Classification-Segmentation Pipeline for MRI via Transfer Learning and Residual Networks

Duong-Trung, Nghia; Ha, Dung Ngoc Le; Huynh, Hiep Xuan

doi:10.15439/2021r14

Cited by 4 publications

(4 citation statements)

References 19 publications

(19 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the YOLOv8 models are trained using their original weights instead of starting from scratch, there's no need for a large dataset. This approach leverages an important advantage known as transfer learning [28,29], which allows for the training of models on custom datasets to refine pre-existing weights [30]. Subsequently, labels were created for these datasets.…”

Section: Datasetsmentioning

confidence: 99%

Integrating YOLOv8-agri and DeepSORT for Advanced Motion Detection in Agriculture and Fisheries

Duong-Trung,

Duong-Trung

2024

EAI Endorsed Trans Ind Net Intel Syst

Self Cite

View full text Add to dashboard Cite

This paper integrates the YOLOv8-agri models with the DeepSORT algorithm to advance object detection and tracking in the agricultural and fisheries sectors. We address the current limitations in object classification by adapting YOLOv8 to the unique demands of these environments, where misclassification can hinder operational efficiency. Through the strategic use of transfer learning on specialized datasets, our study refines the YOLOv8-agri models for precise recognition and categorization of diverse biological entities. Coupling these models with DeepSORT significantly enhances motion tracking, leading to more accurate and reliable monitoring systems. The research outcomes identify the YOLOv8l-agri model as the optimal solution for balancing detection accuracy with training time, making it highly suitable for precision agriculture and fisheries applications. We have publicly made our experimental datasets and trained models publicly available to foster reproducibility and further research. This initiative marks a step forward in applying sophisticated computer vision techniques to real-world agricultural and fisheries management.

show abstract

Section: Datasetsmentioning

confidence: 99%

Integrating YOLOv8-agri and DeepSORT for Advanced Motion Detection in Agriculture and Fisheries

Duong-Trung,

Duong-Trung

2024

EAI Endorsed Trans Ind Net Intel Syst

Self Cite

View full text Add to dashboard Cite

show abstract

“…Convolutional Neural Networks (CNNs), a type of artificial neural network, have become essential in various computer vision operations, and are receiving more attention across different published studies. For example, artificial intelligence association into brain magnetic resonance imaging (MRI) in cancer diagnosis [20], audiogram classification method [21], and cancerous region detection in the prostate [22]. The basic layers in a CNN include: convolution layer, pooling layer, fully connected layer, which are changed in number and arrangement to create suitable training models for different problems.…”

Section: B Related Workmentioning

confidence: 99%

“…In this century of growing technology, imaging-based diagnostic techniques have been widely used, especially computer-aided diagnostic (CAD) systems that have been developed to help physicians in the diagnostic and therapeutic process [8][9][10]20]. In contrast, to the conventional thyroid diagnostic methods mentioned above, CAD methods used ultrasound images of thyroid nodules as input and provide information about nodules (benign or malignant) [2][3][4][5][6], thus limiting the unevenness of different doctor9s qualifications at different hospital levels and leading to faster initial diagnosis.…”

Section: Introductionmentioning

confidence: 99%

A Study on Thyroid Nodule Image Classification System Using Small Amount of Training Samples

Hoang¹,

Pham²,

Trinh³

2022

Proceedings of the Seventh International Conference on Research in Intelligent and Computing in Engineering

View full text Add to dashboard Cite

4To reduce errors caused by traditional diagnostic methods that rely heavily on physician experience, the diagnostic systems based on computer-aided have been researched and developed to assist physicians in diagnosing thyroid disease. Therefore, performance of the computer systems plays an important role to improve the quality of diagnostic processes. Although there has been a number of publish related to this issue, those studies still have limitations in which needing large data sets for training classification models is considered the most concerning limitation of previous studies. To solve this limitations, in this work, a classification method using artificial intelligence with a small amount of data to analyze thyroid ultrasound images was proposed. Thus we can save time and effort for data collection and the classification model processing time. Through baseline tests with an open data set, especially thyroid digital image database (TDID), the proposed method has improved the limitations of previous methods.

show abstract

“…[30], pixel restoration, and many more. Moreover, this technique is also quite successful in various healthcare applications such as classification-Segmentation Pipeline for MRI [31], Retinal Blood Vessel Segmentation [33], abnormality detection, classification from medical images, and mammogram image analysis [34]. Dhivya et al [17] proposed a model for enhanced tumor classification using pretrained neural network VGG16.…”

mentioning

confidence: 99%

Transfer Learning Assisted Classification of Artefacts Removed and Contrast Improved Digital Mammograms

Oza¹,

Sharma²,

Patel³

2022

SCPE

View full text Add to dashboard Cite

Mammograms are essential radiological images used to diagnose breast cancer well in advance. However, an accurate diagnosis also depends on the quality of mammogram images. Therefore, removal of artefacts and mammogram enhancement are necessary pre-processing steps. Artefact removal helps exclude unsolicited regions in the mammograms and limits the search for suspicious regions without excessive impact from the background. Mammogram enhancements improve apparent visual details and improve some features of an image. In this paper, we propose a method for mammogram pre-processing. These pre-processed mammograms are then fed into Deep Convolutional Neural Network for the classification process. Two approaches are used and compared to classify mammograms; Training model from scratch and Transfer Learning. Transfer Learning is an excellent approach to dealing with the small-sized training set, allowing us to consume the extendibility of deep learning entirely. By employing VGG16 as a pre-trained network on the pre-processed MIAS dataset, we improved training accuracy (96.14\%) compared to the model developed from scratch and other strategies described in the literature.

show abstract

Classification-Segmentation Pipeline for MRI via Transfer Learning and Residual Networks

Cited by 4 publications

References 19 publications

Integrating YOLOv8-agri and DeepSORT for Advanced Motion Detection in Agriculture and Fisheries

Integrating YOLOv8-agri and DeepSORT for Advanced Motion Detection in Agriculture and Fisheries

A Study on Thyroid Nodule Image Classification System Using Small Amount of Training Samples

Transfer Learning Assisted Classification of Artefacts Removed and Contrast Improved Digital Mammograms

Contact Info

Product

Resources

About