Siti Noraini Sulaiman - Academia.edu (original) (raw)
Papers by Siti Noraini Sulaiman
Journal of electrical and electronic systems research, Nov 1, 2022
Ischaemic heart disease is caused by the blockage of blood flow to the heart and led to an increa... more Ischaemic heart disease is caused by the blockage of blood flow to the heart and led to an increase in the number of deaths in Malaysia with a total of 18,267 deaths reported in 2018. The use of late gadolinium enhancement (LGE) contrast agents in cardiac magnetic resonance imaging (CMR) for the evaluation of post-MI patients have demonstrated an incremental prognostic value. The LGE allows for direct visualisation of scarred myocardial tissue as a region interest (ROI) that is being enhanced. By assessing the enhanced ROI in the left ventricular (LV), the myocardial scar can be detected and diagnosed. However, the current approach for detecting myocardial scar in LV from CMR images is done visually by the radiologist and is time-consuming and subject to variability. Therefore, it is proposed to incorporate computer-aided diagnosis using a deep learning approach based on the YOLO algorithm for automatically locating the LV region that will improve the diagnostic accuracy of the myocardial scar tissue via LGE data acquired in post-MI patients. In this work, a total of 159 images from 10 subjects are selected and split into three datasets i.e. training, validating and testing datasets with the ratio of 80%, 10% and 10% respectively. The deep learning techniques based on YOLOv2 and YOLOv3 with three different solvers (ADAM, RMSProp and SGDM) are used to evaluate the performance of the automated LV localization from the CMR images. The highest localization accuracy is obtained from YOLOv2 using an ADAM solver with an average precision (AP) of 100% and mean intersection over union (IoU) of 89%.
Journal of electrical and electronic systems research, Dec 21, 2020
Lung cancer is a common cause of death among people throughout the world. Lung cancer detection c... more Lung cancer is a common cause of death among people throughout the world. Lung cancer detection can be done in several ways, such as radiography, magnetic resonance imaging (MRI) and computed tomography (CT). These methods take up a lot of resources in terms of time and money. However, CT has good for lung cancer detection, offers a lower cost, short imaging time and widespread availability. Early diagnosis of lung cancer can help doctors to treat patients in order to reduce the number of mortalities. This paper presents designation of thorax and nonthorax regions for lung cancer detection in CT Scan images using deep learning. The primary aim of this research is to propose an intelligent, fast and accurate method for lung cancer detection. As initial stage we proposed a thorax and non-thorax slice detection for CT scan images using deep convolutional neural network (DCNN) so that later it can be used to simplify the process of lung cancer detection. The proposed method involved the development of DCNN network architecture. It comprises the following steps which involves designed the convolution layer, activation function, max pooling, fully-connected layer and output size. We present three DCNN structures to find the most effective network for thorax and non-thorax region detection. All networks were trained using 12866 images and validate the performance using 5514 images. Simulation results showed that DCNN 2 and DCNN 3 were able to classify the thorax and non-thorax regions with good performance. The most efficient network is the DCNN with fivelayer structure (DCNN 2). This DCNN model achieved an accuracy of 99.42% with moderate duration of training time.
Jurnal teknologi, Feb 23, 2023
Lung lesion identification is an important aspect of an early lung cancer diagnosis. Early identi... more Lung lesion identification is an important aspect of an early lung cancer diagnosis. Early identification of lung cancer may assist physicians in treating patients. This paper uses computed tomography scan images to present a lung lesion identification geometrical feature. From the previous studies, lung segmentation is particularly challenging because differences in pulmonary inflation with an elastic chest wall can result in significant variability in volumes and margins when attempting to automate lung segmentation. Besides, the features used to describe a lung lesion focus on image features which are geometric, appearance, texture, and others. This study develops an image processing technique that uses image segmentation algorithms to segment lung lesions in computed tomography images. The suggested approach includes the following stages, which require image processing techniques: data collection, image segmentation, and performance evaluation. The computed tomography scan images were collected from Advanced Medical and Dental Institute (AMDI), Universiti Sains Malaysia database. As a contribution to biomedical engineering, this study has successfully calculated the performance of the image processing method for lung segmentation, which gets an average accuracy of 99.38%, recall is 99.45%, and F-score is 99.6. The lung lesion segmentation approach based on the object's size could help investigate image abnormality for medical analysis. From the study, 80% of the total lesion identification using the proposed method was correctly predicted when compared with the radiologist's lesion mark. The experiment results found clear support for the next stage of this research.
International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Jun 1, 2024
Accurate segmentation of myocardial scar tissue on late gadolinium enhancement-magnetic cardiac r... more Accurate segmentation of myocardial scar tissue on late gadolinium enhancement-magnetic cardiac resonance imaging (LGE-CMR) is exceptionally vital for clinical applications, enabling precise diagnosis and effective treatment of various cardiac diseases, such as myocardial infarction and cardiomyopathies. However, the ventricle (LV) variations in the size and shape, artifacts, and image resolution of LGE-CMR has made automatic segmentation of myocardial scar tissue more challenging. While many existing approaches delineate the LV myocardium region using multi-modal segmentation, these models may be computationally complex and suffer from misalignment. Therefore, this study proposed an automatic dual-stage DeepLabV3+ based approach tailored for myocardial scar segmentation on short-axis LGE-MRI exclusively. To segment myocardial scar tissue, the second stage employs the segmented LV chamber from the previous stage. The encoder part of the framework utilizes a MobileNetV2 and ResNet50 backbone for the first and second segmentation, respectively, aiming for optimal resolution of feature maps. Both stages tailor an improved Atrous Spatial Pyramid Pooling module in the DeepLabV3+ model with fine-tuned dilated atrous rates to effectively extract the LV chamber and myocardial scar from the complex LGE-MRI background. Based on the results, the proposed dual-stage network recorded an outstanding segmentation performance, with mean Dice score of 96.02% for LV chamber segmentation and 68.01% for scar tissue extraction.
