Biology subjects

calhoun, v.

Publications and source records attributed to calhoun, v..

2 recordsLinked to original sources

bioRxiv2019

Connectivity domain analysis: characterization of connectivity frequency profile using filter banks

Studying dynamic functional connectivity (dFC) has been the focus of many studies in recent years. The most commonly used estimator for dFC uses a sliding window in combination with a connectivity estimator such as Pearson correlation. Here, we propose a new approach to estimate connectivity while preserving its full frequency range and subsequently examine both static and dynamic connectivity in one unified approach. This approach which we call filter banked connectivity (FBC), implements frequency tiling directly in the connectivity domain contrary to other studies where frequency tiling is done in the activity domain. This leads to more accurate modeling, and a unified approach to capture connectivity ranging from static to highly dynamic, avoiding the need to pick a specific band as in a sliding window approach. First, we demonstrated that our proposed approach, can estimate connectivity at frequencies that sliding window approach fails. Next we evaluated the ability of the approach to identify group differences by using the FBC approach to estimate dFNC in a resting fMRI data set including schizophrenia patients (SZ, n=151) and typical controls (TC, n=163). To summarize the results, we used k-means to cluster the FBC values into different clusters. Some states showed very weak low frequency strength and as such SWPC was not well suited to capture them. Additionally, we found that SZs tend to spend more time in states exhibiting higher frequencies and engaging the default mode network and its anticorrelations with other networks compared to TCs which spent more time in lower frequency states which primarily includes strong intercorrelations within the sensorimotor domains. In summary, the proposed approach offers a novel way to estimate connectivity while unifying static and dynamic connectivity analyses and can provide additional otherwise missed information about the frequency profile of connectivity patterns.

neuroscience↗

bioRxiv2019

Decentralized Distribution-sampled Classification Models with Application to Brain Imaging

0.1 backgroundIn this age of big data, large data stores allow researchers to compose robust models that are accurate and informative. In many cases, the data are stored in separate locations requiring data transfer between local sites, which can cause various practical hurdles, such as privacy concerns or heavy network load. This is especially true for medical imaging data, which can be constrained due to the health insurance portability and accountability act (HIPAA). Medical imaging datasets can also contain many thousands or millions of features, requiring heavy network load.\n\n0.2 New MethodOur research expands upon current decentralized classification research by implementing a new singleshot method for both neural networks and support vector machines. Our approach is to estimate the statistical distribution of the data at each local site and pass this information to the other local sites where each site resamples from the individual distributions and trains a model on both locally available data and the resampled data.\n\n0.3 ResultsWe show applications of our approach to handwritten digit classification as well as to multi-subject classification of brain imaging data collected from patients with schizophrenia and healthy controls. Overall, the results showed comparable classification accuracy to the centralized model with lower network load than multishot methods.\n\n0.4 Comparison with Existing MethodsMany decentralized classifiers are multishot, requiring heavy network traffic. Our model attempts to alleviate this load while preserving prediction accuracy.\n\n0.5 ConclusionsWe show that our proposed approach performs comparably to a centralized approach while minimizing network traffic compared to multishot methods.\n\n0.6 HighlightsO_LIA novel yet simple approach to decentralized classification\nC_LIO_LIReduces total network load compared to current multishot algorithms\nC_LIO_LIMaintains a prediction accuracy comparable to the centralized approach\nC_LI

bioinformatics↗