Nursuriati Jamil
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Nursuriati Jamil has a total of 113 co-authors. Here are the top co-authors by number of shared publications:
- Noraini Seman
Universiti Teknologi MARA
14 shared publications Raseeda Hamzah
10 shared publicationsNorizah Ardi
Universiti Teknologi MARA
8 shared publicationsIzzad Ramli
Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, Malaysia
7 shared publicationsZainab Abu Bakar
7 shared publications- Syed Mohd Zahid Syed Zainal Ariffin
Universiti Teknologi MARA
6 shared publications
Latest Publications
- Open Access
- 13 May 2025
- Published by Faculty of Electrical Engineering, Computer Science and Information Technology Osijek in International journal of electrical and computer engineering systems
- Vol. 16 (5) , 347-363
3D-based Convolutional Neural Networks for Medical Image Segmentation
Medical image segmentation is essential for disease screening and diagnosis, particularly through techniques like anatomical and lesion segmentation that can be used to isolate critical regions of interest. However, manual segmentation is labor-intensive, costly, and susceptible to subjective bias, underscoring the need for automation. Deep learning, particularly convolutional neural networks (CNNs), has significantly advanced segmentation accuracy and efficiency. With the introduction of 3D imaging, research has evolved from 2D CNNs to 3D CNNs, which leverage inter-slice information to improve segmentation precision. This paper aims to provide a literature review of studies published between 2018 and 2024 on platforms such as Google Scholar and ScienceDirect, where the identified relevant research are "3D segmentation" and "3D medical imaging". This study outlines the key stages of 3D CNN segmentation that include preprocessing, region-of-interest extraction, and post-processing. Furthermore, this study emphasizes the application of 3D CNN architectures to complex lung imaging scenarios, such as lung cancer and COVID-19. Although 3D CNNs outperform 2D CNNs in preserving spatial continuity across slices, they present notable limitations. Key challenges include heavy computational and high memory demands, as well as a dependency on large annotated datasets, which are often scarce in medical imaging. Additionally, effective multiscale feature learning remains a challenging issue, with current architectures struggling to generalize the features of interest across several usage variations. To further improve the segmentation performance, future research should prioritize developing adaptive algorithms and fostering interdisciplinary collaboration between computer scientists and medical professionals to design efficient and scalable models, designed specifically for clinical applications. This future research direction will enhance diagnostic accuracy and segmentation quality in 3D medical imaging.
- 13 May 2025
- Published by Faculty of Electrical Engineering, Computer Science and Information Technology Osijek in International journal of electrical and computer engineering systems
- Vol. 16 (5) , 347-363
Open Access3D-based Convolutional Neural Networks for Medical Image Segmentation
Medical image segmentation is essential for disease screening and diagnosis, particularly through techniques like anatomical and lesion segmentation that can be used to isolate critical regions of interest. However, manual segmentation is labor-intensive, costly, and susceptible to subjective bias, underscoring the need for automation. Deep learning, particularly convolutional neural networks (CNNs), has significantly advanced segmentation accuracy and efficiency. With the introduction of 3D imaging, research has evolved from 2D CNNs to 3D CNNs, which leverage inter-slice information to improve segmentation precision. This paper aims to provide a literature review of studies published between 2018 and 2024 on platforms such as Google Scholar and ScienceDirect, where the identified relevant research are "3D segmentation" and "3D medical imaging". This study outlines the key stages of 3D CNN segmentation that include preprocessing, region-of-interest extraction, and post-processing. Furthermore, this study emphasizes the application of 3D CNN architectures to complex lung imaging scenarios, such as lung cancer and COVID-19. Although 3D CNNs outperform 2D CNNs in preserving spatial continuity across slices, they present notable limitations. Key challenges include heavy computational and high memory demands, as well as a dependency on large annotated datasets, which are often scarce in medical imaging. Additionally, effective multiscale feature learning remains a challenging issue, with current architectures struggling to generalize the features of interest across several usage variations. To further improve the segmentation performance, future research should prioritize developing adaptive algorithms and fostering interdisciplinary collaboration between computer scientists and medical professionals to design efficient and scalable models, designed specifically for clinical applications. This future research direction will enhance diagnostic accuracy and segmentation quality in 3D medical imaging.