Journal of Advanced Research in Applied Sciences and Engineering Technology, Mar 26, 2024
Most supermarket don't have a crowd traffic counting systems to track and counting customer enter... more Most supermarket don't have a crowd traffic counting systems to track and counting customer entering their shop lot. In practise, these methods are useful for businesses in managing customer flow and optimize staffing and marketing efforts. Furthermore, the information can be used to estimate the popularity of the shop with relation to people entering the shop lot. The information also useful for the shop owner in determining the renting value. Therefore, the paper presents a system for tracking and counting people in a retail store using on the edge device, a Jetson Nano board. The comparison of the performance of two algorithms for people detecting of YOLOv5 and MobileNet-SSD are used in this work, YOLOv5 is a state-of-the-art object detection model that is known for its accuracy, but it is computationally intensive and may not be suitable for running on resource-constrained devices such as the Jetson Nano. MobileNet-SSD is a lightweight object detection model that is designed to run efficiently on mobile devices and embedded systems. Next is to track people using SORT, SORT is a real-time multi-object tracking algorithm that is based on the Kalman filter and the Hungarian algorithm. The results show that YOLOv5 was able to achieve the highest accuracy in detecting and counting people, but it was slower than the other two algorithms. MobileNet-SSD was the fastest algorithm, but it had lower accuracy compared to YOLOv5. In conclusion, the choice of algorithm will depend on the tradeoff between accuracy and computational resources, and SORT is a good option for realtime people counting on resource-constrained devices.
Indonesian journal of electrical engineering and computer science, May 1, 2024
Intelligence algorithm systems rely on a large dataset to effectively extract significant feature... more Intelligence algorithm systems rely on a large dataset to effectively extract significant features that can recognize patterns for classification purposes and extensively utilized to assist the physicians in diagnosis of lung cancer. Extracting valuable features from the available dataset is crucial, especially in cases where additional real data may not be readily accessible. In this context, we propose a novel method called feature extraction based on centroid (FE_CXY) for lesion localization, utilizing a statistical approach. The approach begins with a segmentation process that employs image processing techniques to extract features of interest which is data centroid. This extracted data is then used to compute statistical measurements, revealing hidden patterns that contribute to distinguishing between lesion and non-lesion locations. The method's efficiency is reflected in the development of robust models with improved performance in localizing lung lesions. The study's statistical findings strongly indicate that FE_CXY plays a crucial role as an important feature for detecting lesion localization supported by a student's t-test, which identifies a statistically significant difference in the patterns between lesion and non-lesion localization (p<0.05). By incorporating this method into lung cancer detection systems, we anticipate improved accuracy and efficacy, thereby benefiting early diagnosis and treatment planning.
Bioengineering
Mass detection in mammograms has a limited approach to the presence of a mass in overlapping dens... more Mass detection in mammograms has a limited approach to the presence of a mass in overlapping denser fibroglandular breast regions. In addition, various breast density levels could decrease the learning system’s ability to extract sufficient feature descriptors and may result in lower accuracy performance. Therefore, this study is proposing a textural-based image enhancement technique named Spatial-based Breast Density Enhancement for Mass Detection (SbBDEM) to boost textural features of the overlapped mass region based on the breast density level. This approach determines the optimal exposure threshold of the images’ lower contrast limit and optimizes the parameters by selecting the best intensity factor guided by the best Blind/Reference-less Image Spatial Quality Evaluator (BRISQUE) scores separately for both dense and non-dense breast classes prior to training. Meanwhile, a modified You Only Look Once v3 (YOLOv3) architecture is employed for mass detection by specifically assigni...
International Journal of Integrated Engineering
This paper is presented a system that monitor the river water level by using computer vision with... more This paper is presented a system that monitor the river water level by using computer vision with image processing and IoT. This system is developed to detect riverbank level and river water level by applying image processing where edge detection technique is applied on both images captured by video camera. The flood severity level is determined by comparing the river water level and the riverbank level. Then, the determined flood severity is upload to IoT platform. A notification is sent the people when the flood severity level reached certain critical level via Telegram app which one of the social media applications. The available Raspberry Pi 3 Model B is usedas a controller in this system hardware device with the Raspberry Pi 5MP camera module.The IoT platform used is Ubidots where the user can be notified through it. The main contribution of this work is on the integration of computer vision with IoT Cloud as an early flood monitoring system in responding to climate change by d...
