TPCNN: Two-path convolutional neural network for tumor and liver segmentation in CT images using a novel encoding approach

Aghamohammadi, Amirhossein; Ranjbarzadeh, Ramin; Naiemi, Fatemeh; Mogharrebi, Marzieh; Dorosti, Shadi; Bendechache, Malika

doi:10.1016/j.eswa.2021.115406

Cited by 68 publications

(18 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ability to detect arrangements of data or characteristics is called pattern recognition. In other words, Pattern recognition can be categorized as a classification method based on knowledge already obtained or on statistical information mined from patterns or their representation and is implemented in the domain of computer vision for countless applications like recommendation systems, data mining, and biological imaging 28–31 …”

Section: Methodsmentioning

confidence: 99%

“…In other words, Pattern recognition can be categorized as a classification method based on knowledge already obtained or on statistical information mined from patterns or their representation and is implemented in the domain of computer vision for countless applications like recommendation systems, data mining, and biological imaging. [28][29][30][31] F I G U R E 2 A graphical demonstration of utilizing clustering method to two various time span for a user In today's pattern recognition strategies and their applications in many fields, the convolutional neural network (CNN) structures illustrate a massive breakthrough in data analyzing and processing. The CNN models principally deduce the relation between some key details, textural content and are utilized at the core of every model from data mining to the prediction of visiting new sites by people.…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

DeePOF: A hybrid approach of deep convolutional neural network and friendship to Point‐of‐Interest (POI) recommendation system in location‐based social networks

Safavi

Jalali

2022

Concurrency and Computation

View full text Add to dashboard Cite

Today, millions of active users spend a percentage of their time on location-based social networks like Yelp and Gowalla and share their rich information. They can easily learn about their friends' behaviors and where they are visiting and be influenced by their style. As a result, the existence of personalized recommendations and the investigation of meaningful features of users and Point of Interests (POIs), given the challenges of rich contents and data sparsity, is a substantial task to accurately recommend the POIs and interests of users in location-based social networks (LBSNs). This work proposes a novel pipeline of POI recommendations named DeePOF based on deep learning and the convolutional neural network. This approach only takes into consideration the influence of the most similar pattern of friendship instead of the friendship of all users. The mean-shift clustering technique is used to detect similarity. The most similar friends' spatial and temporal features are fed into our deep CNN technique. The output of several proposed layers can predict latitude and longitude and the ID of subsequent appropriate places, and then using the friendship interval of a similar pattern, the lowest distance venues are chosen. This combination method is estimated on two popular datasets of LBSNs. Experimental results demonstrate that analyzing similar friendships could make recommendations more accurate and the suggested model for recommending a sequence of top-k POIs outperforms state-of-the-art approaches.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Convolutional Neural Networkmentioning

confidence: 99%

DeePOF: A hybrid approach of deep convolutional neural network and friendship to Point‐of‐Interest (POI) recommendation system in location‐based social networks

Safavi

Jalali

2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…It is important to mention that each fully connected layer is followed by an activation function such as ReLU except the last nonlinear function that is usually different from the others. The last activation function, which is applied in the classification task, is Softmax to obtain a probability of the input being in the specified class [ 58 , 63 , 64 ].…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

Investigation of Effectiveness of Shuffled Frog-Leaping Optimizer in Training a Convolution Neural Network