- Open Access
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 15 (2) , 1949-1960
Revisiting self-supervised contrastive learning for imbalanced classification
Class imbalance remains a formidable challenge in machine learning, particularly affecting fields that depend on accurate classification across skewed datasets, such as medical imaging and software defect prediction. Traditional approaches often fail to adequately address the underrepresentation of minority classes, leading to models that exhibit high performance on majority classes but have poor performance on critical minority classes. Self-supervised contrastive learning has become an extremely encouraging method for this issue, enabling the utilization of unlabeled data to generate robust and generalizable models. This paper reviews the advancements in self-supervised contrastive learning for imbalanced classification, focusing on methodologies that enhance model performance through innovative contrastive loss functions and data augmentation strategies. By pulling similar instances closer and pushing dissimilar ones apart, these techniques help mitigate the biases inherent in imbalanced datasets. We critically analyze the effectiveness of these methods in diverse scenarios and propose future research directions aimed at refining these approaches for broader application in real-world settings. This review serves as a guide for researchers exploring the potential of contrastive learning to address class imbalances, highlighting recent successes and identifying crucial gaps that need addressing.
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 15 (2) , 1949-1960
Open AccessRevisiting self-supervised contrastive learning for imbalanced classification
Class imbalance remains a formidable challenge in machine learning, particularly affecting fields that depend on accurate classification across skewed datasets, such as medical imaging and software defect prediction. Traditional approaches often fail to adequately address the underrepresentation of minority classes, leading to models that exhibit high performance on majority classes but have poor performance on critical minority classes. Self-supervised contrastive learning has become an extremely encouraging method for this issue, enabling the utilization of unlabeled data to generate robust and generalizable models. This paper reviews the advancements in self-supervised contrastive learning for imbalanced classification, focusing on methodologies that enhance model performance through innovative contrastive loss functions and data augmentation strategies. By pulling similar instances closer and pushing dissimilar ones apart, these techniques help mitigate the biases inherent in imbalanced datasets. We critically analyze the effectiveness of these methods in diverse scenarios and propose future research directions aimed at refining these approaches for broader application in real-world settings. This review serves as a guide for researchers exploring the potential of contrastive learning to address class imbalances, highlighting recent successes and identifying crucial gaps that need addressing.
- Open Access
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (2) , 1420-1429
Optimizing deep learning models from multi-objective perspective via Bayesian optimization
Optimizing hyperparameters is crucial for enhancing the performance of deep learning (DL) models. The process of configuring optimal hyperparameters, known as hyperparameter tuning, can be performed using various methods. Traditional approaches like grid search and random search have significant limitations. In contrast, Bayesian optimization (BO) utilizes a surrogate model and an acquisition function to intelligently navigate the hyperparameter space, aiming to provide deeper insights into performance disparities between naïve and advanced methods. This study evaluates BO's efficacy compared to baseline methods such as random search, manual search, and grid search across multiple DL architectures, including multi-layer perceptron (MLP), convolutional neural network (CNN), and LeNet, applied to the Modified National Institute of Standards and Technology (MNIST) and CIFAR-10 datasets. The findings indicate that BO, employing the tree-structured parzen estimator (TPE) search method and expected improvement (EI) acquisition function, surpasses alternative methods in intricate DL architectures such as LeNet and CNN. However, grid search shows superior performance in smaller DL architectures like MLP. This study also adopts a multi-objective (MO) perspective, balancing conflicting performance objectives such as accuracy, F1 score, and model size (parameter count). This MO assessment offers a comprehensive understanding of how these performance metrics interact and influence each other, leading to more informed hyperparameter tuning decisions.