Journal of Electrical & Electronic Systems Research
In large-scale medical imaging, selecting the best image to extract relevant imaging biomarkers f... more In large-scale medical imaging, selecting the best image to extract relevant imaging biomarkers for image assessment is crucial. Segmentation of the left ventricle (LV) and myocardium are performed in computer-aided analysis usually at short-axis slices of cardiac magnetic resonance (MR) image to quantify cardiovascular disease assessment, such as myocardial scarring, LV ejection fraction and LV mass. The need to correctly identify a short-axis slice range for efficient quantification is preferred for automatic classification of the slice range of interest. The goal of this research is to establish an image processing method for the segmentation of Left ventricle scar from late gadolinium-enhanced (LGE) MR images. In order to achieve the main purpose, the work is divided into two parts, the first is to identify the cardiac Left ventricle segment (LVS) in the stack of short-axis LGE MR images and the second part; detecting the scar in between the LV myocardium area. This paper will present the outcomes of the first part by utilizing a deep convolutional neural network (DCNN) to construct an automatic system for classifying LVS and Non-Left Ventricle Segment (Non-LVS) in MR images. The same image dataset will be used for a comparative analysis with six DCNN models designed from scratch and three famed pre-trained networks, Alexnet, GoogleNet and SqueezeNet. Each model is trained up to 35 epochs using the Cardiac Atlas dataset and cross-validation method. The outcome from this work demonstrated that the DCNN3Y performs well over small training data with an average accuracy of 94.49%. Whereas SqueezeNet outperformed the three pre-trained networks with an average accuracy of 96.96%. It has also been discovered that increasing the number of filters and their subsequent configuration slightly influences the network's performance. This produces very promising results showing that it is ready to be used in the second part of this research. The outcome of this research can compensate for the deficiencies of manual detection in the original image detection system, increase detection efficiency, reduce detection misjudgments, and advance the development of automated and intelligent detection in the medical field.
Journal of Electrical & Electronic Systems Research
Screening mammography has been clinically practiced as a common method for monitoring any potenti... more Screening mammography has been clinically practiced as a common method for monitoring any potential breast diseases, especially in denser breasts among women. The advancement in medical imaging technology proved that the integration of artificial intelligence had given a significant impact on the breast screening process. However, due to various demographic patient backgrounds in clinical profile and non-standardized configurations of the developed intelligence models, such application is incapable of being applied by a health practitioner. Commonly, the deep learning method trained on a single network classifier has resulted in lower performance accuracy. With the motivation to improve the performance of classifying the dense breast, this paper proposes an improved classification model using deep learning approach of Convolutional Neural Network (CNN) and Support Vector Machine (SVM) employed on different sets of publicly established mammogram images of craniocaudal (CC) and medio-lateral (MLO) views. In this study, pre-trained CNN models, namely GoogleNet, ResNet50, ResNet101 and AlexNet, are used with SVM as a classifier. The proposed method for the classification of breast density region is superior to the existing methods as indicated from model performance quantitative results of accuracy, precision and area under the curve (AUC) of its receiver operating characteristic (ROC) curves. Significant improvement in model performance has been obtained using ResNet50 and GoogleNet with SVM classifier with > 94% accuracy and AUC > 0.95. Furthermore, the model is proved to have good feature extraction capabilities to work for various breast density images that can be further explored into detecting malignancy from screening mammogram images.
International Journal of Electrical and Computer Engineering (IJECE)
Lung cancer is the leading cause of cancer death among people worldwide. The primary aim of this ... more Lung cancer is the leading cause of cancer death among people worldwide. The primary aim of this research is to establish an image processing method for lung cancer detection. This paper focuses on lung region segmentation from computed tomography (CT) scan images. In this work, a new procedure for lung region segmentation is proposed. First, the lung CT scan images will undergo an image thresholding stage before going through two morphological reconstruction and masking stages. In between morphological and masking stages, object extraction, border change, and object elimination will occur. Finally, the lung field will be annotated. The outcomes of the proposed procedure and previous lung segmentation methods i.e., the modified watershed segmentation method is compared with the ground truth images for performance evaluation that will be carried out both in qualitative and quantitative manners. Based on the analyses, the new proposed procedure for lung segmentation, denotes better pe...
2017 International Conference on Electrical, Electronics and System Engineering (ICEESE), 2017
White matter hyperintensities (WMH) are small regions of high signal intensity that are observabl... more White matter hyperintensities (WMH) are small regions of high signal intensity that are observable on the white matter region of the brain through magnetic resonance imaging images. Generally, the medical expert conducts a white matter hyperintensities analysis to investigate brain tissue abnormality using manual or semi-automatic methods. However, those methods are prone to error and they establish unreliable results as different in rating scales. In this paper, a fully automatic method is proposed to identify WMH using the multimodal technique which combining image segmentation and enhancement. This method is introduced as an unsupervised method to automatically segment WMH on MRI images of T2-weighted and FLAIR sequences. Subsequently, the processed sequences are integrated by overlying the mapping images in order to map the most precise WMH regions. The accuracy of the WMH regions identification is assessed through the similarity index between automated and manual approach. The experimental results show that the proposed method has achieved significant results to detect exact WMH area. The proposed method is suitable to be implemented in analyzing white matter hyperintensities identification and it may serves as a computer-aided tool for radiologists.
Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Breast can... more Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Breast cancer is well known as a mortal disease. In United States, it is reported as the seccond leading cause of death in women after lung cancer
2022 IEEE 11th International Conference on Communication Systems and Network Technologies (CSNT)
IEEE Access
This study optimized the latest YOLOv5 framework, including its subset models, with training on d... more This study optimized the latest YOLOv5 framework, including its subset models, with training on different datasets that differed in image contrast and cloudiness to assess model performances based on quantitative metrics and image processing speed. The hyperparameter in the feature-extraction phase was configured based on the learning rate and momentum and further improved based on the adaptive moment estimation (ADAM) optimizer and the function reducing-learning-rate-on-plateau to optimize the model's training scheme. The optimized YOLOv5s achieved a better performance, with a mean average precision of 98.6% and a high inference speed of 106 frames per second. The ADAM optimizer with a detailed learning rate (0.0001) and momentum (0.99) fine-tuning yielded a sufficient convergence rate (0.69% at 55th epoch) to assist YOLOv5s in attaining a more precise detection for underwater objects. INDEX TERMS Image processing speed, object recognition, optimization model, tuning hyper-parameter, underwater imaging.
2021 6th IEEE International Conference on Recent Advances and Innovations in Engineering (ICRAIE), 2021
Cold Chain Management (CCM) can be described as the supply chain network of facilities and distri... more Cold Chain Management (CCM) can be described as the supply chain network of facilities and distribution that are controlled by temperature and humidity throughout the supply chain stages and entities. In CCM, the product will undergo the delivery process from the storeroom until it reaches the market cold room. However, the likelihood of accidents occurring during the process of delivery such as the breakdown of vehicles and the collapse of the freezing system can affect product quality. Therefore, an immediate response must be taken to prevent these setbacks from happening by implementing a real-time CCM system by utilizing the Sigfox UnaShield that stack into Arduino MakerUNO that read the temperature value and then sent to the Sigfox network. The data will be displayed in the Ubidots dashboard. An encouraging result is obtained on the prototyped used for the CCM monitoring using Ubidots Cloud and Xperanti Sigfox network.
Available online 30 March 2021 Dyslexia is developed by neurobiological in origin which is catego... more Available online 30 March 2021 Dyslexia is developed by neurobiological in origin which is categorized as learning disorder that affect the ability to read, spell, write and speak. The most common dyslexia symptom can easily be identified through the handwriting pattern. There are many intelligence and computational methods that have been proposed, and they have provided various and different performance to evaluate the proposed system ability. However, system performances are varied and nonstandardized in each assesment on dyslexic children to validate the presence of dyslexia symptom. The recent deep learning models have been employed to improve the assesment performance and (the models/ they have shown) shows significant output to detect and classify the present of dyslexia symptoms among school children. Therefore, there is a crucial need in deep learning, specifically for Convolutional Neural Network ( CNN) to validate performances of different networks, so that the most perfor...
![Research paper thumbnail of Sistem klasifikasi pintar untuk mikrokalsifikasi barah payudara / Siti Noraini Sulaiman … [et al.]](https://attachments.academia-assets.com/97193620/thumbnails/1.jpg)
](Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Pengesanan... more Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Pengesanan dari peringkat awal dapat membantu mengurangkan kadar kematian. Mammografi dan ultrabunyi adalah dua kaedah yang sering digunakan untuk mengesan barah payudara. Namun masalah yang timbul ialah para doktor tidak dapat membuat keputusan dengan tepat kerana imej kabur, tidak jelas atau bercampur hingar. Oleh itu satu teknik pemprosesan imej dibangunkan sebagai satu perantara berguna bagi para doktor menentukan keputusan mereka. Dalam penyelidikan ini, satu sistem pemprosesan imej secara berkomputer telah dibangunkan bagi memproses imej-imej mammogram bertujuan mengesan mikrokalsifikasi, petanda paling umum kehadiran barah payudara. Tiga teknik prapemprosesan digunakan iaitu peningkatan imej, penurasan, dan peruasan mikrokalsifikasi. Dalam peruasan mikrokalsifikasi, teknik pengesanan pinggir digunakan bertujuan untuk meningkatkan ciri-ciri diagnostiknya dan seterusnya mengira bilangan mikro...
Journal of Electrical & Electronic Systems Research, 2021
This paper presents a lesion of lung cancer detection exist in CT scan images using watershed seg... more This paper presents a lesion of lung cancer detection exist in CT scan images using watershed segmentation. Lung cancer is a disease of uncontrolled cell growth in tissues of the lung. It seems to be the common cause of death among people throughout the world. Therefore, diagnosis of lung cancer at early stage can help doctors to treat patients and keep them alive. The main aim of this research is to establish an image processing method for segmentation of lung cancer from CT scan images by using image processing techniques. In order to achieve the main aims, the work is divided into two parts, first is obtaining lung region from CT scan images and second is detecting the lesion of lung cancer. Hence, this paper will present the outcome of the second part. The lung lobes and nodules or lesion in CT image are segmented using two techniques which are convolution watershed and modified watershed within two stage approaches. Firstly, the image will undergo threshold, clustering and imag...
Electrical Engineering & Electromechanics, 2021
Introduction. Known vibrational energy harvesting methods use a source of vibration to harvest el... more Introduction. Known vibrational energy harvesting methods use a source of vibration to harvest electric energy. Piezoelectric material works as a sensing element converted mechanical energy (vibration) to electrical energy (electric field). The existing piezoelectric energy harvesting (PEHs) devices have low sensitivity, low energy conversion, and low bandwidth. The novelty of the proposed work consists of the design of PEH’s structure. Air cavity was implemented in the design where it is located under the sensing membrane to improve sensitivity. Another novelty is also consisting in the design structure where the flexural membrane was located at the top of electrodes. The third novelty is a new design structure of printed circuit board (PCB). The purpose of improvised design is to increase the stress in between the edges of PEH and increase energy conversion. With the new structure of PCB, it will work as a substrate that absorbs surrounding vibration energy and transfers it to sen...
