Deep Learning Approach To Improved Image Quality For Medical Diagnostics, Olivia Loesch, Katie Leyba, Halyley Chan, Craig Goergen

Discovery Undergraduate Interdisciplinary Research Internship

The United Nation’s health-related Sustainable Development Goals are difficult to achieve in low- and middle-income countries due to workforce shortages and inadequate health surveillance systems. However, with the growth of artificial intelligence (AI) and computer algorithms, it is possible to apply AI to healthcare technologies to improve progress towards these UN standards. This project aims at using and improving computer algorithms and deep learning to aid in the extraction of important structural and functional information from murine carotid artery ultrasound and photoacoustic images. First, we created a large database of simulated photoacoustic images to optimize the algorithms. These images were …

Inter-Subject Correlation While Listening To Minimalist Music: A Study Of Electrophysiological And Behavioral Responses To Steve Reich’S Piano Phase, Tysen Dauer, Duc T. Nguyen, Nick Gang, Jacek P. Dmochowski, Jonathan Berger, Blair Kaneshiro

Publications and Research

Musical minimalism utilizes the temporal manipulation of restricted collections of rhythmic, melodic, and/or harmonic materials. One example, Steve Reich’s Piano Phase, offers listeners readily audible formal structure with unpredictable events at the local level. For example, pattern recurrences may generate strong expectations which are violated by small temporal and pitch deviations. A hyper-detailed listening strategy prompted by these minute deviations stands in contrast to the type of listening engagement typically cultivated around functional tonal Western music. Recent research has suggested that the inter-subject correlation (ISC) of electroencephalographic (EEG) responses to natural audio-visual stimuli objectively indexes a state of “engagement,” demonstrating …

Single-Molecule Localization Microscopy Of 3d Orientation And Anisotropic Wobble Using A Polarized Vortex Point Spread Function, Tianben Ding, Matthew D. Lew

Electrical & Systems Engineering Publications and Presentations

Within condensed matter, single fluorophores are sensitive probes of their chemical environments, but it is difficult to use their limited photon budget to image precisely their positions, 3D orientations, and rotational diffusion simultaneously. We demonstrate the polarized vortex point spread function (PSF) for measuring these parameters, including characterizing the anisotropy of a molecule’s wobble, simultaneously from a single image. Even when imaging dim emitters (∼500 photons detected), the polarized vortex PSF can obtain 12 nm localization precision, 4°–8° orientation precision, and 26° wobble precision. We use the vortex PSF to measure the emission anisotropy of fluorescent beads, the wobble dynamics …

Correlation Of Acute Radiation Dermatitis To Tissue Oxygenation In Radiation Therapy Treated Breast Cancer Subjects, Edwin Alexander Robledo

FIU Electronic Theses and Dissertations

Over 95% of radiation therapy (RT) treated breast cancer subjects undergo an adverse skin reaction known as radiation dermatitis (RD). Assessment of severity or grading of RD is clinically visual and hence subjective. Our objective is to determine sub-clinical tissue oxygenation (StO₂) changes in response to RT treatment in breast cancer subjects using near-infrared spectroscopic imaging and correlate these changes to RD grading. A WIRB approved 6-8 week longitudinal pilot study was carried out on 10 RT-treated subjects at Miami Cancer Institute. Significant changes (p < 0.05) in StO₂ of irradiated and contralateral chest wall and axilla regions with weeks of …

Dataset Of Concurrent Eeg, Ecg, And Behavior With Multiple Doses Of Transcranial Electrical Stimulation, Nigel Gebodh, Zeinab Esmaeilpour, Abhishek Datta, Marom Bikson

