Showing 1–28 of 28 results for author: D'Souza, N

Multimodal Sleep Apnea Detection with Missing or Noisy Modalities

Authors: Hamed Fayyaz, Abigail Strang, Niharika S. D'Souza, Rahmatollah Beheshti

Abstract: Polysomnography (PSG) is a type of sleep study that records multimodal physiological signals and is widely used for purposes such as sleep staging and respiratory event detection. Conventional machine learning methods assume that each sleep study is associated with a fixed set of observed modalities and that all modalities are available for each sample. However, noisy and missing modalities are a… ▽ More Polysomnography (PSG) is a type of sleep study that records multimodal physiological signals and is widely used for purposes such as sleep staging and respiratory event detection. Conventional machine learning methods assume that each sleep study is associated with a fixed set of observed modalities and that all modalities are available for each sample. However, noisy and missing modalities are a common issue in real-world clinical settings. In this study, we propose a comprehensive pipeline aiming to compensate for the missing or noisy modalities when performing sleep apnea detection. Unlike other existing studies, our proposed model works with any combination of available modalities. Our experiments show that the proposed model outperforms other state-of-the-art approaches in sleep apnea detection using various subsets of available data and different levels of noise, and maintains its high performance (AUROC>0.9) even in the presence of high levels of noise or missingness. This is especially relevant in settings where the level of noise and missingness is high (such as pediatric or outside-of-clinic scenarios). △ Less

Submitted 24 February, 2024; originally announced February 2024.

Room-temperature quantum sensing with photoexcited triplet electrons in organic crystals

Authors: Harpreet Singh, Noella D'Souza, Keyuan Zhong, Emanuel Druga, Julianne Oshiro, Brian Blankenship, Jeffrey A. Reimer, Jonathan D. Breeze, Ashok Ajoy

Abstract: Quantum sensors have notably advanced high-sensitivity magnetic field detection. Here, we report quantum sensors constructed from polarized spin-triplet electrons in photoexcited organic chromophores, specifically focusing on pentacene-doped para-terphenyl (${\approx}$0.1%). We demonstrate essential quantum sensing properties at room temperature: electronic optical polarization and state-dependent… ▽ More Quantum sensors have notably advanced high-sensitivity magnetic field detection. Here, we report quantum sensors constructed from polarized spin-triplet electrons in photoexcited organic chromophores, specifically focusing on pentacene-doped para-terphenyl (${\approx}$0.1%). We demonstrate essential quantum sensing properties at room temperature: electronic optical polarization and state-dependent fluorescence contrast, by leveraging differential pum** and relaxation rates between triplet and ground states. We measure high optically detected magnetic resonance (ODMR) contrast ${\approx}16.8\%$ of the triplet states at room temperature, along with long coherence times under spin echo and CPMG sequences, $T_2{=}2.7μ$s and $T_2^{DD}{=}18.4μ$s respectively, limited only by the triplet lifetimes. The material offers several advantages for quantum sensing, including the ability to grow large ($cm$-scale) crystals at low cost, the absence of paramagnetic impurities, and the diamagnetism of electronic states used for sensing when not optically illuminated. Utilizing pentacene as a representative of a broader class of spin triplet-polarizable organic molecules, this study highlights new potential for quantum sensing in chemical systems. △ Less

Submitted 21 February, 2024; originally announced February 2024.

Nanoscale engineering and dynamical stabilization of mesoscopic spin textures

Authors: Kieren Harkins, Christoph Fleckenstein, Noella D'Souza, Paul M. Schindler, David Marchiori, Claudia Artiaco, Quentin Reynard-Feytis, Ushoshi Basumallick, William Beatrez, Arjun Pillai, Matthias Hagn, Aniruddha Nayak, Samantha Breuer, Xudong Lv, Maxwell McAllister, Paul Reshetikhin, Emanuel Druga, Marin Bukov, Ashok Ajoy

Abstract: Thermalization phenomena, while ubiquitous in quantum systems, have traditionally been viewed as obstacles to be mitigated. In this study, we demonstrate the ability, instead, to harness thermalization to dynamically engineer and stabilize structured quantum states in a mesoscopically large ensemble of spins. Specifically, we showcase the capacity to generate, control, stabilize, and read out 'she… ▽ More Thermalization phenomena, while ubiquitous in quantum systems, have traditionally been viewed as obstacles to be mitigated. In this study, we demonstrate the ability, instead, to harness thermalization to dynamically engineer and stabilize structured quantum states in a mesoscopically large ensemble of spins. Specifically, we showcase the capacity to generate, control, stabilize, and read out 'shell-like' spin texture with interacting $ {}^{ 13}\mathrm{C}$ nuclear spins in diamond, wherein spins are polarized oppositely on either side of a critical radius. The texture spans several nanometers and encompasses many hundred spins. We capitalize on the thermalization process to impose a quasi-equilibrium upon the generated texture; as a result, it is highly stable, immune to spin diffusion, and endures over multiple-minute long periods -- over a million times longer than the intrinsic interaction scale of the spins. Additionally, the texture is created and interrogated without locally controlling or probing the nuclear spins. These features are accomplished using an electron spin as a nanoscale injector of spin polarization, and employing it as a source of spatially varying dissipation, allowing for serial readout of the emergent spin texture. Long-time stabilization is achieved via prethermalization to a Floquet-induced Hamiltonian under the electronic gradient field. Our work presents a new approach to robust nanoscale spin state engineering and paves the way for new applications in quantum simulation, quantum information science, and nanoscale imaging. △ Less

Submitted 9 October, 2023; originally announced October 2023.