Journal of electrical and electronic systems research, Nov 1, 2022
Ischaemic heart disease is caused by the blockage of blood flow to the heart and led to an increa... more Ischaemic heart disease is caused by the blockage of blood flow to the heart and led to an increase in the number of deaths in Malaysia with a total of 18,267 deaths reported in 2018. The use of late gadolinium enhancement (LGE) contrast agents in cardiac magnetic resonance imaging (CMR) for the evaluation of post-MI patients have demonstrated an incremental prognostic value. The LGE allows for direct visualisation of scarred myocardial tissue as a region interest (ROI) that is being enhanced. By assessing the enhanced ROI in the left ventricular (LV), the myocardial scar can be detected and diagnosed. However, the current approach for detecting myocardial scar in LV from CMR images is done visually by the radiologist and is time-consuming and subject to variability. Therefore, it is proposed to incorporate computer-aided diagnosis using a deep learning approach based on the YOLO algorithm for automatically locating the LV region that will improve the diagnostic accuracy of the myocardial scar tissue via LGE data acquired in post-MI patients. In this work, a total of 159 images from 10 subjects are selected and split into three datasets i.e. training, validating and testing datasets with the ratio of 80%, 10% and 10% respectively. The deep learning techniques based on YOLOv2 and YOLOv3 with three different solvers (ADAM, RMSProp and SGDM) are used to evaluate the performance of the automated LV localization from the CMR images. The highest localization accuracy is obtained from YOLOv2 using an ADAM solver with an average precision (AP) of 100% and mean intersection over union (IoU) of 89%.
Journal of electrical and electronic systems research, Dec 21, 2020
Lung cancer is a common cause of death among people throughout the world. Lung cancer detection c... more Lung cancer is a common cause of death among people throughout the world. Lung cancer detection can be done in several ways, such as radiography, magnetic resonance imaging (MRI) and computed tomography (CT). These methods take up a lot of resources in terms of time and money. However, CT has good for lung cancer detection, offers a lower cost, short imaging time and widespread availability. Early diagnosis of lung cancer can help doctors to treat patients in order to reduce the number of mortalities. This paper presents designation of thorax and nonthorax regions for lung cancer detection in CT Scan images using deep learning. The primary aim of this research is to propose an intelligent, fast and accurate method for lung cancer detection. As initial stage we proposed a thorax and non-thorax slice detection for CT scan images using deep convolutional neural network (DCNN) so that later it can be used to simplify the process of lung cancer detection. The proposed method involved the development of DCNN network architecture. It comprises the following steps which involves designed the convolution layer, activation function, max pooling, fully-connected layer and output size. We present three DCNN structures to find the most effective network for thorax and non-thorax region detection. All networks were trained using 12866 images and validate the performance using 5514 images. Simulation results showed that DCNN 2 and DCNN 3 were able to classify the thorax and non-thorax regions with good performance. The most efficient network is the DCNN with fivelayer structure (DCNN 2). This DCNN model achieved an accuracy of 99.42% with moderate duration of training time.
Jurnal teknologi, Feb 23, 2023
Lung lesion identification is an important aspect of an early lung cancer diagnosis. Early identi... more Lung lesion identification is an important aspect of an early lung cancer diagnosis. Early identification of lung cancer may assist physicians in treating patients. This paper uses computed tomography scan images to present a lung lesion identification geometrical feature. From the previous studies, lung segmentation is particularly challenging because differences in pulmonary inflation with an elastic chest wall can result in significant variability in volumes and margins when attempting to automate lung segmentation. Besides, the features used to describe a lung lesion focus on image features which are geometric, appearance, texture, and others. This study develops an image processing technique that uses image segmentation algorithms to segment lung lesions in computed tomography images. The suggested approach includes the following stages, which require image processing techniques: data collection, image segmentation, and performance evaluation. The computed tomography scan images were collected from Advanced Medical and Dental Institute (AMDI), Universiti Sains Malaysia database. As a contribution to biomedical engineering, this study has successfully calculated the performance of the image processing method for lung segmentation, which gets an average accuracy of 99.38%, recall is 99.45%, and F-score is 99.6. The lung lesion segmentation approach based on the object's size could help investigate image abnormality for medical analysis. From the study, 80% of the total lesion identification using the proposed method was correctly predicted when compared with the radiologist's lesion mark. The experiment results found clear support for the next stage of this research.
International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, Jun 1, 2024
Accurate segmentation of myocardial scar tissue on late gadolinium enhancement-magnetic cardiac r... more Accurate segmentation of myocardial scar tissue on late gadolinium enhancement-magnetic cardiac resonance imaging (LGE-CMR) is exceptionally vital for clinical applications, enabling precise diagnosis and effective treatment of various cardiac diseases, such as myocardial infarction and cardiomyopathies. However, the ventricle (LV) variations in the size and shape, artifacts, and image resolution of LGE-CMR has made automatic segmentation of myocardial scar tissue more challenging. While many existing approaches delineate the LV myocardium region using multi-modal segmentation, these models may be computationally complex and suffer from misalignment. Therefore, this study proposed an automatic dual-stage DeepLabV3+ based approach tailored for myocardial scar segmentation on short-axis LGE-MRI exclusively. To segment myocardial scar tissue, the second stage employs the segmented LV chamber from the previous stage. The encoder part of the framework utilizes a MobileNetV2 and ResNet50 backbone for the first and second segmentation, respectively, aiming for optimal resolution of feature maps. Both stages tailor an improved Atrous Spatial Pyramid Pooling module in the DeepLabV3+ model with fine-tuned dilated atrous rates to effectively extract the LV chamber and myocardial scar from the complex LGE-MRI background. Based on the results, the proposed dual-stage network recorded an outstanding segmentation performance, with mean Dice score of 96.02% for LV chamber segmentation and 68.01% for scar tissue extraction.
