Explaining deep learning-based representations of resting state functional connectivity data: focusing on interpreting nonlinear patterns in autism spectrum disorder.

Background: Resting state Functional Magnetic Resonance Imaging fMRI (rs-fMRI) has been used extensively to study brain function in psychiatric disorders, yielding insights into brain organization. However, the high dimensionality of the rs-fMRI data presents significant challenges for data analysis. Variational autoencoders (VAEs), a type of neural network, have been instrumental in extracting low-dimensional latent representations of resting state functional connectivity (rsFC) patterns, thereby addressing the complex nonlinear structure of rs-fMRI data. Despite these advances, interpreting these latent representations remains a challenge. This paper aims to address this gap by developing explainable VAE models and testing their utility using rs-fMRI data in autism spectrum disorder (ASD). Methods: One-thousand one hundred and fifty participants (601 healthy controls [HC] and 549 patients with ASD) were included in the analysis. RsFC correlation matrices were extracted from the preprocessed rs-fMRI data using the Power atlas, which includes 264 regions of interest (ROIs). Then VAEs were trained in an unsupervised manner. Lastly, we introduce our latent contribution scores to explain the relationship between estimated representations and the original rs-fMRI brain measures. Results: We quantified the latent contribution scores for both the ASD and HC groups at the network level. We found that both ASD and HC groups share the top network connectivitives contributing to all estimated latent components. For example, latent 0 was driven by rsFC within ventral attention network (VAN) in both the ASD and HC. However, we found significant differences in the latent contribution scores between the ASD and HC groups within the VAN for latent 0 and the sensory/somatomotor network for latent 2. Conclusion: This study introduced latent contribution scores to interpret nonlinear patterns identified by VAEs. These scores effectively capture changes in each observed rsFC feature as the estimated latent representation changes, enabling an explainable deep learning model that better understands the underlying neural mechanisms of ASD.

Optimizing Contingency Management with Reinforcement Learning.

Contingency Management (CM) is a psychological treatment that aims to change behavior with financial incentives. In substance use disorders (SUDs), deployment of CM has been enriched by longstanding discussions around the cost-effectiveness of prized-based and voucher-based approaches. In prize-based CM, participants earn draws to win prizes, including small incentives to reduce costs, and the number of draws escalates depending on the duration of maintenance of abstinence. In voucher-based CM, participants receive a predetermined voucher amount based on specific substance test results. While both types have enhanced treatment outcomes, there is room for improvement in their cost-effectiveness: the voucher-based system requires enduring financial investment; the prize-based system might sacrifice efficacy. Previous work in computational psychiatry of SUDs typically employs frameworks wherein participants make decisions to maximize their expected compensation. In contrast, we developed new frameworks that clinical decision-makers choose actions, CM structures, to reinforce the substance abstinence behavior of participants. We consider the choice of the voucher or prize to be a sequential decision, where there are two pivotal parameters: the prize probability for each draw and the escalation rule determining the number of draws. Recent advancements in Reinforcement Learning, more specifically, in off-policy evaluation, afforded techniques to estimate outcomes for different CM decision scenarios from observed clinical trial data. We searched CM schemas that maximized treatment outcomes with budget constraints. Using this framework, we analyzed data from the Clinical Trials Network to construct unbiased estimators on the effects of new CM schemas. Our results indicated that the optimal CM schema would be to strengthen reinforcement rapidly in the middle of the treatment course. Our estimated optimal CM policy improved treatment outcomes by 32% while maintaining costs. Our methods and results have broad applications in future clinical trial planning and translational investigations on the neurobiological basis of SUDs.

Temporal Generative Models for Learning Heterogeneous Group Dynamics of Ecological Momentary Data.

One of the goals of precision psychiatry is to characterize mental disorders in an individualized manner, taking into account the underlying dynamic processes. Recent advances in mobile technologies have enabled the collection of Ecological Momentary Assessments (EMAs) that capture multiple responses in real-time at high frequency. However, EMA data is often multi-dimensional, correlated, and hierarchical. Mixed-effects models are commonly used but may require restrictive assumptions about the fixed and random effects and the correlation structure. The Recurrent Temporal Restricted Boltzmann Machine (RTRBM) is a generative neural network that can be used to model temporal data, but most existing RTRBM approaches do not account for the potential heterogeneity of group dynamics within a population based on available covariates. In this paper, we propose a new temporal generative model, the Heterogeneous-Dynamics Restricted Boltzmann Machine (HDRBM), to learn the heterogeneous group dynamics and demonstrate the effectiveness of this approach on simulated and real-world EMA data sets. We show that by incorporating covariates, HDRBM can improve accuracy and interpretability, explore the underlying drivers of the group dynamics of participants, and serve as a generative model for EMA studies.

Explaining Deep Learning-Based Representations of Resting State Functional Connectivity Data: Focusing on Interpreting Nonlinear Patterns in Autism Spectrum Disorder.