Comments: 8 + 32 pages

Demonstrating a long-coherence dual-rail erasure qubit using tunable transmons

Authors: Harry Levine, Arbel Haim, Jimmy S. C. Hung, Nasser Alidoust, Mahmoud Kalaee, Laura DeLorenzo, E. Alex Wollack, Patricio Arrangoiz-Arriola, Amirhossein Khalajhedayati, Rohan Sanil, Hesam Moradinejad, Yotam Vaknin, Aleksander Kubica, David Hover, Shahriar Aghaeimeibodi, Joshua Ari Alcid, Christopher Baek, James Barnett, Kaustubh Bawdekar, Przemyslaw Bienias, Hugh Carson, Cliff Chen, Li Chen, Harut Chinkezian, Eric M. Chisholm , et al. (88 additional authors not shown)

Abstract: Quantum error correction with erasure qubits promises significant advantages over standard error correction due to favorable thresholds for erasure errors. To realize this advantage in practice requires a qubit for which nearly all errors are such erasure errors, and the ability to check for erasure errors without dephasing the qubit. We demonstrate that a "dual-rail qubit" consisting of a pair of… ▽ More Quantum error correction with erasure qubits promises significant advantages over standard error correction due to favorable thresholds for erasure errors. To realize this advantage in practice requires a qubit for which nearly all errors are such erasure errors, and the ability to check for erasure errors without dephasing the qubit. We demonstrate that a "dual-rail qubit" consisting of a pair of resonantly coupled transmons can form a highly coherent erasure qubit, where transmon $T_1$ errors are converted into erasure errors and residual dephasing is strongly suppressed, leading to millisecond-scale coherence within the qubit subspace. We show that single-qubit gates are limited primarily by erasure errors, with erasure probability $p_\text{erasure} = 2.19(2)\times 10^{-3}$ per gate while the residual errors are $\sim 40$ times lower. We further demonstrate mid-circuit detection of erasure errors while introducing $< 0.1\%$ dephasing error per check. Finally, we show that the suppression of transmon noise allows this dual-rail qubit to preserve high coherence over a broad tunable operating range, offering an improved capacity to avoid frequency collisions. This work establishes transmon-based dual-rail qubits as an attractive building block for hardware-efficient quantum error correction. △ Less

Submitted 20 March, 2024; v1 submitted 17 July, 2023; originally announced July 2023.

Comments: 9+13 pages, 16 figures

Journal ref: Physical Review X 14, 011051 (2024)

MaxCorrMGNN: A Multi-Graph Neural Network Framework for Generalized Multimodal Fusion of Medical Data for Outcome Prediction

Authors: Niharika S. D'Souza, Hongzhi Wang, Andrea Giovannini, Antonio Foncubierta-Rodriguez, Kristen L. Beck, Orest Boyko, Tanveer Syeda-Mahmood

Abstract: With the emergence of multimodal electronic health records, the evidence for an outcome may be captured across multiple modalities ranging from clinical to imaging and genomic data. Predicting outcomes effectively requires fusion frameworks capable of modeling fine-grained and multi-faceted complex interactions between modality features within and across patients. We develop an innovative fusion a… ▽ More With the emergence of multimodal electronic health records, the evidence for an outcome may be captured across multiple modalities ranging from clinical to imaging and genomic data. Predicting outcomes effectively requires fusion frameworks capable of modeling fine-grained and multi-faceted complex interactions between modality features within and across patients. We develop an innovative fusion approach called MaxCorr MGNN that models non-linear modality correlations within and across patients through Hirschfeld-Gebelein-Renyi maximal correlation (MaxCorr) embeddings, resulting in a multi-layered graph that preserves the identities of the modalities and patients. We then design, for the first time, a generalized multi-layered graph neural network (MGNN) for task-informed reasoning in multi-layered graphs, that learns the parameters defining patient-modality graph connectivity and message passing in an end-to-end fashion. We evaluate our model an outcome prediction task on a Tuberculosis (TB) dataset consistently outperforming several state-of-the-art neural, graph-based and traditional fusion techniques. △ Less

Submitted 13 July, 2023; originally announced July 2023.

Comments: To appear in ML4MHD workshop at ICML 2023

mSPD-NN: A Geometrically Aware Neural Framework for Biomarker Discovery from Functional Connectomics Manifolds

Authors: Niharika S. D'Souza, Archana Venkataraman

Abstract: Connectomics has emerged as a powerful tool in neuroimaging and has spurred recent advancements in statistical and machine learning methods for connectivity data. Despite connectomes inhabiting a matrix manifold, most analytical frameworks ignore the underlying data geometry. This is largely because simple operations, such as mean estimation, do not have easily computable closed-form solutions. We… ▽ More Connectomics has emerged as a powerful tool in neuroimaging and has spurred recent advancements in statistical and machine learning methods for connectivity data. Despite connectomes inhabiting a matrix manifold, most analytical frameworks ignore the underlying data geometry. This is largely because simple operations, such as mean estimation, do not have easily computable closed-form solutions. We propose a geometrically aware neural framework for connectomes, i.e., the mSPD-NN, designed to estimate the geodesic mean of a collections of symmetric positive definite (SPD) matrices. The mSPD-NN is comprised of bilinear fully connected layers with tied weights and utilizes a novel loss function to optimize the matrix-normal equation arising from Fréchet mean estimation. Via experiments on synthetic data, we demonstrate the efficacy of our mSPD-NN against common alternatives for SPD mean estimation, providing competitive performance in terms of scalability and robustness to noise. We illustrate the real-world flexibility of the mSPD-NN in multiple experiments on rs-fMRI data and demonstrate that it uncovers stable biomarkers associated with subtle network differences among patients with ADHD-ASD comorbidities and healthy controls. △ Less

Submitted 27 March, 2023; originally announced March 2023.

