View All Items

Title: INVESTIGATING AND IMPROVING FAIRNESS AND BIAS IN MACHINE LEARNING MODELS FOR DERMATOLOGY.
Creator: Corbin, Adam, Marques, Oge, Florida Atlantic University, Department of Computer and Electrical Engineering and Computer Science, College of Engineering and Computer Science
Abstract/Description: Advancements in Artificial Intelligence (AI) and Machine Learning (ML) have significantly improved their application in dermatology. However, bias issues in AI systems can result in missed diagnoses and disparities in healthcare, especially for individuals with different skin types. This dissertation aims to investigate and improve the fairness and bias in machine learning models for dermatology by evaluating and enhancing their performance across different Fitzpatrick skin types. The...
Show moreAdvancements in Artificial Intelligence (AI) and Machine Learning (ML) have significantly improved their application in dermatology. However, bias issues in AI systems can result in missed diagnoses and disparities in healthcare, especially for individuals with different skin types. This dissertation aims to investigate and improve the fairness and bias in machine learning models for dermatology by evaluating and enhancing their performance across different Fitzpatrick skin types. The technical contributions of the dissertation include generating metadata for Fitzpatrick Skin Type using Individual Typology Angle; outlining best practices for Explainable AI (XAI) and the use of colormaps; developing and enhancing ML models through skin color transformation and extending the models to include XAI methods for better interpretation and improvement of fairness and bias; and providing a list of steps for successful application of deep learning in medical image analysis. The research findings of this dissertation have the potential to contribute to the development of fair and unbiased AI/ML models in dermatology. This can ultimately lead to better health outcomes and reduced healthcare costs, particularly for individuals with different skin types.
Show less
Date Issued: 2023
PURL: http://purl.flvc.org/fau/fd/FA00014131
Subject Headings: Diagnostic Imaging, Machine learning, Dermatology, Artificial intelligence
Format: Document (PDF)

Title: FEDERATED LEARNING FOR MEDICAL IMAGE CLASSIFICATION.
Creator: Blazanovic, Danica, Zhu, Xingquan, Florida Atlantic University, Department of Computer and Electrical Engineering and Computer Science, College of Engineering and Computer Science
Abstract/Description: Machine learning (ML) has traditionally been used to make predictive models by training on local data. However, due to concerns regarding privacy, it is not always possible to collect and combine data from different sources. On the other hand, if there are insufficient data available, it might not be possible to construct accurate models to produce meaningful outcomes. This is where Federated Learning comes to the rescue. Federated Learning (FL) represents a sophisticated distributed machine...
Show moreMachine learning (ML) has traditionally been used to make predictive models by training on local data. However, due to concerns regarding privacy, it is not always possible to collect and combine data from different sources. On the other hand, if there are insufficient data available, it might not be possible to construct accurate models to produce meaningful outcomes. This is where Federated Learning comes to the rescue. Federated Learning (FL) represents a sophisticated distributed machine learning strategy that enables multiple devices hosted at different institutions such as hospitals, to collaboratively train a global model while ensuring that their respective data remains securely stored on-premises. It addresses privacy concerns and data protection regulations, because raw data does not need to be shared or centralized during the training process. This thesis research studies how two different FL architectures, centralized and decentralized FL, affect medical image classification. To study and validate the findings, skin cancer images dataset is used in a federated learning setting with five sites/clients, and a center for centralized FL. Experimental results show that using both centralized and decentralized (peer to peer) version of FL for classification of skin cancer images outperforms using the traditional ML. In addition, two different FL settings, centralized federated learning (CFL) and decentralized federated learning (DFL), are compared using different data distributions across sites/clients. Our study shows that the best accuracy (95.14%) was achieved with the DFL model when tested on the original dataset (without adding bias to the class distributions). This asserts that class distribution imbalance between sites has a significant impact to the federated learning.
Show less
Date Issued: 2023
PURL: http://purl.flvc.org/fau/fd/FA00014205
Subject Headings: Medical imaging, Diagnostic Imaging--classification, Machine learning
Format: Document (PDF)

Title: Technological abandonment: the lived experience of women having an abnormal prenatal ultrasound.
Creator: Gottlieb, Jeanne C., Graduate College
Date Issued: 2013-04-12
PURL: http://purl.flvc.org/fcla/dt/3361303
Subject Headings: Patient-Centered Care, Fetus--Abnormalities, Diagnostic ultrasonic imaging
Format: Document (PDF)

Title: Artificial Intelligence Based Electrical Impedance Tomography for Local Tissue.
Creator: Rao, Manasa, Pandya, Abhijit S., Florida Atlantic University, College of Engineering and Computer Science, Department of Computer and Electrical Engineering and Computer Science
Abstract/Description: This research aims at proposing the use of Electrical Impedance Tomography (EIT), a non-invasive technique that makes it possible to measure two or three dimensional impedance of living local tissue in a human body which is applied for medical diagnosis of diseases. In order to achieve this, electrodes are attached to the part of human body and an image of the conductivity or permittivity of living tissue is deduced from surface electrodes. In this thesis we have worked towards alleviating...
Show moreThis research aims at proposing the use of Electrical Impedance Tomography (EIT), a non-invasive technique that makes it possible to measure two or three dimensional impedance of living local tissue in a human body which is applied for medical diagnosis of diseases. In order to achieve this, electrodes are attached to the part of human body and an image of the conductivity or permittivity of living tissue is deduced from surface electrodes. In this thesis we have worked towards alleviating drawbacks of EIT such as estimating parameters by incorporating it in an electrode structure and determining a solution to spatial distribution of bio-impedance to a close proximity. We address the issue of initial parameter estimation and spatial resolution accuracy of an electrode structure by using an arrangement called "divided electrode" for measurement of bio-impedance in a cross section of a local tissue. Its capability is examined by computer simulations, where a distributed equivalent circuit is utilized as a model for the cross section tissue. Further, a novel hybrid model is derived which is a combination of artificial intelligence based gradient free optimization technique and numerical integration in order to estimate parameters. This arne! iorates the achievement of spatial resolution of equivalent circuit model to the closest accuracy.
Show less
Date Issued: 2008
PURL: http://purl.flvc.org/fau/fd/FA00012544
Subject Headings: Electrical impedance tomography, Diagnostic imaging--Data processing, Computational intelligence
Format: Document (PDF)

