Gustavo Rohde

About

Professor Gustavo Rohde is associated with the Imaging and Data Science Laboratory at the University of Virginia, where he advances the science of imaging, signal analysis, and data-driven discovery. His work focuses on transforming signals, images, and data into meaningful knowledge through the development of rigorous mathematical, computational, and machine learning methods. His research includes transport-based representations, digital pathology, transport-based morphometry for biomedical imaging, and mobile computer vision. Professor Rohde has delivered numerous seminars and colloquia, including at the University of Virginia, University of Nebraska-Lincoln, and George Washington University. He has received recognition such as the Mac Wade Professor of Engineering endowed chair at the University of Virginia and became an AIMBE Fellow. His recent contributions involve end-to-end signal classification and MRI-based classification of genetic copy number variations in autism, reflecting his focus on applying advanced mathematical and computational techniques to biomedical imaging and signal analysis.

Research topics

• Artificial Intelligence
• Computer Science
• Medicine
• Pathology
• Machine Learning
• Ophthalmology
• Cardiology
• Psychology
• Statistics
• Internal medicine

Selected publications

• Massive image-based single-cell profiling reveals high levels of circulating platelet aggregates in patients with COVID-19

  Nature Communications · 2021 · 82 citations

  A characteristic clinical feature of COVID-19 is the frequent incidence of microvascular thrombosis. In fact, COVID-19 autopsy reports have shown widespread thrombotic microangiopathy characterized by extensive diffuse microthrombi within peripheral capillaries and arterioles in lungs, hearts, and other organs, resulting in multiorgan failure. However, the underlying process of COVID-19-associated microvascular thrombosis remains elusive due to the lack of tools to statistically examine platelet…

  DOI

• Enabling early detection of osteoarthritis from presymptomatic cartilage texture maps via transport-based learning

  Proceedings of the National Academy of Sciences · 2020 · 63 citations

  Senior authorCorresponding

  Many diseases have no visual cues in the early stages, eluding image-based detection. Today, osteoarthritis (OA) is detected after bone damage has occurred, at an irreversible stage of the disease. Currently no reliable method exists for OA detection at a reversible stage. We present an approach that enables sensitive OA detection in presymptomatic individuals. Our approach combines optimal mass transport theory with statistical pattern recognition. Eighty-six healthy individuals were selected f…

  DOI

• Video-Based Facial Weakness Analysis

  IEEE Transactions on Biomedical Engineering · 2021 · 21 citations

  Senior authorCorresponding

  OBJECTIVE: Facial weakness is a common sign of neurological diseases such as Bell's palsy and stroke. However, recognizing facial weakness still remains as a challenge, because it requires experience and neurological training. METHODS: We propose a framework for facial weakness detection, which models the temporal dynamics of both shape and appearance-based features of each target frame through a bi-directional long short-term memory network (Bi-LSTM). The system is evaluated on a "in-the-wild"v…

  PublisherDOI

• Discovering the gene-brain-behavior link in autism via generative machine learning

  Science Advances · 2024-06-12 · 16 citations

  articleOpen accessSenior authorCorresponding

  Autism is traditionally diagnosed behaviorally but has a strong genetic basis. A genetics-first approach could transform understanding and treatment of autism. However, isolating the gene-brain-behavior relationship from confounding sources of variability is a challenge. We demonstrate a novel technique, 3D transport-based morphometry (TBM), to extract the structural brain changes linked to genetic copy number variation (CNV) at the 16p11.2 region. We identified two distinct endophenotypes. In d…

  PublisherDOI

• End-to-End Signal Classification in Signed Cumulative Distribution Transform Space

  IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-03-01 · 15 citations

  articleOpen accessSenior author

  This paper presents a new end-to-end signal classification method using the signed cumulative distribution transform (SCDT). We adopt a transport generative model to define the classification problem. We then make use of mathematical properties of the SCDT to render the problem easier in transform domain, and solve for the class of an unknown sample using a nearest local subspace (NLS) search algorithm in SCDT domain. Experiments show that the proposed method provides high accuracy classificatio…

  PublisherOA PDFDOI

Recent grants

• Utility of Effusion Cytology and Image Analysis in the Diagnosis of Mesothelioma

  NIH · $193k · 2014–2017

• CIF: Small: Transport and other Lagrangian transforms for signal analysis and discrimination

  NSF · $500k · 2014–2017

• ABI: Innovation: Analyzing Neuroglial Cell Dynamics in their Natural Environment with Video Microscopy

  NSF · $629k · 2018–2022

Frequent coauthors

• Soheil Kolouri

  Vanderbilt University

  53 shared

• Yan Zhuang

  National Institutes of Health Clinical Center

  45 shared

• Abu Hasnat Mohammad Rubaiyat

  University of Virginia

  43 shared

• Mohammad Shifat‐E‐Rabbi

  University of Virginia

  41 shared

• Keisuke Goda

  University of California, Los Angeles

  41 shared

• Shiying Li

  32 shared

• Xuwang Yin

  University of Virginia

  32 shared

• John A. Ozolek

  29 shared

Labs

• IMAGING AND DATA SCIENCE LABPI

Awards & honors

• Mac Wade Professor of Engineering endowed chair at the Unive…
• AIMBE Fellow

Similar researchers at University of Virginia

• Todd Bauerh-61
• Benjamin (BJ) Purowh-50
• Dave Kashatush-34