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (2) , 1420-1429
Open AccessOptimizing deep learning models from multi-objective perspective via Bayesian optimization
Optimizing hyperparameters is crucial for enhancing the performance of deep learning (DL) models. The process of configuring optimal hyperparameters, known as hyperparameter tuning, can be performed using various methods. Traditional approaches like grid search and random search have significant limitations. In contrast, Bayesian optimization (BO) utilizes a surrogate model and an acquisition function to intelligently navigate the hyperparameter space, aiming to provide deeper insights into performance disparities between naïve and advanced methods. This study evaluates BO's efficacy compared to baseline methods such as random search, manual search, and grid search across multiple DL architectures, including multi-layer perceptron (MLP), convolutional neural network (CNN), and LeNet, applied to the Modified National Institute of Standards and Technology (MNIST) and CIFAR-10 datasets. The findings indicate that BO, employing the tree-structured parzen estimator (TPE) search method and expected improvement (EI) acquisition function, surpasses alternative methods in intricate DL architectures such as LeNet and CNN. However, grid search shows superior performance in smaller DL architectures like MLP. This study also adopts a multi-objective (MO) perspective, balancing conflicting performance objectives such as accuracy, F1 score, and model size (parameter count). This MO assessment offers a comprehensive understanding of how these performance metrics interact and influence each other, leading to more informed hyperparameter tuning decisions.
- Open Access
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (2) , 925-932
DualVitOA: A dual vision transformer-based model for osteoarthritis grading using x-ray images
Knee osteoarthritis (OA) is a primary factor contributing to reduced activity and physical impairment in older individuals. Early identification and treatment of knee OA can assist patients in delaying the advancement of the condition. Currently, knee OA is detected early using X-ray images and assessed based on the Kellgren-Lawrence (KL) grading system. Doctors' assessments are subjective and can vary among different doctors. The automatic knee OA grading and diagnosis can assist doctors and help doctors reduce their workload. A new novel network called dual-vision transformer (ViT) OA is proposed to automatically diagnose knee OA. The network utilizes pre-processing technologies to process the data before doing classification operations using the Dual-ViT network. The suggested network outperformed neural networks like ResNet, DenseNet, visual geometry group (VGG), inception, and ViT in terms of accuracy and mean absolute error (MAE), and achieved an accuracy of 78.4 and MAE of 0.471, demonstrating its effectiveness.
- 1 April 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (2) , 925-932
Open AccessDualVitOA: A dual vision transformer-based model for osteoarthritis grading using x-ray images
Knee osteoarthritis (OA) is a primary factor contributing to reduced activity and physical impairment in older individuals. Early identification and treatment of knee OA can assist patients in delaying the advancement of the condition. Currently, knee OA is detected early using X-ray images and assessed based on the Kellgren-Lawrence (KL) grading system. Doctors' assessments are subjective and can vary among different doctors. The automatic knee OA grading and diagnosis can assist doctors and help doctors reduce their workload. A new novel network called dual-vision transformer (ViT) OA is proposed to automatically diagnose knee OA. The network utilizes pre-processing technologies to process the data before doing classification operations using the Dual-ViT network. The suggested network outperformed neural networks like ResNet, DenseNet, visual geometry group (VGG), inception, and ViT in terms of accuracy and mean absolute error (MAE), and achieved an accuracy of 78.4 and MAE of 0.471, demonstrating its effectiveness.
- Open Access
- 1 February 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (1) , 109-118
Convolutional neural network modelling for autistic individualized education chatbot
The traditional education system for autistic kids needs integration with computer technology that embraces artificial intelligence to help school instructors and management. An application that enables the teacher to retrieve information from a trusted source is essential since the information is only sometimes available on time. Thus, developing a chatbot application that utilizes natural language processing can enhance the management of autistic schools and will help individualized education for autistic students. This research uses a deep learning model that utilizes a convolutional neural network to develop a chatbot as a teaching assist tool for teachers. The results show that the chatbot has achieved ˜0.03% loss when trained with different epoch numbers. In terms of usability, the chatbot achieves mean system usability scores of 80.48 ± 13.03. This may open opportunities for more effective individualized education for students with special needs and increase the potential to improve inclusive education for disabled students. It is useful to include future actions that enable the simplification of the use of this chatbot tool in a wide range of contexts. To close the education gap for children with disabilities, chatbots could help people with communication disabilities and could also significantly enhance the rate of communication.
