Bioimaging and Biomedical Optics Commons^™

Robustness Of Ai-Driven Histopathology Under Real-World Adversarial Examples, Ruchik N. Yajnik

Theses

This robustness of histopathology classification models under adversarial and real-world perturbations resembling clinical artifacts is being investigated.

Using whole-slide images from the CAMELYON17 cohort, four representative architectures—ResNet-18, ResNet-50, HIPT-2MLP, and ViT-B/16 —are benchmarked across controlled pixel-level distortions and artifact-like transformations. Adversarial methods include iterative Fast Gradient Sign, Projected Gradient Descent, Salt-and-Pepper noise, and the Adversarial Watermark—Stain Shift (AWSS). Three defense strategies—Randomized Smoothing, Adversarial Training, and an Artifact Detector—are evaluated for their ability to preserve diagnostic accuracy and model reliability. Structured perturbations consistently degrade performance, with transformer-based models showing the greatest sensitivity. The benchmark developed here offers a reproducible framework for …

Untrained Position-Encoded Multilayer Perceptron Network For Structured Illumination Microscopy Reconstruction, Sahil Sharma, Leonidas Zimianitis, Krishnendu Samanta, Balpreet Singh Ahluwalia, Joby Joseph, Dushan N. Wadduwage

Computer Science Faculty Publications

Structured Illumination Microscopy (SIM) enables super-resolution imaging by encoding high-frequency spatial information through patterned light. While traditional Fourier-based reconstruction methods are prone to artifacts under suboptimal conditions, recent deep learning approaches often require large training datasets and lack adaptability across different imaging setups. In this work, we present Position Encoded Multi-Layer Perceptron (PEM) network that leverages implicit neural representations (INRs) and SIM forward-model-driven modeling to reconstruct super-resolved images without any training data. PEM-SIM represents each spatial coordinate as a combination of sinusoidal functions across multiple frequencies, enabling rich encoding of fine spatial detail. A forward model grounded in SIM image …

Precision Medicine Assessment Of The Radiographic Defect Angle Of The Intrabony Defect In Periodontal Lesions By Deep Learning Of Bitewing Radiographs, Patricia Angela R. Abu, Yi Cheng Mao, Yuan Jin Lin, Chien Kai Chao, Yi He Lin, Bo Siang Wang, Chiung An Chen, Shih Lun Chen, Tsung Yi Chen, Kuo Chen Li

Department of Information Systems & Computer Science Faculty Publications

In dental diagnosis, evaluating the severity of periodontal disease by analyzing the radiographic defect angle of the intrabony defect is essential for effective treatment planning. However, dentists often rely on clinical examinations and manual analysis, which can be time-consuming and labor-intensive. Due to the high recurrence rate of periodontal disease after treatment, accurately evaluating the radiographic defect angle of the intrabony defect is vital for implementing targeted interventions, which can improve treatment outcomes and reduce recurrence. This study aims to streamline clinical practices and enhance patient care in managing periodontal disease by determining its severity based on the analysis of …

Automated Segmentation Of Autofluorescence Lifetime Images Using A Convolutional Neural Network, Nicole Hooten

Biomedical Engineering Undergraduate Honors Theses

Chronic wounds affect nearly 1 out of every 50 people in the United States, decreasing quality of life and putting people with potential comorbidities at high risk for obtaining an infection. Wound healing progress in clinical settings is measured by tracking wound size and there are currently no non-invasive, quantitative measurement techniques. Previous studies have proposed nicotinamide adenine dinucleotide (NADH) autofluorescence lifetime (FLIM) imaging of the wound edge as a method to quantify the wound healing process by connecting the role of NADH in cellular metabolism to wound healing stage. However, evaluation of FLIM images is heavily subjective, as a …

Efficient Scopeformer: Towards Scalable And Rich Feature Extraction For Intracranial Hemorrhage Detection Using Hybrid Convolution And Vision Transformer Networks, Yassine Barhoumi

Theses and Dissertations

The field of medical imaging has seen significant advancements through the use of artificial intelligence (AI) techniques. The success of deep learning models in this area has led to the need for further research. This study aims to explore the use of various deep learning algorithms and emerging modeling techniques to improve training paradigms in medical imaging. Convolutional neural networks (CNNs) are the go-to architecture for computer vision problems, but they have limitations in mapping long-term dependencies within images. To address these limitations, the study explores the use of techniques such as global average pooling and self-attention mechanisms. Additionally, the …

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 …

Identification Of Microscopy Cell Images By Using Convolutional Neural Network Application, Ajis Norfatin Farisya

Student Works (2020-2029)

Breast cancer has been the major factor of cancer death and the second main cause of women’s deaths in the world. The false positive results of this cancer cell detection during the screening test leads to false treatment and emotional disturbance of the patients. Thus, breast cancer cell lines (MCF7) is used as the microscopy image samples together with the Human Bone Osteosarcoma Epithelial Cells (U2OS), and Human Hepatocyte as control to study the effectiveness of convolutional neural network (CNN) as a method of image recognition. The objectives of this study are to determine the ability of convolutional neural network …