Saadi

Sarshar

Sadeghi

et al. 2022

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

One of the leading algorithms and architectures in deep learning is Convolution Neural Network (CNN). It represents a unique method for image processing, object detection, and classification. CNN has shown to be an efficient approach in the machine learning and computer vision fields. CNN is composed of several filters accompanied by nonlinear functions and pooling layers. It enforces limitations on the weights and interconnections of the neural network to create a good structure for processing spatial and temporal distributed data. A CNN can restrain the numbering of free parameters of the network through its weight-sharing property. However, the training of CNNs is a challenging approach. Some optimization techniques have been recently employed to optimize CNN’s weight and biases such as Ant Colony Optimization, Genetic, Harmony Search, and Simulated Annealing. This paper employs the well-known nature-inspired algorithm called Shuffled Frog-Leaping Algorithm (SFLA) for training a classical CNN structure (LeNet-5), which has not been experienced before. The training method is investigated by employing four different datasets. To verify the study, the results are compared with some of the most famous evolutionary trainers: Whale Optimization Algorithm (WO), Bacteria Swarm Foraging Optimization (BFSO), and Ant Colony Optimization (ACO). The outcomes demonstrate that the SFL technique considerably improves the performance of the original LeNet-5 although using this algorithm slightly increases the training computation time. The results also demonstrate that the suggested algorithm presents high accuracy in classification and approximation in its mechanism.

show abstract

“…In addition, traditional approaches tend not to have enough capacity to adapt to the pixel-level segmentation tasks, especially when the segmentation target varies greatly. Contrary to traditional approaches, Deep learning-based methods such as FCN [5] and UNet [6] have shown a strong ability to fit the complex nonlinear segmentation tasks and they have achieved the best performance on many tasks such as liver [7][8][9], kidney [10,11], Cardiomyopathy [12] and small tissues [13]. However, when conducting pancreas segmentation, deep learning-based networks seem to be easily confused by the complex and variable background as the pancreas often accounts for less than 0.5% of the network input [14].…”

Section: Introductionmentioning

confidence: 99%

CMFCUNet: cascaded multi-scale feature calibration UNet for pancreas segmentation

et al. 2022

View full text Add to dashboard Cite

Segmenting the pancreas from abdominal CT scans is challenging since it often takes up a relatively small region. Researchers suggested leveraging coarse-to-fine approaches to cope with this challenge. However, the coarse-scaled segmentation and the fine-scaled segmentation are either trained separately utilizing the coordinates located by the coarse-scaled segmentation mask to crop the fine-scaled segmentation input, or trained jointly utilizing the coarse-scaled segmentation mask to enhance the finescaled segmentation input. We argued that these two solutions are complementary to some extent and can promote each other to improve the performance of pancreas segmentation. In addition, the backbone in the coarse-scaled segmentation and fine-scaled segmentation is mostly based on UNet or UNet-like networks, where the multi-scale features transmitted from the encoder to the decoder have not been explored for vertical calibration before. In this paper, we propose a cascaded multi-scale feature calibration UNet (CMFCUNet) for pancreas segmentation where the multi-scale features in the backbone of each scaled segmentation are calibrated vertically in a pixel-wise fashion. Besides, the coarse-scaled segmentation and the fine-scaled segmentation are connected by leveraging a designed dual enhancement module (DEM). Experiments are conducted on the public NIH pancreas dataset. First, when leveraging CMFCUNet, our method increased by over 3% on the Jaccard index (JI) Article Title and nearly 1% on dice similarity coefficient (DSC) which surpasses all existing pancreas segmentation approaches. In addition, our experiments demonstrate that CMFCUNet improved the coarse-to-fine segmentation framework and outperformed the mainstream coarse-to-fine pancreas segmentation approaches. Furthermore, we also conducted ablation studies to analyze the effectiveness of the backbone (MFCUNet) and the DEM.

show abstract

TPCNN: Two-path convolutional neural network for tumor and liver segmentation in CT images using a novel encoding approach

Cited by 68 publications

References 87 publications

DeePOF: A hybrid approach of deep convolutional neural network and friendship to Point‐of‐Interest (POI) recommendation system in location‐based social networks

DeePOF: A hybrid approach of deep convolutional neural network and friendship to Point‐of‐Interest (POI) recommendation system in location‐based social networks

Investigation of Effectiveness of Shuffled Frog-Leaping Optimizer in Training a Convolution Neural Network

CMFCUNet: cascaded multi-scale feature calibration UNet for pancreas segmentation

Contact Info

Product

Resources

About