Doctor of Philosophy (PhD), Ohio University, 2018, Electrical Engineering & Computer Science (Engineering and Technology)

Changes in the morphology of a skin lesion is indicative of melanoma, a deadly type of skin cancer. This dissertation proposes a temporal analysis method to monitor the vascularity, pigmentation, size and other critical morphological attributes of the lesion. Digital images of a skin lesion acquired during follow-up imaging sessions are input to the proposed system. The images are pre-processed to normalize variations introduced over time. The vascularity is modelled as the skin images' red channel information and its changes by the Kullback-Leibler (KL) divergence of the probability density function approximation of histograms. The pigmentation is quantified as textural energy, changes in the energy and pigment coverage in the lesion. An optical flow field and divergence measure indicates the magnitude and direction of global changes in the lesion. Sub-surface change is predicted based on the surface skin lesion image with a novel approach. Changes in key morphological features such as lesions' shape, color, texture, size, and border regularity are computed. Future trends of the skin lesions features are estimated by an auto-regressive predictor. Finally, the features extracted using deep convolutional neural networks and the hand-crafted lesion features are compared with classification metrics. An accuracy of 80.5%, specificity of 98.14%, sensitivity of 76.9% with a deep learning neural network is achieved. Experimental results show the potential of the proposed method to monitor a skin lesion in real-time during routine skin exams.

Committee: Mehmet Celenk Ph.D. (Advisor); Savas Kaya Ph.D. (Committee Member); Jundong Liu Ph.D. (Committee Member); Razvan Bunescu Ph.D. (Committee Member); Xiaoping Shen Ph.D. (Committee Member); Sergio Lopez-Permouth Ph.D. (Committee Member) Subjects: Computer Science; Electrical Engineering; Medical Imaging; Oncology

Master of Science in Biomedical Engineering (MSBME), Wright State University, 2016, Biomedical Engineering

This thesis introduces the concept of implementing greyscale analysis, also known as intensity analysis, on endobronchial ultrasound (EBUS) images for the purposes of diagnosing peripheral lung tumors. The statistical methodology of using greyscale and histogram analysis allows the characterization of lung tissue in EBUS images. Regions of interest (ROI) will be analyzed in MATLAB and a feature vector will be created. A feature vector of first-order, second-order and histogram greyscale analysis will be created and used for the classification of malignant vs benign peripheral lung tumors. The tools that were implemented were MedCalc for the initial statistical analysis of receiver operating curves (ROC), Multiple Regression and MATLAB for the machine learning and ROI collection. Feature analysis, multiple regression and machine learning methods were used to better classify the malignant and benign EBUS images. The classification is assessed with a confusion matrix, ROC curve, accuracy, sensitivity and specificity. It was found that minimum pixel value, contrast and energy are the best determining factors to discriminate between benign and malignant EBUS images.

Committee: Ulas Sunar Ph.D. (Advisor); Jason Parker Ph.D. (Committee Member); Jaime Ramirez-Vick Ph.D. (Committee Member) Subjects: Biomedical Engineering; Biomedical Research; Biostatistics; Computer Engineering; Engineering; Health Care; Medical Imaging

Master of Science in Computer Engineering, University of Dayton, 2023, Electrical and Computer Engineering

Deep Convolutional Neural Networks demonstrate state-of-art performance in computer vision and medical image tasks. However, handling a large-scale image is still a challenging task that usually deals with resizing and patching methods to embed in the lower dimensional space. Recently, Learnable Resizer (LR) has been proposed to analyze large-scale images for computer vision tasks. This study proposes two DCNN models for classification and segmentation tasks constructed with LR in combination with successful classification and segmentation architectures. The performance of the proposed models is evaluated for the Diabetic Retinopathy (DR) analysis and skin cancer segmentation tasks. The proposed model demonstrated better performance than the existing methods for segmentation and classification tasks. For classification tasks, the proposed architectures achieved a 5.34% improvement in accuracy compared to ResNet50. Besides, around 0.62% accuracy over the base model and 0.28% in Intersection-over-Union (IoU) from state-of-the-art performance. The proposed model with the resizer network enhances the capability of the existing R2U-Net for medical image segmentation tasks. Moreover, the proposed methods enable a significant advantage in learning better with a few samples. The experimental results reveal that the proposed models are better than the current approaches.