Publications and Research

We present a dataset combining human-participant high-density electroencephalography (EEG) with physiological and continuous behavioral metrics during transcranial electrical stimulation (tES). Data include within participant application of nine High-Definition tES (HD-tES) types, targeting three cortical regions (frontal, motor, parietal) with three stimulation waveforms (DC, 5 Hz, 30 Hz); more than 783 total stimulation trials over 62 sessions with EEG, physiological (ECG, EOG), and continuous behavioral vigilance/alertness metrics. Experiment 1 and 2 consisted of participants performing a continuous vigilance/alertness task over three 70-minute and two 70.5-minute sessions, respectively. Demographic data were collected, as well as self-reported wellness questionnaires before and after each …

Pairwise Correlation Analysis Of The Alzheimer’S Disease Neuroimaging Initiative (Adni) Dataset Reveals Significant Feature Correlation, Erik D. Huckvale, Matthew W. Hodgman, Brianna B. Greenwood, Devorah O. Stucki, Katrisa M. Ward, Mark T. W. Ebbert, John S. K. Kauwe, The Alzheimer’S Disease Neuroimaging Initiative, The Alzheimer’S Disease Metabolomics Consortium, Justin B. Miller

Sanders-Brown Center on Aging Faculty Publications

The Alzheimer’s Disease Neuroimaging Initiative (ADNI) contains extensive patient measurements (e.g., magnetic resonance imaging [MRI], biometrics, RNA expression, etc.) from Alzheimer’s disease (AD) cases and controls that have recently been used by machine learning algorithms to evaluate AD onset and progression. While using a variety of biomarkers is essential to AD research, highly correlated input features can significantly decrease machine learning model generalizability and performance. Additionally, redundant features unnecessarily increase computational time and resources necessary to train predictive models. Therefore, we used 49,288 biomarkers and 793,600 extracted MRI features to assess feature correlation within the ADNI dataset to determine the …

Artificial Image Objects For Classification Of Breast Cancer Biomarkers With Transcriptome Sequencing Data And Convolutional Neural Network Algorithms, Xiangning Chen, Daniel G. Chen, Zhongming Zhao, Justin M. Balko, Jingchun Chen

School of Medicine Faculty Publications

Background: Transcriptome sequencing has been broadly available in clinical studies. However, it remains a challenge to utilize these data effectively for clinical applications due to the high dimension of the data and the highly correlated expression between individual genes. Methods: We proposed a method to transform RNA sequencing data into artificial image objects (AIOs) and applied convolutional neural network (CNN) algorithms to classify these AIOs. With the AIO technique, we considered each gene as a pixel in an image and its expression level as pixel intensity. Using the GSE96058 (n = 2976), GSE81538 (n = 405), and GSE163882 (n = …

Association Of X-Ray Absorptiometry Body Composition Measurements With Basic Anthropometrics And Mortality Hazard, Nir Y. Krakauer, Jesse C. Krakauer

Publications and Research

Dual-energy X-ray absorptiometry (DEXA) is a non-invasive imaging modality that can estimate whole-body and regional composition in terms of fat, lean, and bone mass. We examined the ability of DEXA body composition measures (whole-body, trunk, and limb fat mass and fat-free mass) to predict mortality in conjunction with basic body measures (anthropometrics), expressed using body mass index (BMI) and a body shape index (ABSI). We used data from the 1999–2006 United States National Health and Nutrition Examination Survey (NHANES), with mortality follow-up to 2015. We found that all DEXA-measured masses were highly correlated with each other and with ABSI and …

Pathcnn: Interpretable Convolutional Neural Networks For Survival Prediction And Pathway Analysis Applied To Glioblastoma, Jung Hun Oh, Wookjin Choi, Euiseong Ko, Mingon Kang, Allen Tannenbaum, Joseph O. Deasy

Computer Science Faculty Research

Motivation: Convolutional neural networks (CNNs) have achieved great success in the areas of image processing and computer vision, handling grid-structured inputs and efficiently capturing local dependencies through multiple levels of abstraction. However, a lack of interpretability remains a key barrier to the adoption of deep neural networks, particularly in predictive modeling of disease outcomes. Moreover, because biological array data are generally represented in a non-grid structured format, CNNs cannot be applied directly. Results: To address these issues, we propose a novel method, called PathCNN, that constructs an interpretable CNN model on integrated multi-omics data using a newly defined pathway image. …

