New Approaches to the Analysis of Modern Imaging Modalities

Jiangmei Xiong Chair
 
Wednesday, Aug 7: 8:30 AM - 10:20 AM
5143 
Contributed Papers 
Oregon Convention Center 
Room: CC-E142 

Main Sponsor

Section on Statistics in Imaging

Presentations

A CAIPI Approach to Decrease Geometry Factor for Simultaneous Multi-Slice Technique in FMRI

FMRI has been a powerful and safe medical imaging tool to study the function of the brain by demonstrating the spatial and temporal changes in brain metabolism in recent decades. To capture brain functionality more efficiently, efforts have been made to accelerate the number of images acquired per unit of time that create each volume image without losing full anatomical structure. The Simultaneous Multi-Slice (SMS) technique provides an alternative reconstruction method where multiple slices are acquired and aliased concurrently. Traditional imaging techniques such as SENSE and GRAPPA can reconstruct an image from less measured data but have their drawbacks. Controlled Aliasing in Parallel Imaging (CAIPI) shifts is a technique where the field-of-view is shifted during image acquisition. We present a novel SMS technique called mSPECS-CAIPI with Through-Plane and In-Plane Acceleration. It combines the image shift method of CAIPI, the CAIPI with view angle tilting technique, and Hadamard phase-encoding. Our proposed approach was applied to a simulation study with preliminary results showing a decrease in the influence of the geometry factor while increasing brain activation detection. 

Keywords

fMRI

SMS

Through-Plane

In-Plane

CAIPI 

View Abstract 3727

Co-Author

Daniel Rowe, Marquette University

First Author

Ke Xu

Presenting Author

Ke Xu

A Two-Stage Approach for Segmenting Spatial Point Patterns Applied to Tumor Immunology

In tumor immunology, clinical regimes corresponding to different stages of disease or responses to treatment may manifest as different spatial arrangements of tumor and immune cells. Spatial point pattern (SPP) modeling can be used to segment tissue images according to these regimes. To this end, we propose a two-stage approach: first, local intensities and pair correlation functions (PCF) are estimated from the SPP of cells within each image, and the PCFs are reduced in dimension via spectral decomposition of the covariance function. Second, the estimates are clustered in a Bayesian hierarchical model with spatially-dependent cluster labels. The clusters correspond to regimes of interest that are present across subjects; the cluster labels segment the images according to those regimes. Through Markov Chain Monte Carlo (MCMC) sampling, we jointly estimate and quantify uncertainty in the cluster assignment and spatial characteristics of each cluster. The number of clusters is found through cross-validation. Simulations demonstrate the performance of the method, and it is applied to a set of multiplex immunofluorescence images of pancreatic tissue. 

Keywords

Bayesian mixture model

Biomedical imaging

Functional data analysis

Tumor immunology

MCMC

Spatial point pattern 

View Abstract 3238

Co-Author(s)

Brian Reich, North Carolina State University
Ana-Maria Staicu, North Carolina State University

First Author

Alvin Sheng, North Carolina State University

Presenting Author

Alvin Sheng, North Carolina State University

A Wasserstein-type Distance for Gaussian Mixtures on Vector Bundles with Applications to Shape Analysis

This paper uses sample data to study the problem of comparing populations on finite-dimensional parallelizable Riemannian manifolds and more general trivial vector bundles. Utilizing triviality, our framework represents populations as mixtures of Gaussians on vector bundles and estimates the population parameters using a mode-based clustering algorithm. We derive a Wasserstein-type metric between Gaussian mixtures, adapted to the manifold geometry, in order to compare estimated distributions. Our contributions include an identifiability result for Gaussian mixtures on manifold domains and a convenient characterization of optimal couplings of Gaussian mixtures under the derived metric. We demonstrate these tools on some example domains, including the pre-shape space of planar closed curves, with applications to the shape space of triangles and populations of nanoparticles. In the nanoparticle application, we consider a sequence of populations of particle shapes arising from a manufacturing process, and utilize the Wasserstein-type distance to perform change-point detection.  