Journal of Advanced Research in Applied Sciences and Engineering Technology, Mar 26, 2024
Most supermarket don't have a crowd traffic counting systems to track and counting customer enter... more Most supermarket don't have a crowd traffic counting systems to track and counting customer entering their shop lot. In practise, these methods are useful for businesses in managing customer flow and optimize staffing and marketing efforts. Furthermore, the information can be used to estimate the popularity of the shop with relation to people entering the shop lot. The information also useful for the shop owner in determining the renting value. Therefore, the paper presents a system for tracking and counting people in a retail store using on the edge device, a Jetson Nano board. The comparison of the performance of two algorithms for people detecting of YOLOv5 and MobileNet-SSD are used in this work, YOLOv5 is a state-of-the-art object detection model that is known for its accuracy, but it is computationally intensive and may not be suitable for running on resource-constrained devices such as the Jetson Nano. MobileNet-SSD is a lightweight object detection model that is designed to run efficiently on mobile devices and embedded systems. Next is to track people using SORT, SORT is a real-time multi-object tracking algorithm that is based on the Kalman filter and the Hungarian algorithm. The results show that YOLOv5 was able to achieve the highest accuracy in detecting and counting people, but it was slower than the other two algorithms. MobileNet-SSD was the fastest algorithm, but it had lower accuracy compared to YOLOv5. In conclusion, the choice of algorithm will depend on the tradeoff between accuracy and computational resources, and SORT is a good option for realtime people counting on resource-constrained devices.
Indonesian journal of electrical engineering and computer science, May 1, 2024
Intelligence algorithm systems rely on a large dataset to effectively extract significant feature... more Intelligence algorithm systems rely on a large dataset to effectively extract significant features that can recognize patterns for classification purposes and extensively utilized to assist the physicians in diagnosis of lung cancer. Extracting valuable features from the available dataset is crucial, especially in cases where additional real data may not be readily accessible. In this context, we propose a novel method called feature extraction based on centroid (FE_CXY) for lesion localization, utilizing a statistical approach. The approach begins with a segmentation process that employs image processing techniques to extract features of interest which is data centroid. This extracted data is then used to compute statistical measurements, revealing hidden patterns that contribute to distinguishing between lesion and non-lesion locations. The method's efficiency is reflected in the development of robust models with improved performance in localizing lung lesions. The study's statistical findings strongly indicate that FE_CXY plays a crucial role as an important feature for detecting lesion localization supported by a student's t-test, which identifies a statistically significant difference in the patterns between lesion and non-lesion localization (p<0.05). By incorporating this method into lung cancer detection systems, we anticipate improved accuracy and efficacy, thereby benefiting early diagnosis and treatment planning.
Bioengineering
Mass detection in mammograms has a limited approach to the presence of a mass in overlapping dens... more Mass detection in mammograms has a limited approach to the presence of a mass in overlapping denser fibroglandular breast regions. In addition, various breast density levels could decrease the learning system’s ability to extract sufficient feature descriptors and may result in lower accuracy performance. Therefore, this study is proposing a textural-based image enhancement technique named Spatial-based Breast Density Enhancement for Mass Detection (SbBDEM) to boost textural features of the overlapped mass region based on the breast density level. This approach determines the optimal exposure threshold of the images’ lower contrast limit and optimizes the parameters by selecting the best intensity factor guided by the best Blind/Reference-less Image Spatial Quality Evaluator (BRISQUE) scores separately for both dense and non-dense breast classes prior to training. Meanwhile, a modified You Only Look Once v3 (YOLOv3) architecture is employed for mass detection by specifically assigni...
International Journal of Integrated Engineering
This paper is presented a system that monitor the river water level by using computer vision with... more This paper is presented a system that monitor the river water level by using computer vision with image processing and IoT. This system is developed to detect riverbank level and river water level by applying image processing where edge detection technique is applied on both images captured by video camera. The flood severity level is determined by comparing the river water level and the riverbank level. Then, the determined flood severity is upload to IoT platform. A notification is sent the people when the flood severity level reached certain critical level via Telegram app which one of the social media applications. The available Raspberry Pi 3 Model B is usedas a controller in this system hardware device with the Raspberry Pi 5MP camera module.The IoT platform used is Ubidots where the user can be notified through it. The main contribution of this work is on the integration of computer vision with IoT Cloud as an early flood monitoring system in responding to climate change by d...
