Deep Learning for Quality Control of Subcortical Brain 3D Shape Models
Authors:
Dmitry Petrov,
Boris A. Gutman Egor Kuznetsov,
Theo G. M. van Erp,
Jessica A. Turner,
Lianne Schmaal,
Dick Veltman,
Lei Wang,
Kathryn Alpert,
Dmitry Isaev,
Artemis Zavaliangos-Petropulu,
Christopher R. K. Ching,
Vince Calhoun,
David Glahn,
Theodore D. Satterthwaite,
Ole Andreas Andreassen,
Stefan Borgwardt,
Fleur Howells,
Nynke Groenewold,
Aristotle Voineskos,
Joaquim Radua,
Steven G. Potkin,
Benedicto Crespo-Facorro,
Diana Tordesillas-Gutierrez,
Li Shen,
Irina Lebedeva
, et al. (48 additional authors not shown)
Abstract:
We present several deep learning models for assessing the morphometric fidelity of deep grey matter region models extracted from brain MRI. We test three different convolutional neural net architectures (VGGNet, ResNet and Inception) over 2D maps of geometric features. Further, we present a novel geometry feature augmentation technique based on a parametric spherical mapping. Finally, we present a…
▽ More
We present several deep learning models for assessing the morphometric fidelity of deep grey matter region models extracted from brain MRI. We test three different convolutional neural net architectures (VGGNet, ResNet and Inception) over 2D maps of geometric features. Further, we present a novel geometry feature augmentation technique based on a parametric spherical mapping. Finally, we present an approach for model decision visualization, allowing human raters to see the areas of subcortical shapes most likely to be deemed of failing quality by the machine. Our training data is comprised of 5200 subjects from the ENIGMA Schizophrenia MRI cohorts, and our test dataset contains 1500 subjects from the ENIGMA Major Depressive Disorder cohorts. Our final models reduce human rater time by 46-70%. ResNet outperforms VGGNet and Inception for all of our predictive tasks.
△ Less
Submitted 4 September, 2018; v1 submitted 30 August, 2018;
originally announced August 2018.
Functional effects of schizophrenia-linked genetic variants on intrinsic single-neuron excitability: A modeling study
Authors:
Tuomo Mäki-Marttunen,
Geir Halnes,
Anna Devor,
Aree Witoelar,
Francesco Bettella,
Srdjan Djurovic,
Yunpeng Wang,
Gaute T. Einevoll,
Ole A. Andreassen,
Anders M. Dale
Abstract:
Background: Recent genome-wide association studies (GWAS) have identified a large number of genetic risk factors for schizophrenia (SCZ) featuring ion channels and calcium transporters. For some of these risk factors, independent prior investigations have examined the effects of genetic alterations on the cellular electrical excitability and calcium homeostasis. In the present proof-of-concept stu…
▽ More
Background: Recent genome-wide association studies (GWAS) have identified a large number of genetic risk factors for schizophrenia (SCZ) featuring ion channels and calcium transporters. For some of these risk factors, independent prior investigations have examined the effects of genetic alterations on the cellular electrical excitability and calcium homeostasis. In the present proof-of-concept study, we harnessed these experimental results for modeling of computational properties on layer V cortical pyramidal cell and identify possible common alterations in behavior across SCZ-related genes.
Methods: We applied a biophysically detailed multi-compartmental model to study the excitability of a layer V pyramidal cell. We reviewed the literature on functional genomics for variants of genes associated with SCZ, and used changes in neuron model parameters to represent the effects of these variants.
Results: We present and apply a framework for examining the effects of subtle single nucleotide polymorphisms in ion channel and Ca2+ transporter-encoding genes on neuron excitability. Our analysis indicates that most of the considered SCZ- related genetic variants affect the spiking behavior and intracellular calcium dynamics resulting from summation of inputs across the dendritic tree.
Conclusions: Our results suggest that alteration in the ability of a single neuron to integrate the inputs and scale its excitability may constitute a fundamental mechanistic contributor to mental disease, alongside with the previously proposed deficits in synaptic communication and network behavior.
△ Less
Submitted 24 September, 2015;
originally announced September 2015.
