Outlier Detection: Techniques and Applications

Suri, Niranjan; M, Narasimha Murty; Athithan, G.

doi:10.1007/978-3-030-05127-3

Cited by 32 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the regression error between our predicted pose and the ground truth, we adopt a widely used evaluation method, mean absolute error (MAE) [44], in our experiments for different models. MAE computes the absolute costs between the real and the predicted slice position or orientation:…”

Section: Comparison Methods and Evaluation Metricsmentioning

confidence: 99%

Automatic Plane Pose Estimation for Cardiac Left Ventricle Coverage Estimation via Deep Adversarial Regression Network

Zhang,

Bronik,

Piechnik

et al. 2024

IEEE Trans. Artif. Intell.

View full text Add to dashboard Cite

Accurate segmentation of the ventricles plays a crucial role in determining cardiac functional parameters such as ventricular volume, ventricular mass, or ejection fraction. However, poor image quality, such as inadequate coverage of the left ventricle (LV) and right ventricle (RV) in cardiac magnetic resonance (CMR) image sequences, can significantly affect the assessment of cardiac function. This study investigates issues related to missing or corrupted imaging planes, which often lead to incomplete ventricle coverage. To address the challenge of estimating ventricle coverage in CMR images regardless of variations in imaging parameters such as device type, magnetic field strength, and protocol execution, we introduce a novel convolutional neural network (CNN) based on adversarial learning. Additionally, we integrate supplementary information (e.g., cross-view image data) as privileged information to enhance the interpretability of our model's predictions and identify potential biases or inaccuracies. This research represents the first attempt to automatically estimate ventricular coverage by identifying missing slices and plane orientations in CMR images using a dataset-agnostic approach. The effectiveness of the proposed model is demonstrated through the evaluation of datasets from three diverse and sizable image acquisition cohorts, demonstrating superior performance compared to existing methods.Impact Statement-Cardiac functional parameters, such as the ejection fraction of both ventricles and cardiac output, are crucial clinical indicators of cardiac function, providing insights into whether it is within normal or abnormal ranges. Accurate calculation of these parameters is based on precise measurements of ventricular volumes at the end of diastole and systole. The accuracy of volume measurements depends on correctly determining the heart's position, including its location, orientation, and size in CMR image sequences, thus establishing the full extent of the left ventricle (LV) and right ventricle (RV).This study introduces a fully automatic detection method to identify missing slices and estimate heart pose parameters in CMR volumes, which is robust across different datasets. Unlike previous research that focused solely on identifying missing base or apex slices to assess ventricular coverage, this study goes

show abstract

Section: Comparison Methods and Evaluation Metricsmentioning

confidence: 99%

Automatic Plane Pose Estimation for Cardiac Left Ventricle Coverage Estimation via Deep Adversarial Regression Network

Zhang,

Bronik,

Piechnik

et al. 2024

IEEE Trans. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…First of all, the distance-based methods identify outliers by calculating pairwise distances for the available data samples. A sample whose average distance from the majority of the other data samples exceeds a certain pre-specified threshold is labeled as an outlier [22], [23]. Angiulli et al [24] followed this approach to identify outliers in a highdimensional space.…”

Section: Related Work a Conventional Outlier Detection Methodsmentioning

confidence: 99%

Dynamic Construction of Outlier Detector Ensembles With Bisecting K-Means Clustering

et al. 2023

View full text Add to dashboard Cite

Outlier detection (OD) is a key problem, for which numerous solutions have been proposed. To deal with the difficulties associated with outlier detection across various domains and data characteristics, ensembles of outlier detectors have recently been employed to improve the performance of individual outlier detectors. In this paper, we follow an ensemble outlier detection approach in which good outlier detectors are selected through an enhanced clustering-based dynamic selection (CBDS) method. In this method, a bisecting K-means clustering algorithm is employed to partition the input data into clusters where every cluster defines a local region of competence. Among the initial pool of detectors, the outputs of the detectors with the most competent local performance were combined through four possible schemes to produce the final OD results. Experimental evaluation and comparison of our method were carried out against four variants of locally selective combination in parallel (LSCP) outlier ensembles. The CBDS-based schemes compare well with the LSCP-based ones on 16 public benchmark datasets and incur considerably lower computational costs. The CBDS method consistently achieved superior average scores of the area under the curve (AUC) of the receiver operating characteristic (ROC), and particularly outperformed the LSCP method on nine of the 16 datasets in terms of the AUC score. In addition, while the CBDS and LSCP methods have similar computational costs on small datasets, the CBDS method achieves significant time savings compared with the LSCP method on large datasets.

show abstract

“…Various preprocessing methods were carried out to prepare the raw data for the ML models. To determine and exclude outliers stemming from the video recording process and/or gait parameter extraction from the videos, the IQR algorithm was used [29]. Accordingly, first lower…”

Section: Plos Onementioning

confidence: 99%

Exploration and demonstration of explainable machine learning models in prosthetic rehabilitation-based gait analysis

Aghababa,

Andrysek

2024

PLoS ONE

View full text Add to dashboard Cite

Quantitative gait analysis is important for understanding the non-typical walking patterns associated with mobility impairments. Conventional linear statistical methods and machine learning (ML) models are commonly used to assess gait performance and related changes in the gait parameters. Nonetheless, explainable machine learning provides an alternative technique for distinguishing the significant and influential gait changes stemming from a given intervention. The goal of this work was to demonstrate the use of explainable ML models in gait analysis for prosthetic rehabilitation in both population- and sample-based interpretability analyses. Models were developed to classify amputee gait with two types of prosthetic knee joints. Sagittal plane gait patterns of 21 individuals with unilateral transfemoral amputations were video-recorded and 19 spatiotemporal and kinematic gait parameters were extracted and included in the models. Four ML models—logistic regression, support vector machine, random forest, and LightGBM—were assessed and tested for accuracy and precision. The Shapley Additive exPlanations (SHAP) framework was applied to examine global and local interpretability. Random Forest yielded the highest classification accuracy (98.3%). The SHAP framework quantified the level of influence of each gait parameter in the models where knee flexion-related parameters were found the most influential factors in yielding the outcomes of the models. The sample-based explainable ML provided additional insights over the population-based analyses, including an understanding of the effect of the knee type on the walking style of a specific sample, and whether or not it agreed with global interpretations. It was concluded that explainable ML models can be powerful tools for the assessment of gait-related clinical interventions, revealing important parameters that may be overlooked using conventional statistical methods.

show abstract

Outlier Detection: Techniques and Applications

Cited by 32 publications

References 0 publications

Automatic Plane Pose Estimation for Cardiac Left Ventricle Coverage Estimation via Deep Adversarial Regression Network

Automatic Plane Pose Estimation for Cardiac Left Ventricle Coverage Estimation via Deep Adversarial Regression Network

Dynamic Construction of Outlier Detector Ensembles With Bisecting K-Means Clustering

Exploration and demonstration of explainable machine learning models in prosthetic rehabilitation-based gait analysis

Contact Info

Product

Resources

About