Comments: Accepted into IPMI 2023

Bayesian Models of Functional Connectomics and Behavior

Authors: Niharika Shimona D'Souza

Abstract: The problem of jointly analysing functional connectomics and behavioral data is extremely challenging owing to the complex interactions between the two domains. In addition, clinical rs-fMRI studies often have to contend with limited samples, especially in the case of rare disorders. This data-starved regimen can severely restrict the reliability of classical machine learning or deep learning desi… ▽ More The problem of jointly analysing functional connectomics and behavioral data is extremely challenging owing to the complex interactions between the two domains. In addition, clinical rs-fMRI studies often have to contend with limited samples, especially in the case of rare disorders. This data-starved regimen can severely restrict the reliability of classical machine learning or deep learning designed to predict behavior from connectivity data. In this work, we approach this problem from the lens of representation learning and bayesian modeling. To model the distributional characteristics of the domains, we first examine the ability of approaches such as Bayesian Linear Regression, Stochastic Search Variable Selection after performing a classical covariance decomposition. Finally, we present a fully bayesian formulation for joint representation learning and prediction. We present preliminary results on a subset of a publicly available clinical rs-fMRI study on patients with Autism Spectrum Disorder. △ Less

Submitted 15 January, 2023; originally announced January 2023.

Fusing Modalities by Multiplexed Graph Neural Networks for Outcome Prediction in Tuberculosis

Authors: Niharika S. D'Souza, Hongzhi Wang, Andrea Giovannini, Antonio Foncubierta-Rodriguez, Kristen L. Beck, Orest Boyko, Tanveer Syeda-Mahmood

Abstract: In a complex disease such as tuberculosis, the evidence for the disease and its evolution may be present in multiple modalities such as clinical, genomic, or imaging data. Effective patient-tailored outcome prediction and therapeutic guidance will require fusing evidence from these modalities. Such multimodal fusion is difficult since the evidence for the disease may not be uniform across all moda… ▽ More In a complex disease such as tuberculosis, the evidence for the disease and its evolution may be present in multiple modalities such as clinical, genomic, or imaging data. Effective patient-tailored outcome prediction and therapeutic guidance will require fusing evidence from these modalities. Such multimodal fusion is difficult since the evidence for the disease may not be uniform across all modalities, not all modality features may be relevant, or not all modalities may be present for all patients. All these nuances make simple methods of early, late, or intermediate fusion of features inadequate for outcome prediction. In this paper, we present a novel fusion framework using multiplexed graphs and derive a new graph neural network for learning from such graphs. Specifically, the framework allows modalities to be represented through their targeted encodings, and models their relationship explicitly via multiplexed graphs derived from salient features in a combined latent space. We present results that show that our proposed method outperforms state-of-the-art methods of fusing modalities for multi-outcome prediction on a large Tuberculosis (TB) dataset. △ Less

Submitted 25 October, 2022; originally announced October 2022.

Comments: Accepted into MICCAI 2022

A Matrix Autoencoder Framework to Align the Functional and Structural Connectivity Manifolds as Guided by Behavioral Phenotypes

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Deana Crocetti, Nicholas Wymbs, Joshua Robinson, Stewart Mostofsky, Archana Venkataraman

Abstract: We propose a novel matrix autoencoder to map functional connectomes from resting state fMRI (rs-fMRI) to structural connectomes from Diffusion Tensor Imaging (DTI), as guided by subject-level phenotypic measures. Our specialized autoencoder infers a low dimensional manifold embedding for the rs-fMRI correlation matrices that mimics a canonical outer-product decomposition. The embedding is simultan… ▽ More We propose a novel matrix autoencoder to map functional connectomes from resting state fMRI (rs-fMRI) to structural connectomes from Diffusion Tensor Imaging (DTI), as guided by subject-level phenotypic measures. Our specialized autoencoder infers a low dimensional manifold embedding for the rs-fMRI correlation matrices that mimics a canonical outer-product decomposition. The embedding is simultaneously used to reconstruct DTI tractography matrices via a second manifold alignment decoder and to predict inter-subject phenotypic variability via an artificial neural network. We validate our framework on a dataset of 275 healthy individuals from the Human Connectome Project database and on a second clinical dataset consisting of 57 subjects with Autism Spectrum Disorder. We demonstrate that the model reliably recovers structural connectivity patterns across individuals, while robustly extracting predictive and interpretable brain biomarkers in a cross-validated setting. Finally, our framework outperforms several baselines at predicting behavioral phenotypes in both real-world datasets. △ Less

Submitted 9 July, 2021; v1 submitted 29 May, 2021; originally announced May 2021.

A Multi-Task Deep Learning Framework to Localize the Eloquent Cortex in Brain Tumor Patients Using Dynamic Functional Connectivity

Authors: Naresh Nandakumar, Niharika Shimona D'souza, Komal Manzoor, Jay J. Pillai, Sachin K. Gujar, Haris I. Sair, Archana Venkataraman

Abstract: We present a novel deep learning framework that uses dynamic functional connectivity to simultaneously localize the language and motor areas of the eloquent cortex in brain tumor patients. Our method leverages convolutional layers to extract graph-based features from the dynamic connectivity matrices and a long-short term memory (LSTM) attention network to weight the relevant time points during cl… ▽ More We present a novel deep learning framework that uses dynamic functional connectivity to simultaneously localize the language and motor areas of the eloquent cortex in brain tumor patients. Our method leverages convolutional layers to extract graph-based features from the dynamic connectivity matrices and a long-short term memory (LSTM) attention network to weight the relevant time points during classification. The final stage of our model employs multi-task learning to identify different eloquent subsystems. Our unique training strategy finds a shared representation between the cognitive networks of interest, which enables us to handle missing patient data. We evaluate our method on resting-state fMRI data from 56 brain tumor patients while using task fMRI activations as surrogate ground-truth labels for training and testing. Our model achieves higher localization accuracies than conventional deep learning approaches and can identify bilateral language areas even when trained on left-hemisphere lateralized cases. Hence, our method may ultimately be useful for preoperative map** in tumor patients. △ Less

Submitted 17 November, 2020; originally announced November 2020.

