Comparison of five cluster validity indices performance in brain [<sup>18</sup>F]<scp>FET</scp>‐<scp>PET</scp> image segmentation using <i>k</i>‐means

Abualhaj, Bedor; Weng, Guoyang; Ong, Melissa; Attarwala, Ali Asgar; Molina, Flavia; Büsing, Karen A.; Glatting, Gerhard

doi:10.1002/mp.12025

Cited by 16 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…1. The silhouette coefficient (SC) [28] is a way to evaluate the clustering effect, which can evaluate clustering results for different algorithms based on the same original data. The SC of clustering results is in the range (−1, 1).…”

Section: Dfcm Clustering Analysismentioning

confidence: 99%

Multi-Class Freeway Congestion and Emission Based on Robust Dynamic Multi-Objective Optimization

2021

View full text Add to dashboard Cite

In order to solve the problem of traffic congestion and emission optimization of urban multi-class expressways, a robust dynamic nondominated sorting multi-objective genetic algorithm DFCM-RDNSGA-III based on density fuzzy c-means clustering method is proposed in this paper. Considering the three performance indicators of travel time, ramp queue and traffic emissions, the ramp metering and variable speed limit control schemes of an expressway are optimized to improve the main road and ramp traffic congestion, therefore achieving energy conservation and emission reduction. In the VISSIM simulation environment, a multi-on-ramp and multi-off-ramp road network is built to verify the performance of the algorithm. The results show that, compared with the existing algorithm NSGA-III, the DFCM-RDNSGA-III algorithm proposed in this paper can provide better ramp metering and variable speed limit control schemes in the process of road network peak formation and dissipation. In addition, the traffic congestion of expressways can be improved and energy conservation as well as emission reduction can also be realized.

show abstract

Section: Dfcm Clustering Analysismentioning

confidence: 99%

Multi-Class Freeway Congestion and Emission Based on Robust Dynamic Multi-Objective Optimization

2021

View full text Add to dashboard Cite

show abstract

“…As for cluster validation, unfortunately, there is no best Cluster Validity Index (CVI) for the k-Shape clustering algorithm. In this study, a proper CVI is used to compare the clustering result of k-Shape clustering algorithm with k-means clustering algorithm [27].…”

Section: A K-shape Clustering Algorithm For Driving Cyclesmentioning

confidence: 99%

A Convolutional Neural Network-Based Driving Cycle Prediction Method for Plug-in Hybrid Electric Vehicles With Bus Route

2020

View full text Add to dashboard Cite

Driving cycle prediction plays a key role in energy management strategy (EMS) for hybrid electric vehicles (HEVs). This paper studies a driving cycle prediction method based on convolutional neural network (CNN). Firstly, the k-shape clustering method is used to group the driving cycle data into six different types. Moreover, this method is compared with the k-means algorithm which is often used for clustering driving cycles. Secondly, CNN is adopted to predict the different types of the driving cycles based on the results of k-Shape clustering. Some basic features are selected to construct the input of the networks with no assistance of human experience. In the process of training neural networks, some high-level features which can describe the information of a driving cycle more accurately are extracted, and the deep neural networks are built, which are different from traditional experience-based driving cycle prediction methods. And then, the better performance of the proposed method is illustrated by making a comparison with the traditional machine learning method. Finally, an adaptive energy management strategy for plug-in hybrid electric buses (PHEB) based on deep learning is given, and simulation results prove the effectiveness of the proposed method. INDEX TERMS Plug-in hybrid electric bus, driving cycle prediction, energy management strategy, deep learning.

show abstract

“…The extracted radiomics features serve as computational biomarkers of image intensity, texture, and morphology descriptors, which are subsequently used by classifiers (e.g., logistic regression, supporting vector machine, random forest, etc.) to distinguish glioma from normal tissue 24–28 . Driven by recent developments in algorithms and increased computational power, deep learning methods, especially convolutional neural networks (CNNs), have been adopted as a new approach for the automatic glioma segmentation 11,29–31 .…”