Committee: Tarek M Taha (Committee Chair); Eric Balster (Committee Member); Chris Yakopcic (Committee Member) Subjects: Artificial Intelligence; Biomedical Research; Computer Engineering; Computer Science; Engineering; Medical Imaging

Master of Science in Biomedical Engineering, Cleveland State University, 2021, Washkewicz College of Engineering

The usage of radiomics for extracting high-dimensional features from radiographic imaging to quantify subtle changes in tissue structure and heterogeneity has shown great potential for disease diagnosis and prognosis. However, radiomic features are known to be impacted by acquisition-related changes (e.g., dose and reconstruction variations in CT scans) as well as technical variations between cohorts (i.e., batch effects due to varying dosage and tube currents). Using features which are not resilient to such imaging variations can result in poor performance of the downstream radiomics classifier models. In this study, we present a framework to systematically identify radiomic features that are resilient to both batch effects and acquisition differences, as well as evaluate the impact of such variations on radiomic model performance. We demonstrate the utility of our approach in the context of distinguishing active Crohn's disease (CD) from healthy controls using a uniquely accrued cohort of 164 CTE scans accrued from a single institution, which included (a) batch effects due to variations in effective dosage and tube current, as well as (b) scans simultaneously acquired at multiple doses and reconstructions (3 variations per patient). Our framework involves systematically evaluating the impact of acquisition variations (based on feature robustness to explicit dose/acquisition changes) and batch effects (based on feature stability to implicit dosage/current variations). Resilient radiomic features identified after accounting for both types of variations yielded the best random forest classifier performance across both discovery (AUC=0.819 ± 0.043) and validation (AUC=0.787) cohorts when using full-dose images; also found to be significantly more generalizable than features that were not optimized for such variations (AUC=0.419 in validation). This subset of radiomic features that were both robust and stable (resilient) also maintained their performance when evaluate (open full item for complete abstract)

Committee: Satish E. Viswanath (Committee Chair); Hongkai Yu (Committee Member); Moo-Yeal Lee (Advisor) Subjects: Biology; Biomedical Engineering; Biomedical Research; Medical Imaging; Radiology

Doctor of Philosophy (PhD), Ohio University, 2020, Computer Science (Engineering and Technology)

In recent years, fully convolutional networks (FCNs) have become the state-of-the-art solutions for various image segmentation tasks. Duo to the network setups, however, the standard FCNs tend to have issues of 1) overlooking small components; and 2) lacking intermediate supervision. To tackle the first issue in FCNs, we develop a two-stage network solution and apply it for white-matter lesion segmentation task. In the first stage, we design three networks with different input sizes and train them with patches from brain MR scans. In the second stage, we process large and small lesion separately, and use ensemble-nets to combine the segmentation results. A number of network setups, including activation functions, training strategy and ensemble paradigms have been explored to improve the segmentation accuracy measured by Dice Similarity Coefficient. The address the second issue, we adopt the notions of network residual and Laplacian pyramids to design a Laplacian module as the building block for new FCNs. The networks are applied to segment follicles and follicular cells. In order to separate individual follicle instances, we take advantage of the topological relationship between follicles and colloid, and use colloid masks as the guidance to identify individual labels. We also utilize Sobel edge maps as a guiding loss to ensure the smoothness of the segmentation results.

Committee: David Juedes (Committee Member); Razvan Bunescu (Committee Member); Chang Liu (Committee Member); Li Xu (Committee Member); Qiliang Wu (Committee Member); Jundong Liu (Advisor) Subjects: Artificial Intelligence

PhD, University of Cincinnati, 2018, Engineering and Applied Science: Aerospace Engineering