Keywords

Optimal Transport

Differential Geometry

Statistical Shape Analysis 

Abstracts


Co-Author(s)

Michael Wilson
Tom Needham, Florida State Board of Administration
Suprateek Kundu, MD Anderson
Chiwoo Park, Florida A&M - Florida State University College of Engineering
Anuj Srivastava, Florida State University

First Author

Michael Wilson

Presenting Author

Michael Wilson

Automated Generation of Building Footprints in Satellite Imagery using Artificial Intelligence

A deep learning algorithm is implemented to identify building footprints in satellite or arial imagery. A Mask Region-Based Convolutional Neural Network (Mask R-CNN) was trained and tested for detection of residential buildings. Annotated dataset for training was generated by sketching bounding boxes across the buildings. Resnet50 was used as the backbone for transfer learning in the model for detection of building footprints. The dataset for re-training and fine-tuning of the transfer network was the 2024 Bay County building footprints for annotation of the 2022 NAIP images taken over Panama City, Florida. The training data was created by making 2851 image chips of size 256 x 256 from the original image using the Bay County footprints as the labels. The learning rate for the model was 1.0000e-04 with an Average Precision Score of 0.72. It was then tested on images in Mexico Beach and other areas of Panama City for a final test and the results were promising. The unregularized footprints show good agreement with the current image used for inference. Looking at the confidence in the predictions, the lowest level recorded in the attribute table was 0.93. 

Keywords

GIS

Artificial Intelligence

Satellite Imagery

Building Footprint

Object Detection

Georeferencing 

View Abstract 3741

Co-Author

Edmund Robbins, Florida Institute of Technology

First Author

Nezamoddin N Kachouie, Florida Institute of Technology-Department of Mathematical Sciences

Presenting Author

Nezamoddin N Kachouie, Florida Institute of Technology-Department of Mathematical Sciences

Exploit Spatially Transcriptomic Data to Infer Cellular Features from Pathology Imaging Data

Digital pathology is a fast-growing field where leveraging machine learning methods uncovers meaningful imaging features, but it faces the hurdles of sparse annotations, particularly for small pathology image segments. We propose a novel approach that employs spatially resolved transcriptomic data for annotations, though it faces challenges like annotation uncertainty from transcriptomic data and inconsistent image resolutions. We established the viability of this approach and developed a novel transfer-learning neural network model, STpath (Spatial Transcriptomics and pathology images), designed to predict cell type proportions or classify tumor microenvironments. Our findings revealed the intrinsic link between pre-trained deep learning features and cell identities in pathology image segments. Tested on two breast cancer datasets, STpath demonstrated robust performance, effectively handling samples with diverse cell type proportions and high-resolution images despite limited training data. As the influx of spatially resolved transcriptomic data continues, we foresee ongoing updates to STpath, shaping it into an invaluable AI tool for pathologists, enhancing diagnostic accuracy. 

Keywords

Digital and computational pathology

Whole slide imaging

Spatially resolved transcriptomics

Artificial Intelligence

Transfer learning

Cell type proportion and tumor microenvironment 

View Abstract 2689

First Author

Zhining Sui

Presenting Author

Zhining Sui

Improved Extrinsic Energy Tests for Projective Shape Analysis

We present improved extrinsic energy tests for projective shape analyses. Our methods make use of novel embeddings for projective shape space and their tangential components data in place of the embedded sample data such as was previously considered in Guo, et al. (2023). We compare our novel methodologies with those based on embedded sample data via simulation studies and consider a real-life application of our methodology. 

Keywords

extrinsic energy tests

projective shape analysis

tangential components 

View Abstract 2690

Co-Author

Victor Patrangenaru, Florida State University

First Author

Robert Paige, Missouri S&T

Presenting Author

Robert Paige, Missouri S&T

Spatial characteristics of Hyperpolarized Xe-129 apparent diffusion coefficient of CF patients.

Microstructural lung damage is the pivotal cause of morbidity and mortality in patients with Cystic fibrosis (CF), a life-shortening genetic disorder. Detection of such lung abnormalities by examining the microstructural changes from Hyperpolarized Xe-129 MRI, are often done by calculating parameters such as apparent diffusion coefficient (ADC). We examined the spatial distribution of ADC parameter to explain the microstructural differences between healthy controls and CF patients.

Using DW-MRI images from 38 healthy controls (age: 17.2 ±9.5 year) and 39 CF patients (age: 15.28±7.62 year), we compared the spatial characteristics of 129Xe ADC in CF patients to healthy controls. Upon doing this, two random effects for the spatial and nonspatial variations in ADC maps were introduced. The prior distribution for the spatial random effect was specified using the conditional autoregressive model. This helps in understanding the microstructural changes in the lungs of CF patients by looking at the ADC map and the voxel level spatial map along with patient level spatial autocorrelation in comparison to healthy controls. The findings will be presented in the meeting. 

Keywords

Conditional autoregressive model

Cystic Fibrosis

apparent diffusion coefficient

MRI images 

View Abstract 2484

Co-Author(s)

A. Bdaiwi, CPIR, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States
Marepalli Rao, University of Cincinnati
Z.I. Cleveland, CPIR, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States
Md. M. Hossain, Cincinnati Children Hospital medical center

First Author

Neelakshi Chatterjee

Presenting Author

Neelakshi Chatterjee