Assessment of Cluster Tendency Methods for Visualizing the Data Partitions

2021 ◽  
pp. 33-40
Author(s):  
M. Suleman Basha ◽  
S. K. Mouleeswaran ◽  
K. Rajendra Prasad
Keyword(s):  
Data Partitions

scholarly journals Systematic review of research design and reporting of imaging studies applying convolutional neural networks for radiological cancer diagnosis

2021 ◽  
Author(s):  
Robert J. O’Shea ◽  
Amy Rose Sharkey ◽  
Gary J. R. Cook ◽  
Vicky Goh
Keyword(s):  
Neural Network ◽  
Systematic Review ◽  
Convolutional Neural Network ◽  
Cancer Diagnosis ◽  
Model Performance ◽  
Network Models ◽  
Imaging Studies ◽  
Neural Network Models ◽  
Eligibility Criteria ◽  
Data Partitions

Abstract Objectives To perform a systematic review of design and reporting of imaging studies applying convolutional neural network models for radiological cancer diagnosis. Methods A comprehensive search of PUBMED, EMBASE, MEDLINE and SCOPUS was performed for published studies applying convolutional neural network models to radiological cancer diagnosis from January 1, 2016, to August 1, 2020. Two independent reviewers measured compliance with the Checklist for Artificial Intelligence in Medical Imaging (CLAIM). Compliance was defined as the proportion of applicable CLAIM items satisfied. Results One hundred eighty-six of 655 screened studies were included. Many studies did not meet the criteria for current design and reporting guidelines. Twenty-seven percent of studies documented eligibility criteria for their data (50/186, 95% CI 21–34%), 31% reported demographics for their study population (58/186, 95% CI 25–39%) and 49% of studies assessed model performance on test data partitions (91/186, 95% CI 42–57%). Median CLAIM compliance was 0.40 (IQR 0.33–0.49). Compliance correlated positively with publication year (ρ = 0.15, p = .04) and journal H-index (ρ = 0.27, p < .001). Clinical journals demonstrated higher mean compliance than technical journals (0.44 vs. 0.37, p < .001). Conclusions Our findings highlight opportunities for improved design and reporting of convolutional neural network research for radiological cancer diagnosis. Key Points • Imaging studies applying convolutional neural networks (CNNs) for cancer diagnosis frequently omit key clinical information including eligibility criteria and population demographics. • Fewer than half of imaging studies assessed model performance on explicitly unobserved test data partitions. • Design and reporting standards have improved in CNN research for radiological cancer diagnosis, though many opportunities remain for further progress.

scholarly journals Data Partitions and Complex Models in Bayesian Analysis: The Phylogeny of Gymnophthalmid Lizards

2004 ◽  
Vol 53 (3) ◽  
pp. 448-469 ◽  
Author(s):  
Todd A. Castoe ◽  
Tiffany M. Doan ◽  
Christopher L. Parkinson
Keyword(s):  
Bayesian Analysis ◽  
Complex Models ◽  
Data Partitions

FEATURE SELECTION BASED ON MINIMUM ERROR MINIMAX PROBABILITY MACHINE

2007 ◽  
Vol 21 (08) ◽  
pp. 1279-1292 ◽  
Author(s):  
ZENGLIN XU ◽  
IRWIN KING ◽  
MICHAEL R. LYU
Keyword(s):  
Feature Selection ◽  
Prediction Accuracy ◽  
Support Vector ◽  
Feature Subset ◽  
Minimum Error ◽  
Automatic Balance ◽  
Classification Framework ◽  
Data Partitions ◽  
Minimax Probability Machine ◽  
Optimal Feature Subset

Feature selection is an important task in pattern recognition. Support Vector Machine (SVM) and Minimax Probability Machine (MPM) have been successfully used as the classification framework for feature selection. However, these paradigms cannot automatically control the balance between prediction accuracy and the number of selected features. In addition, the selected feature subsets are also not stable in different data partitions. Minimum Error Minimax Probability Machine (MEMPM) has been proposed for classification recently. In this paper, we outline MEMPM to select the optimal feature subset with good stability and automatic balance between prediction accuracy and the size of feature subset. The experiments against feature selection with SVM and MPM show the advantages of the proposed MEMPM formulation in stability and automatic balance between the feature subset size and the prediction accuracy.

scholarly journals Plastome phylogenomics of Poaceae: alternate topologies depend on alignment gaps

2019 ◽  
Vol 192 (1) ◽  
pp. 9-20 ◽  
Author(s):  
Melvin R Duvall ◽  
Sean V Burke ◽  
Dylan C Clark
Keyword(s):  
Transition Zone ◽  
Sequence Alignment ◽  
Low Complexity ◽  
Input Matrix ◽  
Plastid Genomes ◽  
Successive Addition ◽  
Evolutionary Radiation ◽  
Data Partitions

Abstract In Poaceae there is an evolutionary radiation of c. 5000 species called the ‘PACMAD’ grasses. Two hypotheses explain deep PACMAD relationships: the ‘aristidoid sister’ and the ‘panicoid sister’ hypotheses. In each case, the named subfamily is sister to all other taxa. These hypotheses were investigated with data partitions from plastid genomes (plastomes) of 169 grasses including five newly sequenced aristidoids. Plastomes were analysed 40 times with successive addition of more gapped positions introduced by sequence alignment, until all such positions were included. Alignment gaps include low complexity, AT-rich regions. Without gaps, the panicoid sister hypothesis (P(ACMAD)) was moderately supported, but as gaps were gradually added into the input matrix, the topology and support values fluctuated through a transition zone with stripping thresholds from 2–11% until a weakly supported aristidoid sister topology was retrieved. Support values for the aristidoid sister topology then rose and plateaued for remaining analyses until all gaps were allowed. The fact that the aristidoid sister hypothesis was retrieved largely when gapped positions were included suggests that this result might be artefactual. Knowledge of the deep PACMAD topology explicitly impacts our understanding of the radiation of PACMAD grasses into open habitats.

Sign in / Sign up

Export Citation Format

Share Document