Rapid Microscopic Fractional Anisotropy Imaging Via An Optimized Linear Regression Formulation., N J J Arezza, D H Y Tse, C A Baron

Medical Biophysics Publications

Water diffusion anisotropy in the human brain is affected by disease, trauma, and development. Microscopic fractional anisotropy (μFA) is a diffusion MRI (dMRI) metric that can quantify water diffusion anisotropy independent of neuron fiber orientation dispersion. However, there are several different techniques to estimate μFA and few have demonstrated full brain imaging capabilities within clinically viable scan times and resolutions. Here, we present an optimized spherical tensor encoding (STE) technique to acquire μFA directly from the 2nd order cumulant expansion of the powder averaged dMRI signal obtained from direct linear regression (i.e. diffusion kurtosis) which requires fewer powder-averaged signals than …

Water Exchange Rate Across The Blood-Brain Barrier Is Associated With Csf Amyloid-Β 42 In Healthy Older Adults, Brian T. Gold, Xingfeng Shao, Tiffany L. Sudduth, Gregory A. Jicha, Donna M. Wilcock, Elayna R. Seago, Danny J. J. Wang

Sanders-Brown Center on Aging Faculty Publications

INTRODUCTION: We tested if water exchange across the blood-brain barrier (BBB), estimated with a noninvasive magnetic resonance imaging (MRI) technique, is associated with cerebrospinal fluid (CSF) biomarkers of Alzheimer's disease (AD) and neuropsychological function.

METHODS: Forty cognitively normal older adults (67–86 years old) were scanned with diffusion‐prepared, arterial spin labeling (DP‐ASL), which estimates water exchange rate across the BBB (k_w). Participants also underwent CSF draw and neuropsychological testing. Multiple linear regression models were run with kw as a predictor of CSF concentrations and neuropsychological scores.

RESULTS: In multiple brain regions, BBB kw was positively associated with CSF amyloid …

Biomedical Applications And Syntheses Of Selected Anthraquinone Dyes, Richard Sirard

Senior Honors Theses

Anthraquinones are aromatic organic compounds that have multiple applications in the biomedical field. Some anthraquinone-based compounds are used as fluorophores to contrast cell nuclei while others act as chemotherapeutic agents. However, there are not many fluorescent anthraquinone cell stains currently available. In this study, commercially available anthraquinone dyes, in addition to other dye families and compounds, were reviewed for their unique properties, advantages, and drawbacks. The development and characterization of three novel anthraquinone fluorophores revealed promising photophysical characteristics, like large Stokes shifts. One of the compounds, RBS3, was chosen for fixed and live cell staining and exhibited desirable biomedical properties. …

A Brief Bibliometric Survey Of Explainable Ai In Medical Field, Nilkanth Mukund Deshpande, Shilpa Shailesh Gite

Library Philosophy and Practice (e-journal)

_{^{Background: This study aims to analyze the work done in the field of explainability related to artificial intelligence, especially in the medical field from 2004 onwards using the bibliometric methods.}}

_{^{Methods: different articles based on the topic leukemia detection were retrieved using one of the most popular database- Scopus. The articles are considered from 2004 onwards. Scopus analyzer is used for different types of analysis including documents by year, source, county and so on. There are other different analysis tools such as VOSviewer Version 1.6.15. This is used for the analysis of different units such as co-authorship, co-occurrences, citation analysis …}}

Preclinical Development Of Single Walled Carbon Nanotube-Based Optical Biosensors, Eric M. Hofferber

Department of Agricultural and Biological Systems Engineering: Dissertations, Theses, and Student Research