Comments: Presented at MLCN 2020 workshop, as a part of MICCAI 2020

A Joint Network Optimization Framework to Predict Clinical Severity from Resting State Functional MRI Data

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Nicholas Wymbs, Stewart H. Mostofsky, Archana Venkataraman

Abstract: We propose a novel optimization framework to predict clinical severity from resting state fMRI (rs-fMRI) data. Our model consists of two coupled terms. The first term decomposes the correlation matrices into a sparse set of representative subnetworks that define a network manifold. These subnetworks are modeled as rank-one outer-products which correspond to the elemental patterns of co-activation… ▽ More We propose a novel optimization framework to predict clinical severity from resting state fMRI (rs-fMRI) data. Our model consists of two coupled terms. The first term decomposes the correlation matrices into a sparse set of representative subnetworks that define a network manifold. These subnetworks are modeled as rank-one outer-products which correspond to the elemental patterns of co-activation across the brain; the subnetworks are combined via patient-specific non-negative coefficients. The second term is a linear regression model that uses the patient-specific coefficients to predict a measure of clinical severity. We validate our framework on two separate datasets in a ten fold cross validation setting. The first is a cohort of fifty-eight patients diagnosed with Autism Spectrum Disorder (ASD). The second dataset consists of sixty three patients from a publicly available ASD database. Our method outperforms standard semi-supervised frameworks, which employ conventional graph theoretic and statistical representation learning techniques to relate the rs-fMRI correlations to behavior. In contrast, our joint network optimization framework exploits the structure of the rs-fMRI correlation matrices to simultaneously capture group level effects and patient heterogeneity. Finally, we demonstrate that our proposed framework robustly identifies clinically relevant networks characteristic of ASD. △ Less

Submitted 27 August, 2020; originally announced September 2020.

Deep sr-DDL: Deep Structurally Regularized Dynamic Dictionary Learning to Integrate Multimodal and Dynamic Functional Connectomics data for Multidimensional Clinical Characterizations

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Deana Crocetti, Nicholas Wymbs, Joshua Robinson, Stewart H. Mostofsky, Archana Venkataraman

Abstract: We propose a novel integrated framework that jointly models complementary information from resting-state functional MRI (rs-fMRI) connectivity and diffusion tensor imaging (DTI) tractography to extract biomarkers of brain connectivity predictive of behavior. Our framework couples a generative model of the connectomics data with a deep network that predicts behavioral scores. The generative compone… ▽ More We propose a novel integrated framework that jointly models complementary information from resting-state functional MRI (rs-fMRI) connectivity and diffusion tensor imaging (DTI) tractography to extract biomarkers of brain connectivity predictive of behavior. Our framework couples a generative model of the connectomics data with a deep network that predicts behavioral scores. The generative component is a structurally-regularized Dynamic Dictionary Learning (sr-DDL) model that decomposes the dynamic rs-fMRI correlation matrices into a collection of shared basis networks and time varying subject-specific loadings. We use the DTI tractography to regularize this matrix factorization and learn anatomically informed functional connectivity profiles. The deep component of our framework is an LSTM-ANN block, which uses the temporal evolution of the subject-specific sr-DDL loadings to predict multidimensional clinical characterizations. Our joint optimization strategy collectively estimates the basis networks, the subject-specific time-varying loadings, and the neural network weights. We validate our framework on a dataset of neurotypical individuals from the Human Connectome Project (HCP) database to map to cognition and on a separate multi-score prediction task on individuals diagnosed with Autism Spectrum Disorder (ASD) in a five-fold cross validation setting. Our hybrid model outperforms several state-of-the-art approaches at clinical outcome prediction and learns interpretable multimodal neural signatures of brain organization. △ Less

Submitted 27 August, 2020; originally announced August 2020.

Ultra high-temperature deformation in a single crystal superalloy: Meso-scale process simulation and micro-mechanisms

Authors: Yuanbo T. Tang, Neil D'Souza, Bryan Roebuck, Phani Karamched, Chinnapat Panwisawas, David M. Collins

Abstract: A mesoscale study of a single crystal nickel-base superalloy subjected to an industrially relevant process simulation has revealed the complex interplay between microstructural development and the micromechanical behaviour. As sample gauge volumes were smaller than the length scale of the highly cored structure of the parent material from which they were produced, their subtle composition differen… ▽ More A mesoscale study of a single crystal nickel-base superalloy subjected to an industrially relevant process simulation has revealed the complex interplay between microstructural development and the micromechanical behaviour. As sample gauge volumes were smaller than the length scale of the highly cored structure of the parent material from which they were produced, their subtle composition differences gave rise to differing work hardening rates, influenced by varying secondary dendrite arm spacings, gamma-prime phase solvus temperatures and a topologically inverted gamma/gamma-prime microstructure. The gamma-prime precipitates possessed a characteristic `X' morphology, resulting from the simultaneously active solute transport mechanisms of thermally favoured octodendritic growth and N-type rafting, indicating creep-type mechanisms were prevalent. High resolution-electron backscatter diffraction (HR-EBSD) characterisation reveals deformation patterning that follows the gamma/gamma-prime microstructure, with high geometrically necessary dislocation density fields localised to the gamma/gamma-prime interfaces; Orowan loo** is evidently the mechanism that mediated plasticity. Examination of the residual elastic stresses indicated the `X' gamma-prime precipitate morphology had significantly enhanced the deformation heterogeneity, resulting in stress states within the gamma channels that favour slip, and that encourage further growth of gamma-prime precipitate protrusions. The combination of such localised plasticity and residual stresses are considered to be critical in the formation of the recrystallisation defect in subsequent post-casting homogenisation heat treatments. △ Less

Submitted 17 August, 2020; originally announced August 2020.