- 1 February 2025
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 14 (1) , 109-118
Open AccessConvolutional neural network modelling for autistic individualized education chatbot
The traditional education system for autistic kids needs integration with computer technology that embraces artificial intelligence to help school instructors and management. An application that enables the teacher to retrieve information from a trusted source is essential since the information is only sometimes available on time. Thus, developing a chatbot application that utilizes natural language processing can enhance the management of autistic schools and will help individualized education for autistic students. This research uses a deep learning model that utilizes a convolutional neural network to develop a chatbot as a teaching assist tool for teachers. The results show that the chatbot has achieved ˜0.03% loss when trained with different epoch numbers. In terms of usability, the chatbot achieves mean system usability scores of 80.48 ± 13.03. This may open opportunities for more effective individualized education for students with special needs and increase the potential to improve inclusive education for disabled students. It is useful to include future actions that enable the simplification of the use of this chatbot tool in a wide range of contexts. To close the education gap for children with disabilities, chatbots could help people with communication disabilities and could also significantly enhance the rate of communication.
- Open Access
- 1 November 2023
- Published by Institute of Advanced Engineering and Science in International Journal of Reconfigurable and Embedded Systems (IJRES)
- Vol. 12 (3) , 442-452
Vehicle detection and classification using three variations of you only look once algorithm
Vehicle detection and classification are essential for advanced driver assistance systems (ADAS) and even traffic camera surveillance. Yet, it is challenging due to complex backgrounds, varying illumination intensities, occlusions, vehicle size, and type variations. This paper aims to apply you only look once (YOLO) since it has been proven to produce high object detection and classification accuracy. There are various versions of YOLO, and their performances differ. An investigation on the detection and classification performance of YOLOv3, YOLOv4, and YOLOv5 has been conducted. The training images were from common objects in context (COCO) and open image, two publicly available datasets. The testing input images were captured on a few highways in two main cities in Malaysia, namely Shah Alam and Kuala Lumpur. These images were captured using a mobile phone camera with different backgrounds during the day and night, representing different illuminations and varying types and sizes of vehicles. The accuracy and speed of detecting and classifying cars, trucks, buses, motorcycles, and bicycles have been evaluated. The experimental results show that YOLOv5 detects vehicles more accurately but slower than its predecessors, namely YOLOv4 and YOLOv3. Future work includes experimenting with newer versions of YOLO.
- 1 November 2023
- Published by Institute of Advanced Engineering and Science in International Journal of Reconfigurable and Embedded Systems (IJRES)
- Vol. 12 (3) , 442-452
Open AccessVehicle detection and classification using three variations of you only look once algorithm
Vehicle detection and classification are essential for advanced driver assistance systems (ADAS) and even traffic camera surveillance. Yet, it is challenging due to complex backgrounds, varying illumination intensities, occlusions, vehicle size, and type variations. This paper aims to apply you only look once (YOLO) since it has been proven to produce high object detection and classification accuracy. There are various versions of YOLO, and their performances differ. An investigation on the detection and classification performance of YOLOv3, YOLOv4, and YOLOv5 has been conducted. The training images were from common objects in context (COCO) and open image, two publicly available datasets. The testing input images were captured on a few highways in two main cities in Malaysia, namely Shah Alam and Kuala Lumpur. These images were captured using a mobile phone camera with different backgrounds during the day and night, representing different illuminations and varying types and sizes of vehicles. The accuracy and speed of detecting and classifying cars, trucks, buses, motorcycles, and bicycles have been evaluated. The experimental results show that YOLOv5 detects vehicles more accurately but slower than its predecessors, namely YOLOv4 and YOLOv3. Future work includes experimenting with newer versions of YOLO.
- Open Access
- 1 August 2023
- Published by Institute of Advanced Engineering and Science in Bulletin of Electrical Engineering and Informatics
- Vol. 12 (4) , 2269-2281
SiulMalaya: an annotated bird audio dataset of Malaysia lowland forest birds for passive acoustic monitoring
The laborious point count method of conducting bird surveys is still a common practice in Malaysia. An alternative method known as passive acoustic monitoring (PAM) is deployed in many countries by placing sound recorders at surveying sites to collect bird sounds. Studies revealed that the number of bird densities counted by human observers was agreeable with those obtained using PAM. However, one of the most essential gaps in conducting PAM is the lack of expert-verified bird-call databases. Therefore, the aim of this study is to construct the first annotated Malaysia lowland forest bird sounds called SiulMalaya to be used as ground-truth datasets for PAM-related activities. The raw bird sounds dataset was downloaded from Macaulay Library using the eBird platform. Data pre-processing was done to produce annotated audio tracks that can be used as training datasets for bird classification. SiulMalaya dataset was further validated by two bird experts from the Department of Wildlife and National Parks, Malaysia. A bird identification experiment was carried out to assess and validate SiulMalaya dataset using a convolutional neural network (CNN) learning model. Even though the accuracy of bird identification is slightly above 50%, the annotated dataset is shown to be viable for PAM-related operations.
