About

Deniz Erdogmus is a COE Distinguished Professor at Northeastern University with a research focus on artificial intelligence, machine learning, signal and image analysis, and their applications in various domains. His work emphasizes human-centric foundational and physical AI, AI for health, quality-of-life, clinical innovation, scientific discovery, and engineered systems. Erdogmus received his BS in Electrical Engineering and Mathematics in 1997 and his MS in Electrical Engineering in 1999 from Middle East Technical University (METU) in Turkey. He earned his PhD in Electrical and Computer Engineering from the University of Florida in 2002. He has served as an associate editor for various journals and has been an active member of IEEE technical committees for MLSP and BISP. Since joining Northeastern University in 2008, Erdogmus has contributed significantly to research in signal processing, machine learning, and AI, leading projects in areas such as cyber-physical systems, biomedical AI, and human-centric AI. His research is conducted through the Cognitive Systems Laboratory (CSL), which is part of multiple institutes and consortia focused on experiential AI, robotics, signal processing, and neurotechnology. Erdogmus has received numerous honors and awards, including the 2024 Distinguished Faculty Award, the 2021 Ruth and Joel Spira Award for Excellence in Teaching, and the 2012 NSF CAREER Award, among others.

Research topics

• Artificial Intelligence
• Computer Science
• Psychology
• Cognitive psychology
• Cognitive science
• Radiology
• Medicine
• Mathematics
• Computer vision
• Neuroscience
• Pathology

Selected publications

• Temporal point process modeling of aggressive behavior onset in psychiatric inpatient youths with autism

  Scientific Reports · 2026-04-18

  articleOpen access

  Aggressive behavior, including aggression towards others and self-injury, occurs in up to 80% of children and adolescents with autism, making it a leading cause of behavioral health referrals and a major driver of healthcare costs. Predicting when autistic youth will exhibit aggression can be challenging due to their communication difficulties. Many are minimally verbal or have poor emotional insight. Recent advances in Machine Learning and wearable biosensing demonstrate the ability to predict aggression within a limited future window (typically one to three minutes) in autistic individuals. However, existing works do not estimate aggression onset probability or the expected number of aggression onsets over longer periods, nor do they provide interpretable insights into onset dynamics. To address these limitations, we apply Temporal Point Processes(TPPs), particularly self-exciting Hawkes processes, to model the timing of aggressive behavior onsets in psychiatric inpatient autistic youth. We benchmark several TPP models by evaluating their goodness-of-fit and predictive metrics. Our results demonstrate that self-exciting TPPs more accurately capture the irregular and clustered nature of aggression onsets, especially compared to traditional Poisson models. These incipient findings suggest that TPPs can provide interpretable, probabilistic forecasts of aggression onset along a time continuum, supporting future clinical decision-making and preemptive intervention.

  PublisherOA PDFDOI

• SSplain: Sparse and Smooth Explainer for Retinopathy of Prematurity Classification

  2026-03-06

  article

  Neural networks are frequently used in medical diagnosis. However, due to their black-box nature, model explainers are used to help clinicians understand better and trust model outputs. This paper introduces an explainer method for classifying Retinopathy of Prematurity (ROP) from fundus images. Previous methods fail to generate explanations that preserve input image structures such as smoothness and sparsity. We introduce Sparse and Smooth Explainer (SSplain), a method that generates pixel-wise explanations while preserving image structures by enforcing smoothness and sparsity. This results in realistic explanations to enhance the understanding of the given black-box model. To achieve this goal, we define an optimization problem with combinatorial constraints and solve it using the Alternating Direction Method of Multipliers (ADMM). Experimental results show that SSplain outperforms commonly used explainers in terms of both post-hoc accuracy and smoothness analyses. Additionally, SSplain identifies features that are consistent with domain-understandable features that clinicians consider as discriminative factors for ROP. We also show SSplain’s generalization by applying it to additional publicly available datasets. Code is available at https://github.com/neu-spiral/SSplain.

  PublisherDOI

• Corticomorphic Hybrid CNN-SNN Architecture for EEG-Based Low-Footprint Low-Latency Auditory Attention Detection

  Annals of Biomedical Engineering · 2026-03-17

  article
  PublisherDOI

• Fast and Robust State Estimation and Tracking via Hierarchical Learning

  IEEE Transactions on Automatic Control · 2025-10-13

  article

  Fast and reliable state estimation and tracking are essential for real-time situation awareness in Cyber-Physical Systems (CPS) operating in tactical environments or complicated civilian environments. Traditional centralized solutions do not scale well whereas existing fully distributed solutions over large networks suffer slow convergence, and are vulnerable to a wide spectrum of communication failures. In this paper, we aim to speed up the convergence and enhance the resilience of state estimation and tracking for large-scale networks using a simple hierarchical system architecture. We propose two “consensus + innovation” algorithms, both of which rely on a novel hierarchical push-sum consensus component. We characterize their convergence rates under a linear local observation model and minimal technical assumptions. We numerically validate our algorithms through simulation studies of underwater acoustic networks and large-scale synthetic networks.