A Deep-Generative Hybrid Model to Integrate Multimodal and Dynamic Connectivity for Predicting Spectrum-Level Deficits in Autism

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Deana Crocetti, Nicholas Wymbs, Joshua Robinson, Stewart Mostofsky, Archana Venkataraman

Abstract: We propose an integrated deep-generative framework, that jointly models complementary information from resting-state functional MRI (rs-fMRI) connectivity and diffusion tensor imaging (DTI) tractography to extract predictive biomarkers of a disease. The generative part of our framework is a structurally-regularized Dynamic Dictionary Learning (sr-DDL) model that decomposes the dynamic rs-fMRI corr… ▽ More We propose an integrated deep-generative framework, that jointly models complementary information from resting-state functional MRI (rs-fMRI) connectivity and diffusion tensor imaging (DTI) tractography to extract predictive biomarkers of a disease. The generative part of our framework is a structurally-regularized Dynamic Dictionary Learning (sr-DDL) model that decomposes the dynamic rs-fMRI correlation matrices into a collection of shared basis networks and time varying patient-specific loadings. This matrix factorization is guided by the DTI tractography matrices to learn anatomically informed connectivity profiles. The deep part of our framework is an LSTM-ANN block, which models the temporal evolution of the patient sr-DDL loadings to predict multidimensional clinical severity. Our coupled optimization procedure collectively estimates the basis networks, the patient-specific dynamic loadings, and the neural network weights. We validate our framework on a multi-score prediction task in 57 patients diagnosed with Autism Spectrum Disorder (ASD). Our hybrid model outperforms state-of-the-art baselines in a five-fold cross validated setting and extracts interpretable multimodal neural signatures of brain dysfunction in ASD. △ Less

Submitted 3 July, 2020; originally announced July 2020.

Integrating Neural Networks and Dictionary Learning for Multidimensional Clinical Characterizations from Functional Connectomics Data

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Nicholas Wymbs, Stewart Mostofsky, Archana Venkataraman

Abstract: We propose a unified optimization framework that combines neural networks with dictionary learning to model complex interactions between resting state functional MRI and behavioral data. The dictionary learning objective decomposes patient correlation matrices into a collection of shared basis networks and subject-specific loadings. These subject-specific features are simultaneously input into a n… ▽ More We propose a unified optimization framework that combines neural networks with dictionary learning to model complex interactions between resting state functional MRI and behavioral data. The dictionary learning objective decomposes patient correlation matrices into a collection of shared basis networks and subject-specific loadings. These subject-specific features are simultaneously input into a neural network that predicts multidimensional clinical information. Our novel optimization framework combines the gradient information from the neural network with that of a conventional matrix factorization objective. This procedure collectively estimates the basis networks, subject loadings, and neural network weights most informative of clinical severity. We evaluate our combined model on a multi-score prediction task using 52 patients diagnosed with Autism Spectrum Disorder (ASD). Our integrated framework outperforms state-of-the-art methods in a ten-fold cross validated setting to predict three different measures of clinical severity. △ Less

Submitted 3 July, 2020; originally announced July 2020.

A Coupled Manifold Optimization Framework to Jointly Model the Functional Connectomics and Behavioral Data Spaces

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Nicholas Wymbs, Stewart Mostofsky, Archana Venkataraman

Abstract: The problem of linking functional connectomics to behavior is extremely challenging due to the complex interactions between the two distinct, but related, data domains. We propose a coupled manifold optimization framework which projects fMRI data onto a low dimensional matrix manifold common to the cohort. The patient specific loadings simultaneously map onto a behavioral measure of interest via a… ▽ More The problem of linking functional connectomics to behavior is extremely challenging due to the complex interactions between the two distinct, but related, data domains. We propose a coupled manifold optimization framework which projects fMRI data onto a low dimensional matrix manifold common to the cohort. The patient specific loadings simultaneously map onto a behavioral measure of interest via a second, non-linear, manifold. By leveraging the kernel trick, we can optimize over a potentially infinite dimensional space without explicitly computing the embeddings. As opposed to conventional manifold learning, which assumes a fixed input representation, our framework directly optimizes for embedding directions that predict behavior. Our optimization algorithm combines proximal gradient descent with the trust region method, which has good convergence guarantees. We validate our framework on resting state fMRI from fifty-eight patients with Autism Spectrum Disorder using three distinct measures of clinical severity. Our method outperforms traditional representation learning techniques in a cross validated setting, thus demonstrating the predictive power of our coupled objective. △ Less

Submitted 3 July, 2020; originally announced July 2020.

Spinodal decomposition versus classical gamma-prime nucleation in a nickel-base superalloy powder: An in-situ neutron diffraction and atomic-scale analysis

Authors: David M Collins, Neil D'Souza, Chinnapat Panwisawas, Chrysanthi Papadaki, Geoff D West, Aleksander Kostka, Paraskevas Kontis

Abstract: Contemporary powder-based polycrystalline nickel-base superalloys inherit microstructures and properties that are heavily determined by the thermo-mechanical treatments during processing. Here, the influence of a thermal exposure alone to an alloy powder is studied to elucidate the controlling formation mechanisms of the strengthening precipitates using a combination of atom probe tomography and i… ▽ More Contemporary powder-based polycrystalline nickel-base superalloys inherit microstructures and properties that are heavily determined by the thermo-mechanical treatments during processing. Here, the influence of a thermal exposure alone to an alloy powder is studied to elucidate the controlling formation mechanisms of the strengthening precipitates using a combination of atom probe tomography and in-situ neutron diffraction. The initial powder comprised a single-phase supersaturated gamma only; from this, the evolution of gamma-prime volume fraction and lattice misfit was assessed. The initial powder notably possessed elemental segregation of Cr and Co and elemental repulsion between Ni, Al and Ti with Cr; here proposed to be a precursor for subsequent gamma to gamma-prime phase transformations. Subsolvus heat treatments yielded a unimodal gamma-prime distribution, formed during heating, with evidence supporting its formation to be via spinodal decomposition. A supersolvus heat treatment led to the formation of this same gamma-prime population during heating, but dissolves as the temperature increases further. The gamma-prime then reprecipitates as a multimodal population during cooling, here forming by classical nucleation and growth. Atom probe characterisation provided intriguing precipitate characteristics, including clear differences in chemistry and microstructure, depending on whether the gamma-prime formed during heating or cooling. △ Less

Submitted 17 October, 2020; v1 submitted 2 April, 2020; originally announced April 2020.