Journal of Electrical & Electronic Systems Research
In large-scale medical imaging, selecting the best image to extract relevant imaging biomarkers f... more In large-scale medical imaging, selecting the best image to extract relevant imaging biomarkers for image assessment is crucial. Segmentation of the left ventricle (LV) and myocardium are performed in computer-aided analysis usually at short-axis slices of cardiac magnetic resonance (MR) image to quantify cardiovascular disease assessment, such as myocardial scarring, LV ejection fraction and LV mass. The need to correctly identify a short-axis slice range for efficient quantification is preferred for automatic classification of the slice range of interest. The goal of this research is to establish an image processing method for the segmentation of Left ventricle scar from late gadolinium-enhanced (LGE) MR images. In order to achieve the main purpose, the work is divided into two parts, the first is to identify the cardiac Left ventricle segment (LVS) in the stack of short-axis LGE MR images and the second part; detecting the scar in between the LV myocardium area. This paper will present the outcomes of the first part by utilizing a deep convolutional neural network (DCNN) to construct an automatic system for classifying LVS and Non-Left Ventricle Segment (Non-LVS) in MR images. The same image dataset will be used for a comparative analysis with six DCNN models designed from scratch and three famed pre-trained networks, Alexnet, GoogleNet and SqueezeNet. Each model is trained up to 35 epochs using the Cardiac Atlas dataset and cross-validation method. The outcome from this work demonstrated that the DCNN3Y performs well over small training data with an average accuracy of 94.49%. Whereas SqueezeNet outperformed the three pre-trained networks with an average accuracy of 96.96%. It has also been discovered that increasing the number of filters and their subsequent configuration slightly influences the network's performance. This produces very promising results showing that it is ready to be used in the second part of this research. The outcome of this research can compensate for the deficiencies of manual detection in the original image detection system, increase detection efficiency, reduce detection misjudgments, and advance the development of automated and intelligent detection in the medical field.
Journal of Electrical & Electronic Systems Research
Screening mammography has been clinically practiced as a common method for monitoring any potenti... more Screening mammography has been clinically practiced as a common method for monitoring any potential breast diseases, especially in denser breasts among women. The advancement in medical imaging technology proved that the integration of artificial intelligence had given a significant impact on the breast screening process. However, due to various demographic patient backgrounds in clinical profile and non-standardized configurations of the developed intelligence models, such application is incapable of being applied by a health practitioner. Commonly, the deep learning method trained on a single network classifier has resulted in lower performance accuracy. With the motivation to improve the performance of classifying the dense breast, this paper proposes an improved classification model using deep learning approach of Convolutional Neural Network (CNN) and Support Vector Machine (SVM) employed on different sets of publicly established mammogram images of craniocaudal (CC) and medio-lateral (MLO) views. In this study, pre-trained CNN models, namely GoogleNet, ResNet50, ResNet101 and AlexNet, are used with SVM as a classifier. The proposed method for the classification of breast density region is superior to the existing methods as indicated from model performance quantitative results of accuracy, precision and area under the curve (AUC) of its receiver operating characteristic (ROC) curves. Significant improvement in model performance has been obtained using ResNet50 and GoogleNet with SVM classifier with > 94% accuracy and AUC > 0.95. Furthermore, the model is proved to have good feature extraction capabilities to work for various breast density images that can be further explored into detecting malignancy from screening mammogram images.
International Journal of Electrical and Computer Engineering (IJECE)
Lung cancer is the leading cause of cancer death among people worldwide. The primary aim of this ... more Lung cancer is the leading cause of cancer death among people worldwide. The primary aim of this research is to establish an image processing method for lung cancer detection. This paper focuses on lung region segmentation from computed tomography (CT) scan images. In this work, a new procedure for lung region segmentation is proposed. First, the lung CT scan images will undergo an image thresholding stage before going through two morphological reconstruction and masking stages. In between morphological and masking stages, object extraction, border change, and object elimination will occur. Finally, the lung field will be annotated. The outcomes of the proposed procedure and previous lung segmentation methods i.e., the modified watershed segmentation method is compared with the ground truth images for performance evaluation that will be carried out both in qualitative and quantitative manners. Based on the analyses, the new proposed procedure for lung segmentation, denotes better pe...
2017 International Conference on Electrical, Electronics and System Engineering (ICEESE), 2017
White matter hyperintensities (WMH) are small regions of high signal intensity that are observabl... more White matter hyperintensities (WMH) are small regions of high signal intensity that are observable on the white matter region of the brain through magnetic resonance imaging images. Generally, the medical expert conducts a white matter hyperintensities analysis to investigate brain tissue abnormality using manual or semi-automatic methods. However, those methods are prone to error and they establish unreliable results as different in rating scales. In this paper, a fully automatic method is proposed to identify WMH using the multimodal technique which combining image segmentation and enhancement. This method is introduced as an unsupervised method to automatically segment WMH on MRI images of T2-weighted and FLAIR sequences. Subsequently, the processed sequences are integrated by overlying the mapping images in order to map the most precise WMH regions. The accuracy of the WMH regions identification is assessed through the similarity index between automated and manual approach. The experimental results show that the proposed method has achieved significant results to detect exact WMH area. The proposed method is suitable to be implemented in analyzing white matter hyperintensities identification and it may serves as a computer-aided tool for radiologists.
Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Breast can... more Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Breast cancer is well known as a mortal disease. In United States, it is reported as the seccond leading cause of death in women after lung cancer
2022 IEEE 11th International Conference on Communication Systems and Network Technologies (CSNT)
IEEE Access
This study optimized the latest YOLOv5 framework, including its subset models, with training on d... more This study optimized the latest YOLOv5 framework, including its subset models, with training on different datasets that differed in image contrast and cloudiness to assess model performances based on quantitative metrics and image processing speed. The hyperparameter in the feature-extraction phase was configured based on the learning rate and momentum and further improved based on the adaptive moment estimation (ADAM) optimizer and the function reducing-learning-rate-on-plateau to optimize the model's training scheme. The optimized YOLOv5s achieved a better performance, with a mean average precision of 98.6% and a high inference speed of 106 frames per second. The ADAM optimizer with a detailed learning rate (0.0001) and momentum (0.99) fine-tuning yielded a sufficient convergence rate (0.69% at 55th epoch) to assist YOLOv5s in attaining a more precise detection for underwater objects. INDEX TERMS Image processing speed, object recognition, optimization model, tuning hyper-parameter, underwater imaging.
2021 6th IEEE International Conference on Recent Advances and Innovations in Engineering (ICRAIE), 2021
Cold Chain Management (CCM) can be described as the supply chain network of facilities and distri... more Cold Chain Management (CCM) can be described as the supply chain network of facilities and distribution that are controlled by temperature and humidity throughout the supply chain stages and entities. In CCM, the product will undergo the delivery process from the storeroom until it reaches the market cold room. However, the likelihood of accidents occurring during the process of delivery such as the breakdown of vehicles and the collapse of the freezing system can affect product quality. Therefore, an immediate response must be taken to prevent these setbacks from happening by implementing a real-time CCM system by utilizing the Sigfox UnaShield that stack into Arduino MakerUNO that read the temperature value and then sent to the Sigfox network. The data will be displayed in the Ubidots dashboard. An encouraging result is obtained on the prototyped used for the CCM monitoring using Ubidots Cloud and Xperanti Sigfox network.
Available online 30 March 2021 Dyslexia is developed by neurobiological in origin which is catego... more Available online 30 March 2021 Dyslexia is developed by neurobiological in origin which is categorized as learning disorder that affect the ability to read, spell, write and speak. The most common dyslexia symptom can easily be identified through the handwriting pattern. There are many intelligence and computational methods that have been proposed, and they have provided various and different performance to evaluate the proposed system ability. However, system performances are varied and nonstandardized in each assesment on dyslexic children to validate the presence of dyslexia symptom. The recent deep learning models have been employed to improve the assesment performance and (the models/ they have shown) shows significant output to detect and classify the present of dyslexia symptoms among school children. Therefore, there is a crucial need in deep learning, specifically for Convolutional Neural Network ( CNN) to validate performances of different networks, so that the most perfor...
Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Pengesanan... more Barah payudara telah dikenalpasti sebagai satu penyakit yang boleh mengakibatkan maut. Pengesanan dari peringkat awal dapat membantu mengurangkan kadar kematian. Mammografi dan ultrabunyi adalah dua kaedah yang sering digunakan untuk mengesan barah payudara. Namun masalah yang timbul ialah para doktor tidak dapat membuat keputusan dengan tepat kerana imej kabur, tidak jelas atau bercampur hingar. Oleh itu satu teknik pemprosesan imej dibangunkan sebagai satu perantara berguna bagi para doktor menentukan keputusan mereka. Dalam penyelidikan ini, satu sistem pemprosesan imej secara berkomputer telah dibangunkan bagi memproses imej-imej mammogram bertujuan mengesan mikrokalsifikasi, petanda paling umum kehadiran barah payudara. Tiga teknik prapemprosesan digunakan iaitu peningkatan imej, penurasan, dan peruasan mikrokalsifikasi. Dalam peruasan mikrokalsifikasi, teknik pengesanan pinggir digunakan bertujuan untuk meningkatkan ciri-ciri diagnostiknya dan seterusnya mengira bilangan mikro...
Journal of Electrical & Electronic Systems Research, 2021
This paper presents a lesion of lung cancer detection exist in CT scan images using watershed seg... more This paper presents a lesion of lung cancer detection exist in CT scan images using watershed segmentation. Lung cancer is a disease of uncontrolled cell growth in tissues of the lung. It seems to be the common cause of death among people throughout the world. Therefore, diagnosis of lung cancer at early stage can help doctors to treat patients and keep them alive. The main aim of this research is to establish an image processing method for segmentation of lung cancer from CT scan images by using image processing techniques. In order to achieve the main aims, the work is divided into two parts, first is obtaining lung region from CT scan images and second is detecting the lesion of lung cancer. Hence, this paper will present the outcome of the second part. The lung lobes and nodules or lesion in CT image are segmented using two techniques which are convolution watershed and modified watershed within two stage approaches. Firstly, the image will undergo threshold, clustering and imag...
Electrical Engineering & Electromechanics, 2021
Introduction. Known vibrational energy harvesting methods use a source of vibration to harvest el... more Introduction. Known vibrational energy harvesting methods use a source of vibration to harvest electric energy. Piezoelectric material works as a sensing element converted mechanical energy (vibration) to electrical energy (electric field). The existing piezoelectric energy harvesting (PEHs) devices have low sensitivity, low energy conversion, and low bandwidth. The novelty of the proposed work consists of the design of PEH’s structure. Air cavity was implemented in the design where it is located under the sensing membrane to improve sensitivity. Another novelty is also consisting in the design structure where the flexural membrane was located at the top of electrodes. The third novelty is a new design structure of printed circuit board (PCB). The purpose of improvised design is to increase the stress in between the edges of PEH and increase energy conversion. With the new structure of PCB, it will work as a substrate that absorbs surrounding vibration energy and transfers it to sen...