Neurobiological Markers For Remission And Persistence Of Childhood Attention-Deficit/Hyperactivity Disorder, Yuyang Luo

Dissertations

Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent neurodevelopmental disorders in children. Symptoms of childhood ADHD persist into adulthood in around 65% of patients, which elevates the risk for a number of adverse outcomes, resulting in substantial individual and societal burden. A neurodevelopmental double dissociation model is proposed based on existing studies in which the early onset of childhood ADHD is suggested to associate with dysfunctional subcortical structures that remain static throughout the lifetime; while diminution of symptoms over development could link to optimal development of prefrontal cortex. Current existing studies only assess basic measures including regional brain activation …

Seeing Eye To Eye: A Machine Learning Approach To Automated Saccade Analysis, Maigh Attre

Honors Scholar Theses

Abnormal ocular motility is a common manifestation of many underlying pathologies particularly those that are neurological. Dynamics of saccades, when the eye rapidly changes its point of fixation, have been characterized for many neurological disorders including concussions, traumatic brain injuries (TBI), and Parkinson’s disease. However, widespread saccade analysis for diagnostic and research purposes requires the recognition of certain eye movement parameters. Key information such as velocity and duration must be determined from data based on a wide set of patients’ characteristics that may range in eye shapes and iris, hair and skin pigmentation [36]. Previous work on saccade analysis has …

Applications Of Machine Learning In Nuclear Imaging And Radiation Detection, Shaikat Mahmood Galib

Doctoral Dissertations

"The main focus of this work is to use machine learning and data mining techniques to address some challenging problems that arise from nuclear data. Specifically, two problem areas are discussed: nuclear imaging and radiation detection. The techniques to approach these problems are primarily based on a variant of Artificial Neural Network (ANN) called Convolutional Neural Network (CNN), which is one of the most popular forms of 'deep learning' technique.

The first problem is about interpreting and analyzing 3D medical radiation images automatically. A method is developed to identify and quantify deformable image registration (DIR) errors from lung CT scans …

Early Emerging Pathogen Detection, Mackenzie Wangenstein

Undergraduate Research & Mentoring Program

A supervised learning technique was employed to identify emerging pathogen species. Portland State University has partnered with the University of New Mexico to take encodings of unknown pathogen molecular structures to determine emerging species.

Pattern Discovery In Brain Imaging Genetics Via Scca Modeling With A Generic Non-Convex Penalty, Lei Du, Kefei Liu, Xiaohui Yao, Jingwen Yan, Shannon L. Risacher, Junwei Han, Lei Guo, Andrew J. Saykin, Li Shen, Michael W. Weiner, Paul Aisen, Ronald Petersen, Clifford R. Jack, William Jagust, John Q. Trojanowki, Arthur W. Toga, Laurel Beckett, Robert C. Green, John Morris, Leslie M. Shaw, Zaven Khachaturian, Greg Sorensen, Maria Carrillo, Lew Kuller, Marc Raichle, Steven Paul, Peter Davies, Howard Fillit, Franz Hefti, David Holtzman, Charles D. Smith, Gregory Jicha, Peter A. Hardy, Partha Sinha, Elizabeth Oates, Gary Conrad

Neurology Faculty Publications

Brain imaging genetics intends to uncover associations between genetic markers and neuroimaging quantitative traits. Sparse canonical correlation analysis (SCCA) can discover bi-multivariate associations and select relevant features, and is becoming popular in imaging genetic studies. The L1-norm function is not only convex, but also singular at the origin, which is a necessary condition for sparsity. Thus most SCCA methods impose ℓ₁-norm onto the individual feature or the structure level of features to pursuit corresponding sparsity. However, the ℓ₁-norm penalty over-penalizes large coefficients and may incurs estimation bias. A number of non-convex penalties are proposed to reduce …

Machine Learning Methods For Medical And Biological Image Computing, Rongjian Li

Computer Science Theses & Dissertations

Medical and biological imaging technologies provide valuable visualization information of structure and function for an organ from the level of individual molecules to the whole object. Brain is the most complex organ in body, and it increasingly attracts intense research attentions with the rapid development of medical and bio-logical imaging technologies. A massive amount of high-dimensional brain imaging data being generated makes the design of computational methods for efficient analysis on those images highly demanded. The current study of computational methods using hand-crafted features does not scale with the increasing number of brain images, hindering the pace of scientific discoveries …

Prediction Of Brain Tumor Progression Using Multiple Histogram Matched Mri Scans, Debrup Banerjee, Loc Tran, Jiang Li, Yuzhong Shen, Frederic Mckenzie, Jihong Wang, Ronald M. Summers (Ed.), Bram Van Ginneken (Ed.)

Electrical & Computer Engineering Faculty Publications

In a recent study [1], we investigated the feasibility of predicting brain tumor progression based on multiple MRI series and we tested our methods on seven patients' MRI images scanned at three consecutive visits A, B and C. Experimental results showed that it is feasible to predict tumor progression from visit A to visit C using a model trained by the information from visit A to visit B. However, the trained model failed when we tried to predict tumor progression from visit B to visit C, though it is clinically more important. Upon a closer look at the MRI scans …