Vahid Ghods
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
Vahid Ghods has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Hassan Khalesi
Department of Electrical and Computer Engineering, Garmsar Branch, Islamic Azad University, Garmsar, Iran
15 shared publications Mohammad Karim Sohrabi
Department of Electrical and Computer Engineering, Se.C. Islamic Azad University, Semnan, Iran
8 shared publications- Ehsanollah Kabir
Department of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran
5 shared publications Arash Sabbaghi
Department of Electrical and Computer Engineering Se.C., Islamic Azad University Semnan Iran
4 shared publications- Elham Khoobjou
Department of Electronic Engineering, Garmsar Branch, Islamic Azad University, Garmsar, Iran
3 shared publications - Fatemeh Naiemi
Department of Electrical and Computer Engineering, Semnan Branch, Islamic Azad University, Semnan, Iran
3 shared publications
Latest Publications
- Open Access
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Image Processing
- Vol. 20 (1)
Realtime Data Augmentation for Breast Cancer Dataset: Dynamic Fine‐Tuning Bounding Box Coordinates and Segmentation Mask
Data augmentation is crucial for training deep learning models in breast cancer detection and segmentation, but conventional methods can cause annotation errors and visual artefacts, which harm instance‐level localisation, generalisation and clinical reliability. This study aims to develop an annotation‐aware, real‐time data augmentation framework that preserves spatial and clinical integrity during geometric transformations to improve model robustness and performance. A real‐time augmentation framework dynamically recalculates bounding box coordinates and segmentation masks during cropping and rotation. Instance‐level annotation correction is performed on‐the‐fly within the training pipeline, without offline preprocessing, additional storage, or generative modelling and is evaluated on the DDSM, INbreast and BUSI datasets. Experimental results show consistent performance gains across all datasets and models, with detection metrics improving by an average of 4.2–4.4% and segmentation accuracy increasing by up to 9.3%. The real‐time implementation achieves low preprocessing latency (≈0.12 s per batch, 3.75 ms per image), enabling high‐throughput training without added computational overhead. By preserving annotation integrity during geometric transformations, the proposed framework provides a computationally efficient and easily integrable solution for breast cancer imaging, with broader applicability to other medical image analysis tasks requiring precise spatial annotations.
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Image Processing
- Vol. 20 (1)
Open AccessRealtime Data Augmentation for Breast Cancer Dataset: Dynamic Fine‐Tuning Bounding Box Coordinates and Segmentation Mask
Data augmentation is crucial for training deep learning models in breast cancer detection and segmentation, but conventional methods can cause annotation errors and visual artefacts, which harm instance‐level localisation, generalisation and clinical reliability. This study aims to develop an annotation‐aware, real‐time data augmentation framework that preserves spatial and clinical integrity during geometric transformations to improve model robustness and performance. A real‐time augmentation framework dynamically recalculates bounding box coordinates and segmentation masks during cropping and rotation. Instance‐level annotation correction is performed on‐the‐fly within the training pipeline, without offline preprocessing, additional storage, or generative modelling and is evaluated on the DDSM, INbreast and BUSI datasets. Experimental results show consistent performance gains across all datasets and models, with detection metrics improving by an average of 4.2–4.4% and segmentation accuracy increasing by up to 9.3%. The real‐time implementation achieves low preprocessing latency (≈0.12 s per batch, 3.75 ms per image), enabling high‐throughput training without added computational overhead. By preserving annotation integrity during geometric transformations, the proposed framework provides a computationally efficient and easily integrable solution for breast cancer imaging, with broader applicability to other medical image analysis tasks requiring precise spatial annotations.
- Open Access
- 1 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-10
A novel D-latch with low energy consumption in quantum-dot cellular automata technology
Today, the implementation of computer systems and digital circuits is done using CMOS (complementary metal oxide semiconductor) technology, but problems such as high-power consumption and physical limitations of CMOS have led researchers to use the emerging technology such as quantum-dot cellular automata (QCA). In this paper, new structures for D-latch in QCA technology are presented. At first, a D-latch with 25 cells, a cross-sectional area of 0.02 µm2, delay of 0.5 clock cycle and 22.89 meV energy consumption is presented, and then, using the proposed D-latch, a reset-based D-latch is also proposed, which has more advantages than the previous structures and has only 29 cells, a cross-sectional area of 0.02µm2 and a delay of 0.5 clock cycles. Then the D-latch with set and reset terminals is proposed, which has better conditions than the previous structures and has only 29 cells, a cross-sectional area of 0.02 µm2 and a delay of 0.5 clock cycle. The proposed D-latch designs have better energy consumption than previous structures. The simulation of the proposed methods has been done with the help of QCADesigner and QCAPro tools and compared with other designs in terms of area and energy consumption.
