEE REGION 10 CONFERENCE (TENCON). IEEE, 2020.

[102] El-Sappagh, Shaker, et al. "Alzheimer's disease progression detection model based on an early fusion of cost-effective multimodal data." Future Generation Computer Systems 115 (2021): 680-699.

[103] Adnan, Niha, et al. "Multi-model deep learning approach for segmentation of teeth and periapical lesions on pantomographs." Oral surgery, oral medicine, oral pathology and oral radiology 138.1 (2024): 196-204.

[104] Lankadasu, Naga Venkata Yaswanth, et al. "Skin Cancer Classification Using a Convolutional Neural Network: An Exploration into Deep Learning." 2024 ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems (ICETSIS). IEEE, 2024.

[105] Mabrouk, Alhassan, et al. "Medical image classification using transfer learning and chaos game optimization on the internet of medical things." Computational Intelligence and Neuroscience 2022.1 (2022): 9112634.

[106] Sharma, Pavan, Hemant Amhia, and Sunil Datt Sharma. "Performance analysis of pre-trained transfer learning models for the classification of the rolling bearing faults." Journal of Physics: Conference Series. Vol. 2070. No. 1. IOP Publishing, 2021.

[107] Mohammed, Mazin Abed, Abdulrahman Abbas Mukhlif, and Belal Al Khateeb. "Breast cancer images classification using a new transfer learning technique." Iraqi

Journal for Computer Science and Mathematics 4.1 (2023): 16.

[108] Desai, Chitra. "Image classification using transfer learning and deep learning." International Journal of Engineering and Computer Science 10.9 (2021): 2539425398.

[109] Lebedev, Georgy, et al. "Application of artificial intelligence methods to recognize pathologies on photographs of fundus." Procedia Computer Science 176 (2020): 1823-1828.

[110] Sudha, V., and T. R. Ganeshbabu. "A Convolutional Neural Network Classifier VGG-19 Architecture for Lesion Detection and Grading in Diabetic Retinopathy Based on Deep Learning." Computers, Materials & Continua 66.1 (2021).

[111] Visuna, Lara, et al. "Computer-aided diagnostic for classifying chest X-ray images using deep ensemble learning." BMC Medical Imaging 22.1 (2022): 178.

[112] Sarwinda, Devvi, et al. "Deep learning in image classification using residual network (ResNet) variants for detection of colorectal cancer." Procedia Computer Science 179 (2021): 423-431.

[113] El Asnaoui, Khalid. "Design ensemble deep learning model for pneumonia disease classification." International Journal of Multimedia Information Retrieval 10.1 (2021): 55-68.

[114] Hammam, Ahmed A., Haytham H. Elmousalami, and Aboul Ella Hassanien. "Stacking deep learning for early COVID-19 vision diagnosis." Big data analytics and artificial intelligence against COVID-19: innovation vision and approach. Springer, Cham, 2020. 297-307.

[115] El Hamdaoui, Halima, et al. "High precision brain tumor classification model based on deep transfer learning and stacking concepts." Indones. J. Electr. Eng. Comput. Sci 24.1 (2021): 167-177.

[116] Azzi, Yamina, Abdelouahab Moussaoui, and Mohand-Tahar Kechadi. "Semantic

segmentation of medical images with deep learning: Overview." Medical Technologies Journal 4.3 (2020): 568-575.

[117] Moorthy, Jayashree, and Usha Devi Gandhi. "A Survey on medical image segmentation based on deep learning techniques." Big Data and Cognitive Computing 6.4 (2022): 117.

[118] Mohimont, Lucas, et al. "Comparison of machine learning and deep learning methods for grape cluster segmentation." International Conference on Smart and Sustainable Agriculture. Cham: Springer International Publishing, 2021.

[119] Jung, Sunguk, et al. "Benchmarking deep learning models for instance segmentation." Applied Sciences 12.17 (2022): 8856.

[120] Scherr, Tim, et al. Best practices in deep learning-based segmentation of microscopy images. KIT Scientific Publishing, 2018.

[121] Kirillov, Alexander, et al. "Panoptic segmentation." Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019.

[122] Elharrouss, Omar, et al. "Panoptic segmentation: A review." arXiv preprint arXiv:2111.10250 (2021).

[123] Khan, Akib Mohammed, et al. "Attresdu-net: medical image segmentation using attention-based residual double u-net." 2023 International Joint Conference on Neural Networks (IJCNN). IEEE, 2023.

[124] Al Nasim, Md Abdullah, et al. "Brain tumor segmentation using enhanced u-net model with empirical analysis." 2022 25th International Conference on Computer and Information Technology (ICCIT). IEEE, 2022.

[125] Guan, Xi, et al. "3D AGSE-VNet: an automatic brain tumor MRI data segmentation framework." BMC medical imaging 22 (2022): 1-18

[126] Zhang, Wenbo, et al. "ME-Net: multi-encoder net framework for brain tumor segmentation." International Journal of Imaging Systems and Technology 31.4

(2021): 1834-1848.

[127] Milletari, Fausto, Nassir Navab, and Seyed-Ahmad Ahmadi. "V-net: Fully convolutional neural networks for volumetric medical image segmentation." 2016 fourth international conference on 3D vision (3DV). Ieee, 2016.

[128] Bahdanau, Dzmitry, Kyunghyun Cho, and Yoshua Bengio. "Neural machine translation by jointly learning to align and translate." arXiv preprint arXiv: 1409.0473 (2014).

[129] Türk, Fuat, Murat Lüy, and Necaattin Bari§fi. "Kidney and renal tumor segmentation using a hybrid V-Net-Based model." Mathematics 8.10 (2020): 1772.

[130] Kuzina, Anna, Evgenii Egorov, and Evgeny Burnaev. "Bayesian generative models for knowledge transfer in MRI semantic segmentation problems." Frontiers in neuroscience 13 (2019): 844.

[131] Shahzadi, Iram, et al. "CNN-LSTM: Cascaded framework for brain tumour classification." 2018IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES). IEEE, 2018.

[132] Liu, Xiangbin, et al. "A review of deep-learning-based medical image segmentation methods." Sustainability 13.3 (2021): 1224.

[133] Wang, Zijian, et al. "Dilated 3d convolutional neural networks for brain mri data classification." Ieee Access 7 (2019): 134388-134398.

[134] Ghaffari, Mina, Arcot Sowmya, and Ruth Oliver. "Automated brain tumour segmentation using cascaded 3d densely-connected u-net." Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: 6th International Workshop, BrainLes 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4, 2020, Revised Selected Papers, Part I 6. Springer International Publishing, 2021.

[135] Brui, Ekaterina, et al. "Deep learning-based fully automatic segmentation of wrist cartilage in MR images." NMR in Biomedicine 33.8 (2020): e4320.