t-viSNE: Interactive Assessment and Interpretation of t-SNE Projections

Chatzimparmpas, Angelos; Martins, Rafael Messias; Kerren, Andreas

doi:10.1109/tvcg.2020.2986996

Cited by 128 publications

(76 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, a previous anatomical study by Williams et al (38) has shown limited variance in ligament insertions between subjects. Finally, t-SNE clusters represent non-linear reductions in dimensionality, which remain difficult to interpret (39). Nevertheless, rapid evolving technology in this field has high potential for future algorithms interpreting these lowdimensional data clusters.…”

Section: Discussionmentioning

confidence: 99%

Towards Automated Ligamentous Injury Evaluation in Syndesmotic Ankle Lesions

Peiffer

Burssens

Mits

et al. 2021

Preprint

View full text Add to dashboard Cite

PurposeForced external rotation is hypothesized as the key mechanism of syndesmotic ankle injuries. This complex trauma pattern ruptures the syndesmotic ligaments, inducing a three-dimensional deviation from the normal distal tibiofibular joint configuration. However, current diagnostic imaging modalities are impeded by a two-dimensionalassessment, without taking into account ligamentous stabilizers. Therefore, our aim is two-fold: (1) to construct an articulated statistical shape model of the normal ankle with inclusion of ligamentous morphometry and (2) to apply this model in the assement of a clinical cohort of paient with syndesmotic ankle injuries.Methods Three-dimensional models of the distal tibiofibular joint were analyzed in asymptomatic controls (N= 76; Mean age 63 +/- 19 years),patients with syndesmotic ankle injury (N = 13; Mean age 35 +/- 15 years), and their healthy contralateral equivalent (N = 13). Subsequently, the statiscal shape model was generated after aligning all ankles based on the distal tibia. The position of the syndesmotic ligaments was predicted based on previously validated iterative shortest path calculation methodology. Evaluation of the model was described by means of accuracy, compactness and generalization. Canonical Correlation Analysis was performed to assess the influence of syndesmotic lesions on the distal tibiofibular joint congruency.ResultsOur presented model contained an accuracy of 0.23 +/- 0.028 mm. Mean prediction accuracy of ligament insertions was 0.53 +/- 12 mm. A statistically significant difference in anterior syndesmotic distance was found between ankles with syndesmotic lesions and healthy controls (95% CI [ 0.32 , 3.29], p = 0.017). There was a significant correlation between presence of syndesmotic injury and the morphological distal tibiofibular configuration (r = 0.873, p <0,001). ConclusionIn this study, we constructed a bony and ligamentous statistical model representing the distal tibiofibular joint Furthermore, the presented model was able to detect an elongation injury of the anterior inferior tibiofibular ligament after traumatic syndesmotic lesions in a clinical patient cohort.

show abstract

Section: Discussionmentioning

confidence: 99%

Towards Automated Ligamentous Injury Evaluation in Syndesmotic Ankle Lesions

Peiffer

Burssens

Mits

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Before the advent of word embedding technology, one-hot representation is a traditional method, but one-hot is too sparse to reflect the interrelationship between words. In addition, the principal component analysis (PCA) 49 and the T-distributed neighborhood embedding algorithm (T-SNE) 50 can be adopted to further reduce the dimension of the distributed representation in the word embedding space, thereby realizing the visualization of word embedding and word meaning induction. In view of this, word embedding technology is utilized in this paper to realize the distribution representation.…”

Section: Openmentioning

confidence: 99%

A deep learning framework combined with word embedding to identify DNA replication origins