High resolution, long-term monitoring of key biological analytes would improve patient outcomes by providing earlier detection of disease states and improved efficacy of treatment. One class of biosensors that have gained much attention in recent years are optical biosensors. Optical probes are attractive biosensors due to their noninvasive nature of detection, as certain light can pass through tissue, water, and blood. Single walled carbon nanotubes (SWNT) are a specific type of optical biosensor that fluoresce in the near infrared range of the electromagnetic spectrum and offer unparalleled spatial and temporal resolution. SWNT have been applied as biosensors in vitro, …

Nanoanalytical Analysis Of Bisphosphonate-Driven Alterations Of Microcalcifications Using A 3d Hydrogel System And In Vivo Mouse Model, Jessica L. Ruiz, Joshua D. Hutcheson, Luis Cardoso, Amirala Bakhshian Nik, Alexandra Condado De Abreu, Tan Pham, Fabrizio Buffolo, Sara Busatto, Stefania Frederici, Andrea Ridolfi, Masanori Aikawa, Sergio Bertazzo, Paolo Bergese, Sheldon Weinbaum, Elena Aikawa

Publications and Research

Vascular calcification predicts atherosclerotic plaque rupture and cardiovascular events. Retrospective studies of women taking bisphosphonates (BiPs), a proposed therapy for vascular calcification, showed that BiPs paradoxically increased morbidity in patients with prior acute cardiovascular events but decreased mortality in event-free patients. Calcifying extracellular vesicles (EVs), released by cells within atherosclerotic plaques, aggregate and nucleate calcification. We hypothesized that BiPs block EV aggregation and modify existing mineral growth, potentially altering microcalcification morphology and the risk of plaque rupture. Three-dimensional (3D) collagen hydrogels incubated with calcifying EVs were used to mimic fibrous cap calcification in vitro, while an ApoE−/− mouse was used …

Penta-Modal Imaging Platform With Oct- Guided Dynamic Focusing For Simultaneous Multimodal Imaging, Arash Dadkhah

FIU Electronic Theses and Dissertations

Complex diseases, such as Alzheimer’s disease, are associated with sequences of changes in multiple disease-specific biomarkers. These biomarkers may show dynamic changes at specific stages of disease progression. Thus, testing/monitoring each biomarker may provide insight into specific disease-related processes, which can result in early diagnosis or even development of preventive measures. Obtaining a comprehensive information of biological tissues requires imaging of multiple optical contrasts, which is not typically offered by a single imaging modality. Thus, combining different contrast mechanisms to achieve simultaneous multimodal imaging is desirable. However, this process is highly challenging due to specific optical and hardware requirements for …

Development Of All-Optical Quantitative Ultrasound Imaging System, Mohamed Abdulrahman Almadi

FIU Electronic Theses and Dissertations

Ultrasound (US) is a well-established deep-tissue imaging modality in biomedicine. It distinguishes different tissue types based on their echogenicity, but this approach provides limited diagnostic sensitivity and accuracy. The majority of the US transducers nowadays rely on lead zirconate titanate (PZT) ceramic elements to transmit and receive ultrasound. Unfortunately, significant limitations arise from these transducers due to their frequency characteristics and complex fabrication process. A recently introduced technique, Quantitative Ultrasound (QUS) Measurement, shows a great promise to improve US-based tissue diagnosis, but it requires a transducer with a large spectrum bandwidth, which is a feature not available in PZT transducers. …

Computational Modelling Enables Robust Multidimensional Nanoscopy, Matthew D. Lew

Electrical & Systems Engineering Publications and Presentations

The following sections are included:

• Present State of Computational Modelling in Fluorescence Nanoscopy

• Recent Contributions to Computational Modelling in Fluorescence Nanoscopy

• Outlook on Computational Modelling in Fluorescence Nanoscopy

• Acknowledgments

• References

Fmri Feature Extraction Model For Adhd Classification Using Convolutional Neural Network, Senuri De Silva, Sanuwani Udara Dayarathna, Gangani Ariyarathne, Dulani Meedeniya, Sampath Jayarathna