Background: Resting state Functional Magnetic Resonance Imaging fMRI (rs-fMRI) has been used to study brain function in psychiatric disorders, yielding insight into brain organization. However, the high dimensionality of the rs-fMRI data presents challenges, and requires dimensionality reduction before applying machine learning techniques. Neural networks, specifically variational autoencoders (VAEs), have been instrumental in extracting low-dimensional latent representations of resting state functional connectivity patterns, addressing the complex nonlinear structure of rs-fMRI. However, interpreting those latent representations remains a challenge. This paper aims to address this gap by creating explainable VAE models and testing their utility using rs-fMRI data in autism spectrum disorder (ASD). Methods: One-thousand one hundred and fifty participants (601 HC and 549 patients with ASD) were included in the analysis. We extracted functional connectivity correlation matrices from the preprocessed rs-fMRI data using Power atlas with 264 ROIs. Then VAEs were trained in an unsupervised fashion. Lastly, we introduce our latent contribution scores to explain the relationship between estimated representations and the original rs-fMRI brain measures. Results: We quantified the latent contribution scores for the ASD and control groups at the network level. We found that both ASD and control groups share the top network connectivity that contribute to all estimated latent components. For example, latent 0 was driven by resting state functional connectivity patterns (rsFC) within ventral attention network in both the ASD and control. However, significant differences in the latent contribution scores between the ASD and control groups were discovered within the ventral attention network in latent 0 and the sensory/somatomotor network in latent 2. Conclusion: This study introduced latent contribution scores to interpret nonlinear patterns identified by VAEs. These scores effectively capture changes in each observed rsFC features as estimated latent representation changes, enabling an explainable deep learning model to better understand the underlying neural mechanism of ASD.

Covariate-informed Representation Learning to Prevent Posterior Collapse of iVAE.

The recently proposed identifiable variational autoencoder (iVAE) framework provides a promising approach for learning latent independent components (ICs). iVAEs use auxiliary covariates to build an identifiable generation structure from covariates to ICs to observations, and the posterior network approximates ICs given observations and covariates. Though the identifiability is appealing, we show that iVAEs could have local minimum solution where observations and the approximated ICs are independent given covariates.-a phenomenon we referred to as the posterior collapse problem of iVAEs. To overcome this problem, we develop a new approach, covariate-informed iVAE (CI-iVAE) by considering a mixture of encoder and posterior distributions in the objective function. In doing so, the objective function prevents the posterior collapse, resulting latent representations that contain more information of the observations. Furthermore, CI-iVAE extends the original iVAE objective function to a larger class and finds the optimal one among them, thus having tighter evidence lower bounds than the original iVAE. Experiments on simulation datasets, EMNIST, Fashion-MNIST, and a large-scale brain imaging dataset demonstrate the effectiveness of our new method.

Axion Dark Matter Search around 4.55 µeV with Dine-Fischler-Srednicki-Zhitnitskii Sensitivity.

We report an axion dark matter search at Dine-Fischler-Srednicki-Zhitnitskii sensitivity with the CAPP-12TB haloscope, assuming axions contribute 100% of the local dark matter density. The search excluded the axion-photon coupling g_{aγγ} down to about 6.2×10^{-16} GeV^{-1} over the axion mass range between 4.51 and 4.59 µeV at a 90% confidence level. The achieved experimental sensitivity can also exclude Kim-Shifman-Vainshtein-Zakharov axion dark matter that makes up just 13% of the local dark matter density. The CAPP-12TB haloscope will continue the search over a wide range of axion masses.

Conditional Wasserstein Generator.

The statistical distance of conditional distributions is an essential element of generating target data given some data as in video prediction. We establish how the statistical distances between two joint distributions are related to those between two conditional distributions for three popular statistical distances: f-divergence, Wasserstein distance, and integral probability metrics. Such characterization plays a crucial role in deriving a tractable form of the objective function to learn a conditional generator. For Wasserstein distance, we show that the distance between joint distributions is an upper bound of the expected distance between conditional distributions, and derive a tractable representation of the upper bound. Based on this theoretical result, we propose a new conditional generator, the conditional Wasserstein generator. Our proposed algorithm can be viewed as an extension of Wasserstein autoencoders (Tolstikhin et al. 2018) to conditional generation or as a Wasserstein counterpart of stochastic video generation (SVG) model by Denton and Fergus (Denton et al. 2018). We apply our algorithm to video prediction and video interpolation. Our experiments demonstrate that the proposed algorithm performs well on benchmark video datasets and produces sharper videos than state-of-the-art methods.

The Mediating Role of Depression Severity on the Relationship Between Suicidal Ideation and Self-Injury in Adolescents With Major Depressive Disorder.