  PublisherDOI

• Single-shot Center-Overlapped EPI for distortion-frEe diffusion MRI (SCOPE)

  Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

  article

  Motivation: Single-shot EPI is prone to image distortion from B0 inhomogeneity and long echo time (TE), especially in diffusion MRI (dMRI). Goal(s): This study proposes a novel single-shot center-overlapped EPI readout, termed SCOPE, designed for distortion-free dMRI by correcting B0-induced distortions in each diffusion direction and reducing TE. Approach: SCOPE uses two overlapped k-space segments with opposing blip directions to derive ∆B0 maps directly from single-shot data, eliminating the need for additional acquisitions. The sequence was tested in vivo on 3T. Results: The SCOPE sequence produced high-fidelity dMRI images with minimal distortion, achieving effective distortion correction comparable to TOPUP-processed images. Impact: We proposed a novel single-shot center-overlapped EPI readout for distortion-free EPI, SCOPE. With overlapped segments, our method estimates the ∆B0 field map from the single-shot data for each diffusion direction. Further, our method has the potential to reduce the TE.

  PublisherDOI

• MarkovType: A Markov Decision Process Strategy for Non-Invasive Brain-Computer Interfaces Typing Systems

  Proceedings of the AAAI Conference on Artificial Intelligence · 2025-04-11 · 1 citations

  articleOpen access

  Brain-Computer Interfaces (BCIs) help people with severe speech and motor disabilities communicate and interact with their environment using neural activity. This work focuses on the Rapid Serial Visual Presentation (RSVP) paradigm of BCIs using noninvasive electroencephalography (EEG). The RSVP typing task is a recursive task with multiple sequences, where users see only a subset of symbols in each sequence. Extensive research has been conducted to improve classification in the RSVP typing task, achieving fast classification. However, these methods struggle to achieve high accuracy and do not consider the typing mechanism in the learning procedure. They apply binary target and non-target classification without including recursive training. To improve performance in the classification of symbols while controlling the classification speed, we incorporate the typing setup into training by proposing a Partially Observable Markov Decision Process (POMDP) approach. To the best of our knowledge, this is the first work to formulate the RSVP typing task as a POMDP for recursive classification. Experiments show that the proposed approach, MarkovType, results in a more accurate typing system compared to competitors. Additionally, our experiments demonstrate that while there is a trade-off between accuracy and speed, MarkovType achieves the optimal balance between these factors compared to other methods.

  PublisherOA PDFDOI

• Exploring Theory-Laden Observations in the Brain Basis of Emotional Experience

  ArXiv.org · 2025-06-30

  preprintOpen access

  In the science of emotion, it is widely assumed that folk emotion categories form a biological and psychological typology, and studies are routinely designed and analyzed to identify emotion-specific patterns. This approach shapes the observations that studies report, ultimately reinforcing the assumption that guided the investigation. Here, we reanalyzed data from one such typologically-guided study that reported mappings between individual brain patterns and group-averaged ratings of 34 emotion categories. Our reanalysis was guided by an alternative view of emotion categories as populations of variable, situated instances, and which predicts a priori that there will be significant variation in brain patterns within a category across instances. Correspondingly, our analysis made minimal assumptions about the structure of the variance present in the data. As predicted, we did not observe the original mappings and instead observed significant variation across individuals. These findings demonstrate how starting assumptions can ultimately impact scientific conclusions and suggest that a hypothesis must be supported using multiple analytic methods before it is taken seriously.

  PublisherOA PDFDOI

• Continuously Optimizing Radar Placement With Model-Predictive Path Integrals

  IEEE Transactions on Aerospace and Electronic Systems · 2025-01-13

  article

  Continuously optimizing sensor placement is essential for precise target localization in various military and civilian applications. While information theory has shown promise in optimizing sensor placement, many studies oversimplify sensor measurement models or neglect dynamic constraints of mobile sensors. To address these challenges, we employ a range measurement model that incorporates radar parameters and radar–target distance, coupled with model-predictive path integral control to manage complex environmental obstacles and dynamic constraints. We compare the proposed approach against stationary radars or simplified range measurement models based on the root-mean-squared error (RMSE) of the cubature Kalman filter estimator for the targets' state. In addition, we visualize the evolving geometry of radars and targets over time, highlighting areas of highest measurement information gain, demonstrating the strengths of the approach. The proposed strategy outperforms stationary radars and simplified range measurement models in target localization, achieving a 38%–74% reduction in the mean RMSE and a 33%–79% reduction in the upper tail of the 90% highest density interval over 500 Monte Carlo trials across all time steps.