- 1 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-10
Open AccessA novel D-latch with low energy consumption in quantum-dot cellular automata technology
Today, the implementation of computer systems and digital circuits is done using CMOS (complementary metal oxide semiconductor) technology, but problems such as high-power consumption and physical limitations of CMOS have led researchers to use the emerging technology such as quantum-dot cellular automata (QCA). In this paper, new structures for D-latch in QCA technology are presented. At first, a D-latch with 25 cells, a cross-sectional area of 0.02 µm2, delay of 0.5 clock cycle and 22.89 meV energy consumption is presented, and then, using the proposed D-latch, a reset-based D-latch is also proposed, which has more advantages than the previous structures and has only 29 cells, a cross-sectional area of 0.02µm2 and a delay of 0.5 clock cycles. Then the D-latch with set and reset terminals is proposed, which has better conditions than the previous structures and has only 29 cells, a cross-sectional area of 0.02 µm2 and a delay of 0.5 clock cycle. The proposed D-latch designs have better energy consumption than previous structures. The simulation of the proposed methods has been done with the help of QCADesigner and QCAPro tools and compared with other designs in terms of area and energy consumption.
- Open Access
- 15 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 87386-87400
BreastCNet: Breast Cancer Detection, Classification, and Localization Convolutional Neural Network With Advanced Optimization Techniques
Breast cancer remains a significant global health issue, primarily affecting females and requiring advanced detection methods to improve patient outcomes. While computer-aided diagnostic methods have progressed, limitations in accuracy persist. This study introduced BreastCNet, a Convolutional Neural Network (CNN) with hyperparameter optimization and a multi-task learning framework, enhancing classification and lesion localization. The BreastCNet integrated CNNs with the Grey Wolf Optimizer (GWO) and Parrot Optimizer (PO) in a dual-optimization approach. GWO fine-tuned neuron counts in dense layers to enhance feature learning, while PO dynamically adjusted the learning rate to improve convergence. Specifically, the parrot optimizer adjusted the learning rate from 0.001 to 0.00156, resulting in a 1.5% improvement in accuracy, and the grey wolf optimizer fine-tuned neuron numbers (Dense_1: 6, Dense_2: 866), contributing to a 2.3% increase in validation accuracy. These optimizations mitigated overfitting, accelerated convergence, and improved generalization, making the BreastCNet computationally efficient and clinically viable. The multi-task learning framework simultaneously performed breast ultrasound image classification (benign/malignant) and lesion localization via bounding box regression. The model achieved 98.10% validation accuracy, an AUC of 0.995, an F1-score of 0.98, and an IoU score of 0.96 for lesion localization on the Breast Ultrasound Images (BUSI) dataset, with comparable improvements on the BUSI, Digital Database for Screening Mammography (DDSM), and INbreast datasets. The BreastCNet is highly likely to advance breast cancer detection and lesion localization, providing precise localization and improved generalization, making it a promising tool for clinical applications.
- 15 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 87386-87400
Open AccessBreastCNet: Breast Cancer Detection, Classification, and Localization Convolutional Neural Network With Advanced Optimization Techniques
Breast cancer remains a significant global health issue, primarily affecting females and requiring advanced detection methods to improve patient outcomes. While computer-aided diagnostic methods have progressed, limitations in accuracy persist. This study introduced BreastCNet, a Convolutional Neural Network (CNN) with hyperparameter optimization and a multi-task learning framework, enhancing classification and lesion localization. The BreastCNet integrated CNNs with the Grey Wolf Optimizer (GWO) and Parrot Optimizer (PO) in a dual-optimization approach. GWO fine-tuned neuron counts in dense layers to enhance feature learning, while PO dynamically adjusted the learning rate to improve convergence. Specifically, the parrot optimizer adjusted the learning rate from 0.001 to 0.00156, resulting in a 1.5% improvement in accuracy, and the grey wolf optimizer fine-tuned neuron numbers (Dense_1: 6, Dense_2: 866), contributing to a 2.3% increase in validation accuracy. These optimizations mitigated overfitting, accelerated convergence, and improved generalization, making the BreastCNet computationally efficient and clinically viable. The multi-task learning framework simultaneously performed breast ultrasound image classification (benign/malignant) and lesion localization via bounding box regression. The model achieved 98.10% validation accuracy, an AUC of 0.995, an F1-score of 0.98, and an IoU score of 0.96 for lesion localization on the Breast Ultrasound Images (BUSI) dataset, with comparable improvements on the BUSI, Digital Database for Screening Mammography (DDSM), and INbreast datasets. The BreastCNet is highly likely to advance breast cancer detection and lesion localization, providing precise localization and improved generalization, making it a promising tool for clinical applications.