Journal ref: Acta Materialia 200 (2020) 959-970

City Planning with Augmented Reality

Authors: Catherine Angelini, Adam S. Williams, Mathew Kress, Edgar Ramos Vieira, Newton D'Souza, Naphtali D. Rishe, Joseph Medina, Francisco R. Ortega

Abstract: We present an early study designed to analyze how city planning and the health of senior citizens can benefit from the use of augmented reality (AR) using Microsoft's HoloLens. We also explore whether AR and VR can be used to help city planners receive real-time feedback from citizens, such as the elderly, on virtual plans, allowing for informed decisions to be made before any construction begins. We present an early study designed to analyze how city planning and the health of senior citizens can benefit from the use of augmented reality (AR) using Microsoft's HoloLens. We also explore whether AR and VR can be used to help city planners receive real-time feedback from citizens, such as the elderly, on virtual plans, allowing for informed decisions to be made before any construction begins. △ Less

Submitted 17 January, 2020; originally announced January 2020.

Comments: This was accepted by Graphics Interface 2019 (GI '2019) HCI track as a poster paper

Rapid Uncertainty Propagation and Chance-Constrained Path Planning for Small Unmanned Aerial Vehicles

Authors: Andrew W. Berning Jr., Anouck Girard, Ilya Kolmanovsky, Sarah N. D'Souza

Abstract: With the number of small Unmanned Aircraft Systems (sUAS) in the national airspace projected to increase in the next few years, there is growing interest in a traffic management system capable of handling the demands of this aviation sector. It is expected that such a system will involve trajectory prediction, uncertainty propagation, and path planning algorithms. In this work, we use linear covar… ▽ More With the number of small Unmanned Aircraft Systems (sUAS) in the national airspace projected to increase in the next few years, there is growing interest in a traffic management system capable of handling the demands of this aviation sector. It is expected that such a system will involve trajectory prediction, uncertainty propagation, and path planning algorithms. In this work, we use linear covariance propagation in combination with a quadratic programming-based collision detection algorithm to rapidly validate declared flight plans. Additionally, these algorithms are combined with a Dynamic, Informed RRT* algorithm, resulting in a computationally efficient algorithm for chance-constrained path planning. Detailed numerical examples for both fixed-wing and quadrotor sUAS models are presented. △ Less

Submitted 6 November, 2019; originally announced November 2019.

Comments: Submitted to Advanced Control for Applications

Energy-efficient switching of nanomagnets for computing: Straintronics and other methodologies

Authors: Noel D'Souza, Ayan Biswas, Hasnain Ahmad, Mohammad Salehi Fashami, Md Mamun Al-Rashid, Vimal Sampath, Dhritiman Bhattacharya, Md Ahsanul Abeed, Jayasimha Atulasimha, Supriyo Bandyopadhyay

Abstract: The need for increasingly powerful computing hardware has spawned many ideas stipulating, primarily, the replacement of traditional transistors with alternate "switches" that dissipate miniscule amounts of energy when they switch and provide additional functionality that are beneficial for information processing. An interesting idea that has emerged recently is the notion of using two-phase (piezo… ▽ More The need for increasingly powerful computing hardware has spawned many ideas stipulating, primarily, the replacement of traditional transistors with alternate "switches" that dissipate miniscule amounts of energy when they switch and provide additional functionality that are beneficial for information processing. An interesting idea that has emerged recently is the notion of using two-phase (piezoelectric/magnetostrictive) multiferroic nanomagnets with bistable (or multi-stable) magnetization states to encode digital information (bits), and switching the magnetization between these states with small voltages (that strain the nanomagnets) to carry out digital information processing. The switching delay is ~1 ns and the energy dissipated in the switching operation can be few to tens of aJ, which is comparable to, or smaller than, the energy dissipated in switching a modern-day transistor. Unlike a transistor, a nanomagnet is "non-volatile", so a nanomagnetic processing unit can store the result of a computation locally without refresh cycles, thereby allowing it to double as both logic and memory. These dual-role elements promise new, robust, energy-efficient, high-speed computing and signal processing architectures (usually non-Boolean and often non-von-Neumann) that can be more powerful, architecturally superior (fewer circuit elements needed to implement a given function) and sometimes faster than their traditional transistor-based counterparts. This topical review covers the important advances in computing and information processing with nanomagnets with emphasis on strain-switched multiferroic nanomagnets acting as non-volatile and energy-efficient switches - a field known as "straintronics". It also outlines key challenges in straintronics. △ Less

Submitted 22 September, 2018; originally announced September 2018.

