Towards algorithmic analytics for large-scale datasets

Bzdok, Danilo; Nichols, Thomas E.; Smith, Stephen M.

doi:10.1038/s42256-019-0069-5

Cited by 68 publications

(55 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yet, their experimental analysis setup may not be able to fully dismiss the critique of insufficient hyperparameter optimization in deep models. In this way, our work provides a critical addition to the existing literature, by lending some support to the idea that kernel models for exploiting nonlinearity might not even be expected to outperform simpler linear models 3,30 . As such, hurdles in exploiting even more sophisticated hierarchical nonlinearity may have been anticipated before the deep learning era in brain imaging.…”

Section: Densitymentioning

confidence: 65%

Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets

et al. 2020

Self Cite

View full text Add to dashboard Cite

Recently, deep learning has unlocked unprecedented success in various domains, especially using images, text, and speech. However, deep learning is only beneficial if the data have nonlinear relationships and if they are exploitable at available sample sizes. We systematically profiled the performance of deep, kernel, and linear models as a function of sample size on UKBiobank brain images against established machine learning references. On MNIST and Zalando Fashion, prediction accuracy consistently improves when escalating from linear models to shallow-nonlinear models, and further improves with deep-nonlinear models. In contrast, using structural or functional brain scans, simple linear models perform on par with more complex, highly parameterized models in age/sex prediction across increasing sample sizes. In sum, linear models keep improving as the sample size approaches~10,000 subjects. Yet, nonlinearities for predicting common phenotypes from typical brain scans remain largely inaccessible to the examined kernel and deep learning methods.

show abstract

Section: Densitymentioning

confidence: 65%

Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, in the present study, we were motivated to recognize potential patterns in variation of brain volume related to measures of inner and outer social groups across 36 brain regions at the population level. In this goal to complement and inform existing studies with a population neuroscience perspective, we employed a fully probabilistic hierarchical model on the UK Biobank dataset ( Bzdok et al., 2019 , 2020 ), which provides uniformly acquired brain scans and information on social behavior of 10 000 participants. This approach enabled us to probe brain–behavior association in an exploratory fashion, thus avoiding strict a priori assumptions about which putative network in the social brain may be most relevant.…”

Section: Introductionmentioning

confidence: 99%

Population variability in social brain morphology for social support, household size and friendship satisfaction

Taebi

Kiesow

Vogeley

et al. 2020

Social Cognitive and Affective Neuroscience

Self Cite

View full text Add to dashboard Cite

The social brain hypothesis proposes that the complexity of human brains has coevolved with increasing complexity of social interactions in primate societies. The present study explored the possible relationships between brain morphology and the richness of more intimate ‘inner’ and wider ‘outer’ social circles by integrating Bayesian hierarchical modeling with a large cohort sample from the UK Biobank resource (n = 10 000). In this way, we examined population volume effects in 36 regions of the ‘social brain’, ranging from lower sensory to higher associative cortices. We observed strong volume effects in the visual sensory network for the group of individuals with satisfying friendships. Further, the limbic network displayed several brain regions with substantial volume variations in individuals with a lack of social support. Our population neuroscience approach thus showed that distinct networks of the social brain show different patterns of volume variations linked to the examined social indices.

show abstract

“…That is, the most obvious sources of variation useful for groupings might reflect sex or age instead of similar disease phenotypes. Thus, biomedical data in particular often needs to be transformed from the original feature space into a more informative embedding space 6,7 . In this paper, we propose a novel space that we believe to be particularly useful for identifying latent subtypes: the space of explanations corresponding to a diagnostic classifier.…”

mentioning

confidence: 99%

Inferring disease subtypes from clusters in explanation space

Schulz

Chapman-Rounds

Verma

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

Identification of disease subtypes and corresponding biomarkers can substantially improve clinical diagnosis and treatment selection. Discovering these subtypes in noisy, high dimensional biomedical data is often impossible for humans and challenging for machines. We introduce a new approach to facilitate the discovery of disease subtypes: Instead of analyzing the original data, we train a diagnostic classifier (healthy vs. diseased) and extract instance-wise explanations for the classifier's decisions. the distribution of instances in the explanation space of our diagnostic classifier amplifies the different reasons for belonging to the same class-resulting in a representation that is uniquely useful for discovering latent subtypes. We compare our ability to recover subtypes via cluster analysis on model explanations to classical cluster analysis on the original data. in multiple datasets with known ground-truth subclasses, particularly on UK Biobank brain imaging data and transcriptome data from the Cancer Genome Atlas, we show that cluster analysis on model explanations substantially outperforms the classical approach. While we believe clustering in explanation space to be particularly valuable for inferring disease subtypes, the method is more general and applicable to any kind of sub-type identification.

show abstract

Towards algorithmic analytics for large-scale datasets

Cited by 68 publications

References 83 publications

Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets

Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets

Population variability in social brain morphology for social support, household size and friendship satisfaction

Inferring disease subtypes from clusters in explanation space

Contact Info

Product

Resources

About