Interaction between a deformable elastic body and an internal or external fluid flow alters the flow pattern. This dissertation describes the effects of elasticity on flow in physiological scenarios. The first part of the thesis describes the influence of soft tissue compliance on flow in the upper airways of pediatric Down syndrome (DS) patients and adolescent Polycystic-Ovarian syndrome patients with obstructive sleep apnea (OSA). Computational fluid dynamics (CFD) of airflow is performed in pre and post-operative geometries of the DS pediatric airway to evaluate effectiveness of a surgery and address the importance of including the subject-specific tissue compliance. A tube law approach and a novel image analysis method are then presented to evaluate the circumferential variation in airway compliance for DS patients. An iterative finite element method is then described to non-invasively estimate patient-specific mechanical properties of the upper airway in these patients. The estimated mechanical properties for a single patient are applied to simulate airway obstruction during inspiratory airflow, before and after surgery. Sensitivity to different flow variables is analyzed and an operating map is created to establish the relationship between tissue elasticity and volumetric airflow. The necessity for performing fluid-structure interaction (FSI) in PCOS subjects with OSA is illustrated through a series of strain maps of upper airway tissue. An inverse methodology based on FSI simulations is described to characterize the soft-palate stiffness in these subjects. Differences in pre and post-operative airflow patterns and tissue motion in a PCOS patient are described using computational modeling and compared with the same for a healthy individual. The second part of the study describes computational FSI modeling of aortic blood flow in Turner syndrome (TS). A continuous measurement tool is developed to automatically compute the longitudinal variation in maximum aort (open full item for complete abstract)

Committee: Ephraim Gutmark Ph.D. (Committee Chair); Shaaban Abdallah Ph.D. (Committee Member); Iris Gutmark-Little (Committee Member); Mark Turner Sc.D. (Committee Member) Subjects: Aerospace Materials

Doctor of Philosophy, Case Western Reserve University, 2017, Applied Mathematics

Image segmentation and registration play active roles in machine vision and medical image analysis of historical data. Individually, the two has seen important research contributions, and the joint treatment of the two problems has become an active area of research. In this thesis we will explore the joint problem of segmenting and registering a template image given a reference image. We formulate the joint problem through an energy functional that integrates two well studied approaches in segmentation and registration: Geodesic Active Contours and nonlinear elastic registration. In the registration regime, the domain is modeled as a St. Venant-Kirchhoff material. We minimize the potential energy of this elastic system using variational methods, and derive an evolution equation which we solve using implicit-explicit integration methods. The numerical discretization of the problem allows us to take advantage of the Fast Fourier Transform. In the segmentation regime, we will adopt an active contours based energy with a weighted total variation penalty on the segmenting front. This particular choice allows for fast solution using the dual formulation of the total variation. The weight of the total variation penalty is an edge stopping function which depends on the deforming template. This allows the segmenting front to accurately track spontaneous changes in the shape of objects embedded in the template image as it deforms.

Committee: Weihong Guo (Committee Chair); Daniela Calvetti (Committee Member); Julia Dobrotsoskya (Committee Member); Erkki Somersalo (Committee Member); Michael Lewicki (Committee Member) Subjects: Applied Mathematics; Biomedical Engineering

MS, Kent State University, 2016, College of Arts and Sciences / Department of Computer Science

In biomedical imaging, segmentation and analysis play an important diagnostic role. Nuclei and nucleoli segmentation and classification have a significant impact on the cancer and tumor diagnostics in biological and medical research studies. Typically, segmentation is difficult in microscopic images because of object shapes and clustering in many samples. In this work we introduce a method that combines simplicity and efficiency. The proposed method utilizes ImageJ framework to automatically segment and classify nuclei and nucleoli after applying some preprocessing techniques to improve the image quality and remove noise. The required preprocessing steps differ based on the kind of segmentation required. Both 2D and 3D segmentation are achieved for the nuclei and nucleoli. The analysis approach provides statistics about volume, area, surface and other properties of the segmented nuclei and nucleoli. The classification process then groups the segmented nuclei and nucleoli based on the previous criteria. Finally, the visualization process shows the results of the proposed method overlaid on the original data set. The proposed method provides a very efficient system for nuclei and nucleoli segmentation and achieves about 98 % accuracy. Furthermore, the plugin is extremely fast when compared to manual segmentation especially with large data sets.

Committee: cheng chang Lu (Advisor); Robert Clements (Committee Member); Austin Melton (Committee Member) Subjects: Bioinformatics; Biomedical Research