Comments: This is a commissioned topical review article published in Nanotechnology

Journal ref: Nanotechnology, vol 44, 442001 (2018)

A Generative-Discriminative Basis Learning Framework to Predict Clinical Severity from Resting State Functional MRI Data

Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Nicholas Wymbs, Stewart Mostofsky, Archana Venkataraman

Abstract: We propose a matrix factorization technique that decomposes the resting state fMRI (rs-fMRI) correlation matrices for a patient population into a sparse set of representative subnetworks, as modeled by rank one outer products. The subnetworks are combined using patient specific non-negative coefficients; these coefficients are also used to model, and subsequently predict the clinical severity of a… ▽ More We propose a matrix factorization technique that decomposes the resting state fMRI (rs-fMRI) correlation matrices for a patient population into a sparse set of representative subnetworks, as modeled by rank one outer products. The subnetworks are combined using patient specific non-negative coefficients; these coefficients are also used to model, and subsequently predict the clinical severity of a given patient via a linear regression. Our generative-discriminative framework is able to exploit the structure of rs-fMRI correlation matrices to capture group level effects, while simultaneously accounting for patient variability. We employ ten fold cross validation to demonstrate the predictive power of our model on a cohort of fifty eight patients diagnosed with Autism Spectrum Disorder. Our method outperforms classical semi-supervised frameworks, which perform dimensionality reduction on the correlation features followed by non-linear regression to predict the clinical scores. △ Less

Submitted 24 July, 2018; originally announced July 2018.

Giant Voltage Manipulation of MgO-based Magnetic Tunnel Junctions via Localized Anisotropic Strain: a Potential Pathway to Ultra-Energy-Efficient Memory Technology

Authors: Zhengyang Zhao, Mahdi Jamali, Noel D'Souza, Delin Zhang, Supriyo Bandyopadhyay, Jayasimha Atulasimha, Jian-** Wang

Abstract: Strain-mediated voltage control of magnetization in piezoelectric/ferromagnetic systems is a promising mechanism to implement energy-efficient spintronic memory devices. Here, we demonstrate giant voltage manipulation of MgO magnetic tunnel junctions (MTJ) on a Pb(Mg1/3Nb2/3)0.7Ti0.3O3 (PMN-PT) piezoelectric substrate with (001) orientation. It is found that the magnetic easy axis, switching field… ▽ More Strain-mediated voltage control of magnetization in piezoelectric/ferromagnetic systems is a promising mechanism to implement energy-efficient spintronic memory devices. Here, we demonstrate giant voltage manipulation of MgO magnetic tunnel junctions (MTJ) on a Pb(Mg1/3Nb2/3)0.7Ti0.3O3 (PMN-PT) piezoelectric substrate with (001) orientation. It is found that the magnetic easy axis, switching field, and the tunnel magnetoresistance (TMR) of the MTJ can be efficiently controlled by strain from the underlying piezoelectric layer upon the application of a gate voltage. Repeatable voltage controlled MTJ toggling between high/low-resistance states is demonstrated. More importantly, instead of relying on the intrinsic anisotropy of the piezoelectric substrate to generate the required strain, we utilize anisotropic strain produced using local gating scheme, which is scalable and amenable to practical memory applications. Additionally, the adoption of crystalline MgO-based MTJ on piezoelectric layer lends itself to high TMR in the strain-mediated MRAM devices. △ Less

Submitted 29 August, 2016; v1 submitted 18 February, 2016; originally announced February 2016.

Incoherent magnetization dynamics in strain mediated switching of magnetostrictive nanomagnets

Authors: Dhritiman Bhattacharya, Md Mamun Al-Rashid, Noel D'Souza, Supriyo Bandyopadhyay, Jayasimha Atulasimha

Abstract: Micromagnetic studies of the magnetization change in magnetostrictive nanomagnets subjected to stress are performed for nanomagnets of different sizes. The interplay between demagnetization, exchange and stress anisotropy energies is used to explain the rich physics of size-dependent magnetization dynamics induced by modulating stress anisotropy in planar nanomagnets. These studies have important… ▽ More Micromagnetic studies of the magnetization change in magnetostrictive nanomagnets subjected to stress are performed for nanomagnets of different sizes. The interplay between demagnetization, exchange and stress anisotropy energies is used to explain the rich physics of size-dependent magnetization dynamics induced by modulating stress anisotropy in planar nanomagnets. These studies have important implications for strain mediated ultralow energy magnetization change in nanomagnets and its application in energy-efficient nanomagnetic computing systems. △ Less

Submitted 14 November, 2015; originally announced November 2015.

Comments: main paper and supplement

Experimental Clocking of Nanomagnets with Strain for Ultra Low Power Boolean Logic

Authors: Noel D'Souza, Mohammad Salehi Fashami, Supriyo Bandyopadhyay, Jayasimha Atulasimha

Abstract: Nanomagnetic implementations of Boolean logic [1,2] have garnered attention because of their non-volatility and the potential for unprecedented energy-efficiency. Unfortunately, the large dissipative losses that take place when nanomagnets are switched with a magnetic field [3], or spin-transfer-torque [4] inhibit the promised energy-efficiency. Recently, there have been experimental reports of ut… ▽ More Nanomagnetic implementations of Boolean logic [1,2] have garnered attention because of their non-volatility and the potential for unprecedented energy-efficiency. Unfortunately, the large dissipative losses that take place when nanomagnets are switched with a magnetic field [3], or spin-transfer-torque [4] inhibit the promised energy-efficiency. Recently, there have been experimental reports of utilizing the Spin Hall effect for switching magnets [5-7], and theoretical proposals for strain induced switching of single-domain magnetostrictive nanomagnets [8-12], that might reduce the dissipative losses significantly. Here, we demonstrate, for the first time, that strain-induced switching of single-domain magnetostrictive nanomagnets of lateral dimensions ~200 nm fabricated on a piezoelectric substrate can implement a nanomagnetic Boolean NOT gate and unidirectional bit information propagation in dipole-coupled nanomagnets chains. This portends ultra-low-energy logic processors and mobile electronics that may be able to operate solely by harvesting energy from the environment without ever requiring a battery. △ Less

Submitted 21 May, 2015; v1 submitted 10 April, 2014; originally announced April 2014.