- 1 August 2023
- Published by Institute of Advanced Engineering and Science in Bulletin of Electrical Engineering and Informatics
- Vol. 12 (4) , 2269-2281
Open AccessSiulMalaya: an annotated bird audio dataset of Malaysia lowland forest birds for passive acoustic monitoring
The laborious point count method of conducting bird surveys is still a common practice in Malaysia. An alternative method known as passive acoustic monitoring (PAM) is deployed in many countries by placing sound recorders at surveying sites to collect bird sounds. Studies revealed that the number of bird densities counted by human observers was agreeable with those obtained using PAM. However, one of the most essential gaps in conducting PAM is the lack of expert-verified bird-call databases. Therefore, the aim of this study is to construct the first annotated Malaysia lowland forest bird sounds called SiulMalaya to be used as ground-truth datasets for PAM-related activities. The raw bird sounds dataset was downloaded from Macaulay Library using the eBird platform. Data pre-processing was done to produce annotated audio tracks that can be used as training datasets for bird classification. SiulMalaya dataset was further validated by two bird experts from the Department of Wildlife and National Parks, Malaysia. A bird identification experiment was carried out to assess and validate SiulMalaya dataset using a convolutional neural network (CNN) learning model. Even though the accuracy of bird identification is slightly above 50%, the annotated dataset is shown to be viable for PAM-related operations.
- Open Access
- 1 August 2023
- Published by Institute of Advanced Engineering and Science in Bulletin of Electrical Engineering and Informatics
- Vol. 12 (4) , 2269-2281
SiulMalaya: an annotated bird audio dataset of Malaysia lowland forest birds for passive acoustic monitoring
The laborious point count method of conducting bird surveys is still a common practice in Malaysia. An alternative method known as passive acoustic monitoring (PAM) is deployed in many countries by placing sound recorders at surveying sites to collect bird sounds. Studies revealed that the number of bird densities counted by human observers was agreeable with those obtained using PAM. However, one of the most essential gaps in conducting PAM is the lack of expert-verified bird-call databases. Therefore, the aim of this study is to construct the first annotated Malaysia lowland forest bird sounds called SiulMalaya to be used as ground-truth datasets for PAM-related activities. The raw bird sounds dataset was downloaded from Macaulay Library using the eBird platform. Data pre-processing was done to produce annotated audio tracks that can be used as training datasets for bird classification. SiulMalaya dataset was further validated by two bird experts from the Department of Wildlife and National Parks, Malaysia. A bird identification experiment was carried out to assess and validate SiulMalaya dataset using a convolutional neural network (CNN) learning model. Even though the accuracy of bird identification is slightly above 50%, the annotated dataset is shown to be viable for PAM-related operations.
- 1 August 2023
- Published by Institute of Advanced Engineering and Science in Bulletin of Electrical Engineering and Informatics
- Vol. 12 (4) , 2269-2281
Open AccessSiulMalaya: an annotated bird audio dataset of Malaysia lowland forest birds for passive acoustic monitoring
The laborious point count method of conducting bird surveys is still a common practice in Malaysia. An alternative method known as passive acoustic monitoring (PAM) is deployed in many countries by placing sound recorders at surveying sites to collect bird sounds. Studies revealed that the number of bird densities counted by human observers was agreeable with those obtained using PAM. However, one of the most essential gaps in conducting PAM is the lack of expert-verified bird-call databases. Therefore, the aim of this study is to construct the first annotated Malaysia lowland forest bird sounds called SiulMalaya to be used as ground-truth datasets for PAM-related activities. The raw bird sounds dataset was downloaded from Macaulay Library using the eBird platform. Data pre-processing was done to produce annotated audio tracks that can be used as training datasets for bird classification. SiulMalaya dataset was further validated by two bird experts from the Department of Wildlife and National Parks, Malaysia. A bird identification experiment was carried out to assess and validate SiulMalaya dataset using a convolutional neural network (CNN) learning model. Even though the accuracy of bird identification is slightly above 50%, the annotated dataset is shown to be viable for PAM-related operations.