- 20 January 2025
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (28) , 34267-34284
Multi-Model emotion recognition from speech using StarGAN, DCNN, and SVM
- 20 January 2025
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (28) , 34267-34284
Multi-Model emotion recognition from speech using StarGAN, DCNN, and SVM
- Open Access
- 8 January 2024
- Published by Springer Nature in Research Square
Data Augmentation Improvement for a Breast Cancer Dataset: Fine-Tuning Bounding Box Coordinates and Segmentation Mask
Breast cancer is one of the leading causes of death among women worldwide, and early detection through medical imaging techniques is crucial for effective treatment. Deep learning models have shown promising results in medical image analysis tasks, but traditional data augmentation methods often do not preserve the accuracy of bounding box and segmentation mask annotations. To address this issue, a proposed method for fine-tuning new coordinates of bounding box and segmentation mask during data augmentation methods cropping and rotation in the breast cancer dataset has been introduced. This method involves generating new images by applying cropping and rotation to the original images and adjusting the coordinates of the bounding box and segmentation mask to match the new image. Experiments conducted on a publicly available breast cancer dataset showed that the proposed method improved the accuracy of the bounding box and segmentation mask annotations while preserving the original information in the image. The proposed method is a promising approach to improve the accuracy of deep learning models for medical image analysis tasks. By dynamically adjusting the coordinates during augmentation, the proposed method can better preserve object shape and improve the accuracy of object detection and segmentation tasks. The approach can be easily integrated into existing data augmentation pipelines and has the potential to improve performance on a range of computer vision applications.
- 8 January 2024
- Published by Springer Nature in Research Square
Open AccessData Augmentation Improvement for a Breast Cancer Dataset: Fine-Tuning Bounding Box Coordinates and Segmentation Mask
Breast cancer is one of the leading causes of death among women worldwide, and early detection through medical imaging techniques is crucial for effective treatment. Deep learning models have shown promising results in medical image analysis tasks, but traditional data augmentation methods often do not preserve the accuracy of bounding box and segmentation mask annotations. To address this issue, a proposed method for fine-tuning new coordinates of bounding box and segmentation mask during data augmentation methods cropping and rotation in the breast cancer dataset has been introduced. This method involves generating new images by applying cropping and rotation to the original images and adjusting the coordinates of the bounding box and segmentation mask to match the new image. Experiments conducted on a publicly available breast cancer dataset showed that the proposed method improved the accuracy of the bounding box and segmentation mask annotations while preserving the original information in the image. The proposed method is a promising approach to improve the accuracy of deep learning models for medical image analysis tasks. By dynamically adjusting the coordinates during augmentation, the proposed method can better preserve object shape and improve the accuracy of object detection and segmentation tasks. The approach can be easily integrated into existing data augmentation pipelines and has the potential to improve performance on a range of computer vision applications.
- 25 November 2023
- Published by Springer Nature in Soft Computing
- Vol. 28 (7-8) , 6039-6057
SAR image classification with convolutional neural network using modified functions
- 25 November 2023
- Published by Springer Nature in Soft Computing
- Vol. 28 (7-8) , 6039-6057
SAR image classification with convolutional neural network using modified functions
- Open Access
- 15 November 2023
- Published by Springer Nature in Research Square
BreastCDNet: Breast Cancer Detection Neural Network, Classification and Localization
Breast cancer remains a significant health concern globally, necessitating advanced detection methods for improved patient outcomes. Convolutional neural networks, CNN, have revolutionized object detection by automating the feature extraction process and allowing for the efficient localization and classification of objects within images. BreastCDNet introduces a novel approach to breast cancer detection using CNN. This pioneering CNN method is tailored specifically for the task of ultrasound breast cancer detection and localization, offering a streamlined and effective approach that significantly enhances diagnostic accuracy. Multi-task learning is leveraged by the proposed model, with simultaneous consideration of breast ultrasound image classification and bounding box regression for lesion localization. Intricate features from medical images are extracted by BreastCDNet architecture, facilitating both classification (benign or malignant) and object localization. Separate data generators are employed for classification and bounding box labels to optimize model training. The model's effectiveness is demonstrated by experimental results on the BUSI dataset, where exceptional classification accuracy and precise bounding box predictions are achieved. Key components, including convolutional layers for feature extraction, max-pooling layers for spatial reduction, fully connected layers for predictions, and specific output layers for classification and bounding box regression, are encompassed by the model architecture. The BreastCDNet model is a high-performance breast cancer detection model that achieved 99.14% training accuracy, 97.70% validation accuracy, 0.99 ROC AUC, and 0.97 F1-score on the training set. It is also capable of accurately localizing breast cancer lesions with an IOU score of 0.95.