Zhang

2021

Sci Rep

View full text Add to dashboard Cite

The DNA replication influences the inheritance of genetic information in the DNA life cycle. As the distribution of replication origins (ORIs) is the major determinant to precisely regulate the replication process, the correct identification of ORIs is significant in giving an insightful understanding of DNA replication mechanisms and the regulatory mechanisms of genetic expressions. For eukaryotes in particular, multiple ORIs exist in each of their gene sequences to complete the replication in a reasonable period of time. To simplify the identification process of eukaryote’s ORIs, most of existing methods are developed by traditional machine learning algorithms, and target to the gene sequences with a fixed length. Consequently, the identification results are not satisfying, i.e. there is still great room for improvement. To break through the limitations in previous studies, this paper develops sequence segmentation methods, and employs the word embedding technique, ‘Word2vec’, to convert gene sequences into word vectors, thereby grasping the inner correlations of gene sequences with different lengths. Then, a deep learning framework to perform the ORI identification task is constructed by a convolutional neural network with an embedding layer. On the basis of the analysis of similarity reduction dimensionality diagram, Word2vec can effectively transform the inner relationship among words into numerical feature. For four species in this study, the best models are obtained with the overall accuracy of 0.975, 0.765, 0.885, 0.967, the Matthew’s correlation coefficient of 0.940, 0.530, 0.771, 0.934, and the AUC of 0.975, 0.800, 0.888, 0.981, which indicate that the proposed predictor has a stable ability and provide a high confidence coefficient to classify both of ORIs and non-ORIs. Compared with state-of-the-art methods, the proposed predictor can achieve ORI identification with significant improvement. It is therefore reasonable to anticipate that the proposed method will make a useful high throughput tool for genome analysis.

show abstract

“…As a solution, t-SNE method will replace the normal distribution used in the low-dimensional space with a t-distribution with 1 of freedom. The probability density function (PDF) of t-distribution is shown in (17) where v is expressed as a degree of freedom. When v=1, it can be simplified to formula (18).…”

Section: 1) Crowding Problemmentioning

confidence: 99%

“…Feature extraction also plays a vital role in data-driven fault diagnosis and dimensionality reduction for the samples or datasets [16]. Nevertheless, the principal component analysis (PCA) is commonly used to find the main components of the original data and establishes a direct relationship between the high and low dimensional data sets, but it cannot capture the non-linear pattern [17]. Therefore, this research uses the t-distribution stochastic neighbor embedding (t-SNE) to reduce the dimensionality of nonlinear data, which can visualize high-dimensional complex signal patterns.…”

Section: Introductionmentioning

confidence: 99%

Induction Motor Fault Classification Based on ROC Curve and t-SNE

Lee

Lin

2021

IEEE Access

View full text Add to dashboard Cite

This paper proposes a novel fault classification method with application to induction motors, which is based on integrating and combining with receiver operating characteristic (ROC) curve and tdistribution stochastic neighbor embedding (t-SNE). According to the feature selection methods of ReliefF, symmetrical uncertainty (SU), and fast correlation-based filter (FCBF), the significant features were verified. Additionally, support vector machine (SVM), k-nearest neighbor (KNN), and decision tree (DT) are also considered as classifiers to identify the simulation results. To begin with, the current signals obtained from distinctive four topologies of working conditions of the motor, which includes healthy, bearing damage, broken rotor bar, and short circuit in stator windings, respectively. The potential feature set is extracted by using Hilbert-Huang transform (HHT) technique. Then, three feature selection methods are adopted to select three optimal feature subsets from the original feature set. Finally, the classification accuracy (ACC) and ROC curve are used to demonstrate the capability of classifiers' recognition. The results showed that the optimal feature subsets significantly reduce the number of selected features and improve the classification ACC and area under the curve (AUC) compared with the original feature set. In conclusion, the proposed method can downgrade the data, demonstrate the scatter plot more intuitively, and identify various types of faults, unlike with other fault diagnosis literature.

show abstract

t-viSNE: Interactive Assessment and Interpretation of t-SNE Projections

Cited by 128 publications

References 59 publications

Towards Automated Ligamentous Injury Evaluation in Syndesmotic Ankle Lesions

Towards Automated Ligamentous Injury Evaluation in Syndesmotic Ankle Lesions

A deep learning framework combined with word embedding to identify DNA replication origins

Induction Motor Fault Classification Based on ROC Curve and t-SNE

Contact Info

Product

Resources

About