  PublisherDOI

• Advancing Multi-Person Tracking for Autism Behavior Analysis: Challenges, Opportunities, and Future Directions in Clinical Settings

  2025-02-28

  article

  Multi-person tracking (MPT) has emerged as a crucial tool for behavior monitoring in clinical settings, offering significant potential for advancing autism research. In this paper, we assess the state of the art in MPT, with a specific focus on its application to autism spectrum disorder (ASD) behavior analysis. Clinical environments present unique challenges for MPT, including frequent occlusions, visually similar appearances such as uniforms, and complex social interactions. We explore these challenges in depth while highlighting recent advancements in tracking algorithms that present opportunities for improving behavioral monitoring in healthcare. Through a comprehensive review, we examine the limitations of existing MPT methods when applied to our clinical dataset, BAV-ASD (behavioral analysis videoset for ASD), discussing their applicability and gaps in addressing the unique demands of clinical autism studies. Our findings emphasize the critical role of MPT in enabling non-intrusive, accurate, and continuous monitoring of challenging behaviors, such as aggression and self-injury, in individuals with ASD. By addressing the limitations of existing methods, this work paves the way for future research aimed at enhancing the reliability and effectiveness of MPT systems in clinical environments, ultimately supporting early intervention strategies and improving out-comes for individuals with ASD <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>Our code can be found at https://github.com/ostadabbas/ASD-Tracking-Challenges. Supported by NSF-CAREER Grant #2143882 and #R01LM014191..

  PublisherDOI

• Exploring Theory-Laden Observations in the Brain Basis of Emotional Experience

  2025-01-24

  preprintOpen access

  In the science of emotion, it is widely assumed that folk emotion categories form a biological and psychological typology, and studies are routinely designed and analyzed to identify emotion-specific patterns. This approach shapes the observations that studies report, ultimately reinforcing the assumption that guided the investigation. Here, we reanalyzed data from one such typologically-guided study that reported mappings between individual brain patterns and group-averaged ratings of 34 emotion categories. Our reanalysis was guided by an alternative view of emotion categories as populations of variable, situated instances, and which predicts a priori that there will be significant variation in brain patterns within a category across instances. Correspondingly, our analysis made minimal assumptions about the structure of the variance present in the data. As predicted, we did not observe the original mappings and instead observed significant variation across individuals. These findings demonstrate how starting assumptions can ultimately impact scientific conclusions and suggest that a hypothesis must be supported using multiple analytic methods before it is taken seriously.

  PublisherOA PDFDOI

Recent grants

• SCH: INT: Collaborative Research: Multimodal Signal Analysis and Data Fusion for Post-traumatic Epilepsy

  NIH · $981k · 2019–2024

• Collaborative Research: CDI-Type I: Computational Models for the Automatic Recognition of Non-Human Primate Social Behaviors

  NSF · $156k · 2010–2014

• HCC-Small: RSVP IconCHAT - A Brain Computer Interface for Icon-based Communication

  NSF · $504k · 2009–2013

• CAREER: Signal Models, Channel Capacity, and Information Rate for Noninvasive Brain Interfaces

  NSF · $521k · 2012–2019

• CPS: TTP Option: Synergy: Collaborative Research: Nested Control of Assistive Robots through Human Intent Inference

  NSF · $603k · 2015–2020

Frequent coauthors

• Jayashree Kalpathy‐Cramer

  University of Colorado Anschutz Medical Campus

  155 shared

• José C. Prı́ncipe

  University of Florida

  137 shared

• Michael F. Chiang

  National Eye Institute

  109 shared

• Tales Imbiriba

  78 shared

• J. Peter Campbell

  75 shared

• Susan Ostmo

  Oregon Health & Science University

  71 shared

• Murat Akçakaya

  University of Pittsburgh

  66 shared

• R.V. Paul Chan

  University of Illinois Chicago

  64 shared

Labs

• Cognitive Systems Laboratory (CSL)PI

Awards & honors

• 2024 Distinguished Faculty Award
• 2021 ECE Ruth and Joel Spira Award for Excellence in Teachin…
• 2021 COE Faculty Research Team Award
• 2019 COE Excellence in Mentoring Award
• 2014 COE Faculty Fellow