Objectives: Suicide is the leading cause of death among adolescents in South Korea, and depression and personality profiles have been identified as significant risk factors for self-injurious behavior. This study examined the influence of depressive mood and temperament/ character on self-injury in adolescents. Methods: A total of 116 adolescents (aged 12-18 years) with a primary diagnosis of major depressive disorder (MDD) and their parents were enrolled in this study. The participants were divided into three groups based on adolescent's self-injury frequency, and their Children's Depression Inventory (CDI), Youth Self-Report (YSR), and Temperament and Character Inventory (TCI) scores were compared. Finally, mediation analysis was conducted to investigate the relationship between suicidal ideation and self-injury. Results: Of study participants, 75.9% answered that they had suicidal ideation, and 55.2% answered that they had engaged in self-injurious behavior in the last six months. There were significant differences in CDI and suicidal ideation among the groups. After adjusting for age and sex, mediation analysis indicated that depressive mood mediated the relationship between suicidal ideation and self-injury. Conclusion: This study emphasizes the importance of evaluating and managing depressive mood severity in adolescents with MDD as these factors partially mediate the transition from suicidal ideation to self-injury.

Bed-mounted laparoscopic surgical robot system with novel positioning arm mechanism.

BACKGROUND: Commercialised laparoscopic surgical robotic systems require a large operating room and can only be used in large hospitals. If the robotic system is to be used in a small- or medium-sized hospital, the occupied volume must be reduced further. METHODS: In this paper, we propose a bed-mounted system that can be installed in a general operating room. Furthermore, we proposed a novel positioning arm suitable for a bed-mounted surgical robot system. RESULTS: The surgical possibility of the proposed bed-mounted system has been verified. Furthermore, the surgical possibility of the proposed system was confirmed using in vivo animal experiments. CONCLUSIONS: A bed-mounted laparoscopic robotic system and a novel positioning arm was proposed. The study's ultimate goal is to enable robotic surgery in small and medium-sized hospitals by introducing the proposed bed-mounted laparoscopic robot system, allowing many people to receive high-quality medical services.

Characterization of Pulse Reverses Electroforming on Hard Gold Coating.

Effect of pulse reverse current (PRC) method on brass coatings electroplated from gold solution was investigated by various plating parameters such as plating duration, the anodic duty cycle, the anodic current density and the cathodic current density. The reversed current results in a significant change in the morphology of electrodeposits, improvement of the overall current efficiency and reduction of deposit porosity. With longer pulses, hemispherical surface features are generated, while larger grains result from shorter pulse widths. The porosity of the plated samples is found to decrease compared with results at the same time-average plating rate obtained from DC or Pulse plating. A major impediment to reducing gold later thickness is the corrosion of the underlying substrate, which is affected by the porosity of the gold layer. Both the morphology and the hydrogen evolution reaction have significant impact on porosity. PRC plating affect hydrogen gold and may oxidize hydrogen produced during the cathodic portion of the waveform. Whether the dissolution of gold and oxidation of hydrogen occur depends on the type of plating bath and the plating conditions adapted. In reversed pulse plating, the amount of excess near-surface cyanide is changed after the cathodic current is applied, and the oxidation of gold under these conditions has not been fully addressed. The effects of the current density, pulse-reverse ratio and brightener concentration of the electroplating process were investigated and optimized for suitable performance.

Fabrication of polyimide nanotubes and carbon nanotubes containing magnetic iron oxide in confinement.

Polyimide nanotubes with tunable wall thickness were fabricated by a precursor impregnation method using an AAO template, and carbon nanotubes containing magnetic iron oxide were obtained using ferric chloride-embedded polyimide precursor by a carbonization process.

Electron-Transfer-Induced Molecular Reorientations: The Benzoquinone/Hydroquinone Reaction at Pd(111)-(square3xsquare3)R30 degrees -I Studied by EC-STM.

The benzoquinone/hydroquinone (Q/H(2)Q) redox reaction has been studied by electrochemical-scanning tunneling microscopy (EC-STM) at a Pd(111)-(square3xsquare3)R30 degrees -I electrode surface in a solution that contained 10(-4) M H(2)Q in 0.05 M H(2)SO(4); iodine-pretreatment of the Pd(111) surface was to prevent chemisorption of organic-derived species. The molecule-resolved EC-STM images indicated that: (i) at a potential where only H(2)Q is present in solution, a self-assembled (square21xsquare21)R10.9 degrees -eta(6)-H(2)Q monolayer is produced in which the H(2)Q molecules are oriented parallel to the surface; (ii) at a potential where partial oxidation (to Q) occurs, a self-assembled (square21xsquare21)R10.9 degrees -eta(6)-QH adlayer is generated, where QH represents quinhydrone, an equimolar mixture of Q and H(2)Q; in this structure, the Q and H(2)Q molecules are oriented vertically, face-to-face, and arranged alternately along a given row, reminiscent of the crystal structure of quinhydrone. The partial oxidation-induced molecular reorientation, which is reversible, most likely arises from favorable Q-H(2)Q face-to-face interactions; that is, complete oxidation would yield only flat-oriented Q species. Unfortunately, at potentials where only Q would be present in solution, I-catalyzed corrosion of the Pd starts to occur, which leads to noise-laden EC-STM images. Copyright 2001 Academic Press.