- Open Access
- 24 May 2023
- Published by Universiti Putra Malaysia in Pertanika Journal of Science and Technology
- Vol. 31 (4) , 1731-1754
Design of a Smart Portable Farming Kit for Indoor Cultivation Using the Raspberry Pi Platform
Pertanika Journal information and guidelines
- 24 May 2023
- Published by Universiti Putra Malaysia in Pertanika Journal of Science and Technology
- Vol. 31 (4) , 1731-1754
Open AccessDesign of a Smart Portable Farming Kit for Indoor Cultivation Using the Raspberry Pi Platform
Pertanika Journal information and guidelines
- Open Access
- 17 March 2023
- Published by MDPI AG in Information
- Vol. 14 (3) , 190
An Improved Co-Training and Generative Adversarial Network (Diff-CoGAN) for Semi-Supervised Medical Image Segmentation
Semi-supervised learning is a technique that utilizes a limited set of labeled data and a large amount of unlabeled data to overcome the challenges of obtaining a perfect dataset in deep learning, especially in medical image segmentation. The accuracy of the predicted labels for the unlabeled data is a critical factor that affects the training performance, thus reducing the accuracy of segmentation. To address this issue, a semi-supervised learning method based on the Diff-CoGAN framework was proposed, which incorporates co-training and generative adversarial network (GAN) strategies. The proposed Diff-CoGAN framework employs two generators and one discriminator. The generators work together by providing mutual information guidance to produce predicted maps that are more accurate and closer to the ground truth. To further improve segmentation accuracy, the predicted maps are subjected to an intersection operation to identify a high-confidence region of interest, which reduces boundary segmentation errors. The predicted maps are then fed into the discriminator, and the iterative process of adversarial training enhances the generators’ ability to generate more precise maps, while also improving the discriminator’s ability to distinguish between the predicted maps and the ground truth. This study conducted experiments on the Hippocampus and Spleen images from the Medical Segmentation Decathlon (MSD) dataset using three semi-supervised methods: co-training, semi-GAN, and Diff-CoGAN. The experimental results demonstrated that the proposed Diff-CoGAN approach significantly enhanced segmentation accuracy compared to the other two methods by benefiting on the mutual guidance of the two generators and the adversarial training between the generators and discriminator. The introduction of the intersection operation prior to the discriminator also further reduced boundary segmentation errors.
- 17 March 2023
- Published by MDPI AG in Information
- Vol. 14 (3) , 190
Open AccessAn Improved Co-Training and Generative Adversarial Network (Diff-CoGAN) for Semi-Supervised Medical Image Segmentation
Semi-supervised learning is a technique that utilizes a limited set of labeled data and a large amount of unlabeled data to overcome the challenges of obtaining a perfect dataset in deep learning, especially in medical image segmentation. The accuracy of the predicted labels for the unlabeled data is a critical factor that affects the training performance, thus reducing the accuracy of segmentation. To address this issue, a semi-supervised learning method based on the Diff-CoGAN framework was proposed, which incorporates co-training and generative adversarial network (GAN) strategies. The proposed Diff-CoGAN framework employs two generators and one discriminator. The generators work together by providing mutual information guidance to produce predicted maps that are more accurate and closer to the ground truth. To further improve segmentation accuracy, the predicted maps are subjected to an intersection operation to identify a high-confidence region of interest, which reduces boundary segmentation errors. The predicted maps are then fed into the discriminator, and the iterative process of adversarial training enhances the generators’ ability to generate more precise maps, while also improving the discriminator’s ability to distinguish between the predicted maps and the ground truth. This study conducted experiments on the Hippocampus and Spleen images from the Medical Segmentation Decathlon (MSD) dataset using three semi-supervised methods: co-training, semi-GAN, and Diff-CoGAN. The experimental results demonstrated that the proposed Diff-CoGAN approach significantly enhanced segmentation accuracy compared to the other two methods by benefiting on the mutual guidance of the two generators and the adversarial training between the generators and discriminator. The introduction of the intersection operation prior to the discriminator also further reduced boundary segmentation errors.