Comments: New version

Exploring Performance, Coherence, and Clocking of Magnetization in Multiferroic Four-State Nanomagnets

Authors: Mohammad Salehi Fashami, Noel D'Souza

Abstract: Nanomagnetic memory and logic are currently seen as promising candidates to replace current digital computing architectures due to its superior energy-efficiency, non-volatility and propensity for highly dense and low-power applications. In this work, we investigate the use of shape engineering (concave and diamond shape) to introduce biaxial anisotropy in single domain nanomagnets, giving rise to… ▽ More Nanomagnetic memory and logic are currently seen as promising candidates to replace current digital computing architectures due to its superior energy-efficiency, non-volatility and propensity for highly dense and low-power applications. In this work, we investigate the use of shape engineering (concave and diamond shape) to introduce biaxial anisotropy in single domain nanomagnets, giving rise to multiple easy and hard axes. Such nanomagnets, with dimensions of 100 nm x 100 nm, double the logic density of conventional two-state nanomagnetic devices by encoding more information (four binary bits:00,11,10,01) per nanomagnet and can be used in memory and logic devices as well as in higher order information processing applications. We study reliability, magnetization switching coherence, and show, for the first time, the use of voltage-induced strain for the clocking of magnetization in these four-state nanomagnets. Critical parameters such as size, thickness, concavity, and geometry of two types of four-state nanomagnets are also investigated. This analytical study provides important insights into achieving reliable and coherent single domain nanomagnets and low-energy magnetization clocking in four-state nanomagnets, paving the way for potential applications in advanced technologies. △ Less

Submitted 27 April, 2017; v1 submitted 10 March, 2014; originally announced March 2014.

An ultrafast image recovery and recognition system implemented with nanomagnets possessing biaxial magnetocrystalline anisotropy

Authors: Noel D'Souza, Jayasimha Atulasimha, Supriyo Bandyopadhyay

Abstract: A circular magnetic disk with biaxial magnetocrystalline anisotropy has four stable magnetization states which can be used to encode a pixel's shade in a black/gray/white image. By solving the Landau-Lifshitz- Gilbert equation, we show that if moderate noise deflects the magnetization slightly from a stable state, it always returns to the original state, thereby automatically de-noising the corrup… ▽ More A circular magnetic disk with biaxial magnetocrystalline anisotropy has four stable magnetization states which can be used to encode a pixel's shade in a black/gray/white image. By solving the Landau-Lifshitz- Gilbert equation, we show that if moderate noise deflects the magnetization slightly from a stable state, it always returns to the original state, thereby automatically de-noising the corrupted image. The same system can compare a noisy input image with a stored image and make a matching decision using magneto-tunneling junctions. These tasks are executed at ultrahigh speeds (~2 ns for a 512\times512 pixel image). △ Less

Submitted 30 September, 2011; originally announced September 2011.

Comments: Submitted to Applied Physics Letters

Journal ref: IEEE Transactions on Nanotechnology, Vol. 11, p. 896, 2012

An Energy-Efficient Bennett Clocking Scheme for 4-State Multiferroic Logic

Authors: Noel D'Souza, Jayasimha Atulasimha, Supriyo Bandyopadhyay

Abstract: Nanomagnets with biaxial magnetocrystalline anisotropy have four stable magnetization orientations that can encode 4-state logic bits (00), (01), (11) and (10). Recently, a 4-state NOR gate derived from three such nanomagnets, interacting via dipole interaction, was proposed. Here, we devise a Bennett clocking scheme to propagate 4-state logic bits unidirectionally between such gates. The nanomagn… ▽ More Nanomagnets with biaxial magnetocrystalline anisotropy have four stable magnetization orientations that can encode 4-state logic bits (00), (01), (11) and (10). Recently, a 4-state NOR gate derived from three such nanomagnets, interacting via dipole interaction, was proposed. Here, we devise a Bennett clocking scheme to propagate 4-state logic bits unidirectionally between such gates. The nanomagnets are assumed to be made of 2-phase strain-coupled magnetostrictive/piezoelectric multiferroic elements, such as nickel and lead zirconate titanate (PZT). A small voltage of 200 mV applied across the piezoelectric layer can generate enough mechanical stress in the magnetostrictive layer to rotate its magnetization away from one of the four stable orientations and implement Bennett clocking. We show that a particular sequence of positive and negative voltages will propagate 4-state logic bits unidirectionally down a chain of such multiferroic nanomagnets for logic flow. △ Less

Submitted 9 May, 2011; originally announced May 2011.

Comments: Manuscript submitted for publication in IEEE Transactions on Nanotechnology

Journal ref: IEEE Transactions on Nanotechnology, vol. 11, no. 2, pp. 418-425, March 2012

Four-state nanomagnetic logic using multiferroics

Authors: Noel D'Souza, Jayasimha Atulasimha, Supriyo Bandyopadhyay

Abstract: The authors show how to implement a 4-state universal logic gate (NOR) using three strain-coupled magnetostrictive-piezoelectric multiferroic nanomagnets (e.g. Ni/PZT) with biaxial magnetocrystalline anisotropy. Two of the nanomagnets encode the 2-state input bits in their magnetization orientations and the third nanomagnet produces the output bit via dipole interaction with the input nanomagnets.… ▽ More The authors show how to implement a 4-state universal logic gate (NOR) using three strain-coupled magnetostrictive-piezoelectric multiferroic nanomagnets (e.g. Ni/PZT) with biaxial magnetocrystalline anisotropy. Two of the nanomagnets encode the 2-state input bits in their magnetization orientations and the third nanomagnet produces the output bit via dipole interaction with the input nanomagnets. A voltage pulse alternating between -0.2 V and +0.2 V is applied to the PZT layer of the third nanomagnet and generates alternating tensile and compressive stress in its Ni layer to produce the output bit, while dissipating ~ 57,000 kT (0.24 fJ) of energy per gate operation. △ Less

Submitted 5 January, 2011; originally announced January 2011.

Journal ref: Journal of Physics D: Applied Physics, Vol. 44, 265001 (2011)