- 15 November 2023
- Published by Springer Nature in Research Square
Open AccessBreastCDNet: Breast Cancer Detection Neural Network, Classification and Localization
Breast cancer remains a significant health concern globally, necessitating advanced detection methods for improved patient outcomes. Convolutional neural networks, CNN, have revolutionized object detection by automating the feature extraction process and allowing for the efficient localization and classification of objects within images. BreastCDNet introduces a novel approach to breast cancer detection using CNN. This pioneering CNN method is tailored specifically for the task of ultrasound breast cancer detection and localization, offering a streamlined and effective approach that significantly enhances diagnostic accuracy. Multi-task learning is leveraged by the proposed model, with simultaneous consideration of breast ultrasound image classification and bounding box regression for lesion localization. Intricate features from medical images are extracted by BreastCDNet architecture, facilitating both classification (benign or malignant) and object localization. Separate data generators are employed for classification and bounding box labels to optimize model training. The model's effectiveness is demonstrated by experimental results on the BUSI dataset, where exceptional classification accuracy and precise bounding box predictions are achieved. Key components, including convolutional layers for feature extraction, max-pooling layers for spatial reduction, fully connected layers for predictions, and specific output layers for classification and bounding box regression, are encompassed by the model architecture. The BreastCDNet model is a high-performance breast cancer detection model that achieved 99.14% training accuracy, 97.70% validation accuracy, 0.99 ROC AUC, and 0.97 F1-score on the training set. It is also capable of accurately localizing breast cancer lesions with an IOU score of 0.95.
- Open Access
- 27 September 2023
- Published by Springer Nature in Silicon
- Vol. 15 (15) , 6761
Correction to: Sensitivity Increment of Metamaterial-based Refractive Index Sensor with Silicon Layers
- 27 September 2023
- Published by Springer Nature in Silicon
- Vol. 15 (15) , 6761
Open AccessCorrection to: Sensitivity Increment of Metamaterial-based Refractive Index Sensor with Silicon Layers
- Open Access
- 9 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 86456-86469
Personality Recognition Based on Handwriting Types Using Fuzzy Inference
Graphology is the psycho-analysis study of personality and individual characteristics of writers based on the handwriting types. Graphology is a time-consuming and complicated task. Due to small number of expert graphologists, computerized measurement of graphology is an essential need. In this research, in addition to extraction of handwriting features, it is tried to categorize them in terms of regularity and inequality of letters and recognize person’s characteristics via graphology tables. This classification includes these types of handwriting: spontaneous, with difference in pen pressure, uncoordinated, coordinated with partial variance, uniform, sharp, regular, with inequality of letter’s inflexion and unequal. In this paper, 140 typical Farsi handwriting samples are used which are interpreted after giving MMPI (Minnesota Multiphasic Personality Inventory) questionnaire. The novel proposed features are used as the Mamdani fuzzy classifier inputs. Variance of baseline curves, variance of distances between words, variance of vertical letter height, variance of pen width, number of zero, acute and obtuse angles and their ratios to the number of text lines and slope angle are used as features for Mamdani fuzzy classifier inputs. In this study, there are nine handwriting types for personality recognition and 24 rules are defined for the nine fuzzy systems. Compared to our ground truth (MMPI results for the handwriting database), the proposed method showed an accuracy of 82.5% for personality recognition that demonstrated promising results.
- 9 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 86456-86469
Open AccessPersonality Recognition Based on Handwriting Types Using Fuzzy Inference
Graphology is the psycho-analysis study of personality and individual characteristics of writers based on the handwriting types. Graphology is a time-consuming and complicated task. Due to small number of expert graphologists, computerized measurement of graphology is an essential need. In this research, in addition to extraction of handwriting features, it is tried to categorize them in terms of regularity and inequality of letters and recognize person’s characteristics via graphology tables. This classification includes these types of handwriting: spontaneous, with difference in pen pressure, uncoordinated, coordinated with partial variance, uniform, sharp, regular, with inequality of letter’s inflexion and unequal. In this paper, 140 typical Farsi handwriting samples are used which are interpreted after giving MMPI (Minnesota Multiphasic Personality Inventory) questionnaire. The novel proposed features are used as the Mamdani fuzzy classifier inputs. Variance of baseline curves, variance of distances between words, variance of vertical letter height, variance of pen width, number of zero, acute and obtuse angles and their ratios to the number of text lines and slope angle are used as features for Mamdani fuzzy classifier inputs. In this study, there are nine handwriting types for personality recognition and 24 rules are defined for the nine fuzzy systems. Compared to our ground truth (MMPI results for the handwriting database), the proposed method showed an accuracy of 82.5% for personality recognition that demonstrated promising results.
- 10 June 2023
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 42 (11) , 6749-6779
Designing of an 8 × 8 Multiplier with New Inexact 4:2 Compressors for Image Processing Applications
- 10 June 2023
- Published by Springer Nature in Circuits, Systems, and Signal Processing
- Vol. 42 (11) , 6749-6779
Designing of an 8 × 8 Multiplier with New Inexact 4:2 Compressors for Image Processing Applications