Section: Introductionmentioning

confidence: 99%

“…to distinguish glioma from normal tissue. [24][25][26][27][28] Driven by recent developments in algorithms and increased computational power, deep learning methods, especially convolutional neural networks (CNNs), have been adopted as a new approach for the automatic glioma segmentation. 11,[29][30][31] The fully convolutional network (FCN), one of the early efforts of CNN-based segmentation models, 32 directly learns the image features at multiple scales by repeated convolutional layers and pooling layers.…”

Section: Introductionmentioning

confidence: 99%

A neural ordinary differential equation model for visualizing deep neural network behaviors in multi‐parametric MRI‐based glioma segmentation

Yang

et al. 2023

Medical Physics

View full text Add to dashboard Cite

PurposeTo develop a neural ordinary differential equation (ODE) model for visualizing deep neural network behavior during multi‐parametric MRI‐based glioma segmentation as a method to enhance deep learning explainability.MethodsBy hypothesizing that deep feature extraction can be modeled as a spatiotemporally continuous process, we implemented a novel deep learning model, Neural ODE, in which deep feature extraction was governed by an ODE parameterized by a neural network. The dynamics of (1) MR images after interactions with the deep neural network and (2) segmentation formation can thus be visualized after solving the ODE. An accumulative contribution curve (ACC) was designed to quantitatively evaluate each MR image's utilization by the deep neural network toward the final segmentation results.The proposed Neural ODE model was demonstrated using 369 glioma patients with a 4‐modality multi‐parametric MRI protocol: T1, contrast‐enhanced T1 (T1‐Ce), T2, and FLAIR. Three Neural ODE models were trained to segment enhancing tumor (ET), tumor core (TC), and whole tumor (WT), respectively. The key MRI modalities with significant utilization by deep neural networks were identified based on ACC analysis. Segmentation results by deep neural networks using only the key MRI modalities were compared to those using all four MRI modalities in terms of Dice coefficient, accuracy, sensitivity, and specificity.ResultsAll Neural ODE models successfully illustrated image dynamics as expected. ACC analysis identified T1‐Ce as the only key modality in ET and TC segmentations, while both FLAIR and T2 were key modalities in WT segmentation. Compared to the U‐Net results using all four MRI modalities, the Dice coefficient of ET (0.784→0.775), TC (0.760→0.758), and WT (0.841→0.837) using the key modalities only had minimal differences without significance. Accuracy, sensitivity, and specificity results demonstrated the same patterns.ConclusionThe Neural ODE model offers a new tool for optimizing the deep learning model inputs with enhanced explainability. The presented methodology can be generalized to other medical image‐related deep‐learning applications.

show abstract

Comparison of five cluster validity indices performance in brain [¹⁸F]FET‐PET image segmentation using k‐means

Cited by 16 publications

References 47 publications

Multi-Class Freeway Congestion and Emission Based on Robust Dynamic Multi-Objective Optimization

Multi-Class Freeway Congestion and Emission Based on Robust Dynamic Multi-Objective Optimization

A Convolutional Neural Network-Based Driving Cycle Prediction Method for Plug-in Hybrid Electric Vehicles With Bus Route

A neural ordinary differential equation model for visualizing deep neural network behaviors in multi‐parametric MRI‐based glioma segmentation

Contact Info

Product

Resources

About

Comparison of five cluster validity indices performance in brain [18F]FET‐PET image segmentation using k‐means

Cited by 16 publications

References 47 publications

Multi-Class Freeway Congestion and Emission Based on Robust Dynamic Multi-Objective Optimization

Multi-Class Freeway Congestion and Emission Based on Robust Dynamic Multi-Objective Optimization

A Convolutional Neural Network-Based Driving Cycle Prediction Method for Plug-in Hybrid Electric Vehicles With Bus Route

A neural ordinary differential equation model for visualizing deep neural network behaviors in multi‐parametric MRI‐based glioma segmentation

Contact Info

Product

Resources

About

Comparison of five cluster validity indices performance in brain [¹⁸F]FET‐PET image segmentation using k‐means