Computer Science Faculty Publications

Biomedical intelligence provides a predictive mechanism for the automatic diagnosis of diseases and disorders. With the advancements of computational biology, neuroimaging techniques have been used extensively in clinical data analysis. Attention deficit hyperactivity disorder (ADHD) is a psychiatric disorder, with the symptomology of inattention, impulsivity, and hyperactivity, in which early diagnosis is crucial to prevent unwelcome outcomes. This study addresses ADHD identification using functional magnetic resonance imaging (fMRI) data for the resting state brain by evaluating multiple feature extraction methods. The features of seed-based correlation (SBC), fractional amplitude of low-frequency fluctuation (fALFF), and regional homogeneity (ReHo) are comparatively applied to …

Characterization & Calibration Of Foresight Ice, Hareem Nisar, Terry M Peters, Elvis C.S. Chen

Robarts Imaging Publications

No abstract provided.

Fast And Accurate Autofocus Control Using Guassian Standard Deviation And Gradient-Based Binning, Peter Dimeo, Lu Sun, Xian Du

Mechanical and Industrial Engineering Faculty Publication Series

We propose a fast and accurate autofbcus algorithm using Gaussian standard deviation and gradient-based binning. Rather than iteratively searching for the optimal focus using an optimization process, the proposed algorithm directly calculates the mean of the Gaussian shaped focus measure (FM) curve to find the optimal focus location and uses the FM curve standard deviation to adapt the motion step size. The calculation only requires 3-4 defocused images to identify the center location of the FM curve. Furthermore, by assigning motion step sizes based on the FM curve standard deviation, the magnitude of the motion step is adaptively controlled according …

Combining Cryo-Em Density Map And Residue Contact For Protein Secondary Structure Topologies, Maytha Alshammari, Jing He

Computer Science Faculty Publications

Although atomic structures have been determined directly from cryo-EM density maps with high resolutions, current structure determination methods for medium resolution (5 to 10 Å) cryo-EM maps are limited by the availability of structure templates. Secondary structure traces are lines detected from a cryo-EM density map for α-helices and β-strands of a protein. A topology of secondary structures defines the mapping between a set of sequence segments and a set of traces of secondary structures in three-dimensional space. In order to enhance accuracy in ranking secondary structure topologies, we explored a method that combines three sources of information: a set …

Bibliometric Review On Applications Of Disease Detection Using Digital Image Processing Techniques, Jayant Jagtap, Rahil Sharma, Aryan Sinha, Nikhil Panda, Amulya Reddy

Library Philosophy and Practice (e-journal)

Advances around the field of deep learning and cognitive computing have allowed mankind to look and solve the problems of the world in a completely new way. Deep learning has been making huge advancements in the field of healthcare, which most importantly focuses upon disease detection and disease prediction. Techniques such as these have been conceptualized the idea of early detection and economical ways of treating the predicted disease in particular. Still, it has been observed that there seems to be no change in the way diagnosis of a particular disease takes place even in the 21^st generation of …

Bibliometric Review On Liver And Tumour Segmentation Using Deep Learning, Jayant Jagtap, Aamir Habeeb, Avinash Jha, Shrey Aggarwal, Khushi Gupta

Library Philosophy and Practice (e-journal)

One of the major organs in the body is liver where tumors occur often. Malignant liver tumors pose a serious hazard to human life and health. Manual segmentation of the liver organ and tumor from computed tomography (CT) scans is difficult, time-consuming, and skewed to the clinician's experience, yet it is essential for hepatic surgical planning. However, due to the following considerations, segmenting liver tumors from computed tomography (CT) images is difficult: In CT pictures, the contrast between the liver tumor and healthy tissues is low, and the boundary is indistinct; the picture of the liver tumor is confusing, with …