A Survey on Graph Representation Learning Methods

Khoshraftar, Shima; An, Aijun

doi:10.1145/3633518

Cited by 20 publications

(2 citation statements)

References 214 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although there is ongoing research about graph representation using machine learning 36 which aims to reduce the sizes of large graphs. Since our graph representation method fits the entire graph into a multidimensional array with the same size as the segmentation data, there was no need to compress the graph using alternative methods.…”

Section: Proposed Frameworkmentioning

confidence: 99%

A new surgical path planning framework for neurosurgery

Kurt Pehlivanoğlu,

Ay,

Eker

et al. 2023

Robotics Computer Surgery

View full text Add to dashboard Cite

BackgroundDespite using a variety of path‐finding algorithms that use tracts, the most significant advancement in this study is considering the values of all brain areas by doing atlas‐based segmentation for a more precise search. Our motivation comes from the literature’s shortcomings in designing and implementing path‐planning methods. Since planning paths with curvatures is a complex problem that requires considering many surgical and physiological constraints, most path‐planning strategies focus on straight paths. There is also a lack of studies that focus on the complete structure of the brain with the tracks, veins, and segmented areas. Instrument dependence is another inadequacy of the methods proposed in the literature.AimsThe aim of this study is to design a new surgical path planning framework that helps to plan the surgical path independently of the instrument, considers the entire structure of the brain, and allows curvilinear surgical paths. Thus, neurosurgeons can generate patient‐specific possible optimal surgical pathways before the neurosurgical procedure.Materials & MethodsThe proposed framework includes different path‐finding algorithms (Dijkstra, A*, and their aggressive variants) that find optimal paths by taking the risk scores (surgeons assessed all the segmented regions, considering the extent of damage. In this evaluation, scores ranged from “0 to 10,” with the most critical areas receiving a score of “10,” while the least possible affected areas were assigned a score of "0") for sensitive brain areas into consideration. For the tract image processing the framework includes fractional anisotropy (FA), relative anisotropy (RA), spherical measure (SM), and linear measure (LM) methods.ResultsThis is the first paper to handle tracts and atlas‐based segmentation of the human brain altogether under a framework for surgical path planning. The framework has a dynamic structure that gives the flexibility to add different path‐finding algorithms and generate different widths of surgical pathways. Moreover, surgeons can update the score table to guarantee minimally invasive surgery. The output file format of all the extracted surgical paths is NRRD, so it can be easily visualised, analysed, or processed over the third part software tools.DiscussionIn this study, we generated many possible surgical pathways then these pathways were evaluated by the surgeons the results were impressive because the framework could identify surgical pathways used in real‐world surgery that correspond to the standard pathways such as anterior transsylvian, trans sulcal, transgyral, and sub‐temporal.ConclusionThis study proposes a new surgical path planning framework for neurosurgery. Moreover, in the future by adding/adopting different parameters (such as operation time, and short and long‐term complications after surgery) to the proposed framework, it would be possible to find new surgical pathways for difficult surgical conditions.

show abstract

Section: Proposed Frameworkmentioning

confidence: 99%

A new surgical path planning framework for neurosurgery

Kurt Pehlivanoğlu,

Ay,

Eker

et al. 2023

Robotics Computer Surgery

View full text Add to dashboard Cite

show abstract

“…To properly interpret and utilize graph data, it is critical to learn meaningful node representations within these complex network topologies. In light of this, Graph Neural Networks (GNNs) have become a powerful paradigm that presents a hopeful resolution to this problem [1]. GNNs facilitate efficient node classification, graph classification, and link prediction, among other tasks, by encoding both the local and global graph structure [2].…”

Section: Introductionmentioning

confidence: 99%

Residual Attention Augmentation Graph Neural Network for Improved Node Classification

Abbas,

Ali,

Smarandache

et al. 2024

Eng. Technol. Appl. Sci. Res.

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) have emerged as a powerful tool for node representation learning within graph structures. However, designing a robust GNN architecture for node classification remains a challenge. This study introduces an efficient and straightforward Residual Attention Augmentation GNN (RAA-GNN) model, which incorporates an attention mechanism with skip connections to discerningly weigh node features and overcome the over-smoothing problem of GNNs. Additionally, a novel MixUp data augmentation method was developed to improve model training. The proposed approach was rigorously evaluated on various node classification benchmarks, encompassing both social and citation networks. The proposed method outperformed state-of-the-art techniques by achieving up to 1% accuracy improvement. Furthermore, when applied to the novel Twitch social network dataset, the proposed model yielded remarkably promising results. These findings provide valuable insights for researchers and practitioners working with graph-structured data.

show abstract

SBGMN: A Multi-view Sign Prediction Network for Bipartite Graphs

Jiang,

Yu,

et al. 2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

A Survey on Graph Representation Learning Methods

Cited by 20 publications

References 214 publications

A new surgical path planning framework for neurosurgery

A new surgical path planning framework for neurosurgery

Residual Attention Augmentation Graph Neural Network for Improved Node Classification

SBGMN: A Multi-view Sign Prediction Network for Bipartite Graphs

Contact Info

Product

Resources

About