Title: AN ARTIFICIAL INTELLIGENCE DRIVEN FRAMEWORK FOR MEDICAL IMAGING.
Creator: Sanghvi, Harshal A., Agarwal, Ankur, Florida Atlantic University, Department of Computer and Electrical Engineering and Computer Science, College of Engineering and Computer Science
Abstract/Description: The major objective of this dissertation was to create a framework which is used for medical image diagnosis. In this diagnosis, we brought classification and diagnosing of diseases through an Artificial Intelligence based framework, including COVID, Pneumonia, and Melanoma cancer through medical images. The algorithm ran on multiple datasets. A model was developed which detected the medical images through changing hyper-parameters. The aim of this work was to apply the new transfer learning...
Show moreThe major objective of this dissertation was to create a framework which is used for medical image diagnosis. In this diagnosis, we brought classification and diagnosing of diseases through an Artificial Intelligence based framework, including COVID, Pneumonia, and Melanoma cancer through medical images. The algorithm ran on multiple datasets. A model was developed which detected the medical images through changing hyper-parameters. The aim of this work was to apply the new transfer learning framework DenseNet-201 for the diagnosis of the diseases and compare the results with the other deep learning models. The novelty in the proposed work was modifying the Dense Net 201 Algorithm, changing hyper parameters (source weights, Batch Size, Epochs, Architecture (number of neurons in hidden layer), learning rate and optimizer) to quantify the results. The novelty also included the training of the model by quantifying weights and in order to get more accuracy. During the data selection process, the data were cleaned, removing all the outliers. Data augmentation was used for the novel architecture to overcome overfitting and hence not producing false absurd results the computational performance was also observed. The proposed model results were also compared with the existing deep learning models and the algorithm was also tested on multiple datasets.
Show less
Date Issued: 2023
PURL: http://purl.flvc.org/fau/fd/FA00014274
Subject Headings: Diagnostic imaging, Artificial intelligence, Deep learning (Machine learning)
Format: Document (PDF)

Title: Computer-aided diagnosis of skin cancers using dermatology images.
Creator: Gilani, Syed Qasim, Marques, Oge, Florida Atlantic University, Department of Computer and Electrical Engineering and Computer Science, College of Engineering and Computer Science
Abstract/Description: Skin cancer is a prevalent cancer that significantly contributes to global mortality rates. Early detection is crucial for a high survival rate. Dermatologists primarily rely on visual inspection to diagnose skin cancers, but this method is inaccurate. Deep learning algorithms can enhance the diagnostic accuracy of skin cancers. However, these algorithms require substantial labeled data for effective training. Acquiring annotated data for skin cancer classification is time-consuming,...
Show moreSkin cancer is a prevalent cancer that significantly contributes to global mortality rates. Early detection is crucial for a high survival rate. Dermatologists primarily rely on visual inspection to diagnose skin cancers, but this method is inaccurate. Deep learning algorithms can enhance the diagnostic accuracy of skin cancers. However, these algorithms require substantial labeled data for effective training. Acquiring annotated data for skin cancer classification is time-consuming, expensive, and necessitates expert annotation. Moreover, skin cancer datasets often suffer from imbalanced data distribution. Generative Adversarial Networks (GANs) can be used to overcome the challenges of data scarcity and lack of labels by automatically generating skin cancer images. However, training and testing data from different distributions can introduce domain shift and bias, impacting the model’s performance. This dissertation addresses this issue by developing deep learning-based domain adaptation models. Additionally, this research emphasizes deploying deep learning models on hardware to enable real-time skin cancer detection, facilitating accurate diagnoses by dermatologists. Deploying conventional deep learning algorithms on hardware is not preferred due to the problem of high resource consumption. Therefore, this dissertation presents spiking neural network-based (SNN) models designed specifically for hardware implementation. SNNs are preferred for their power-efficient behavior and suitability for hardware deployment.
Show less
Date Issued: 2023
PURL: http://purl.flvc.org/fau/fd/FA00014233
Subject Headings: Deep learning (Machine learning), Diagnostic imaging, Skin--Cancer--Diagnosis
Format: Document (PDF)

Title: The Development and Application of Optical Bio-instrumentation to Investigate Metabolic State in Disease Models.
Creator: Ceyhan, Buse Nur, Ranji, Mahsa, Florida Atlantic University, Department of Computer and Electrical Engineering and Computer Science, College of Engineering and Computer Science
Abstract/Description: Metabolic dysfunction can present in conditions like cancer and neurodegeneration. Optical imaging techniques were employed to assist in the diagnosis and understanding of disease pathologies. A cryomesoscopy modality was designed and incorporated into an imaging device to investigate metabolic biomarkers. The new lens design provided higher magnifications and resolution of tissue data. The improved imaging capabilities gave detailed access to structural and biochemical changes that occur in...
Show moreMetabolic dysfunction can present in conditions like cancer and neurodegeneration. Optical imaging techniques were employed to assist in the diagnosis and understanding of disease pathologies. A cryomesoscopy modality was designed and incorporated into an imaging device to investigate metabolic biomarkers. The new lens design provided higher magnifications and resolution of tissue data. The improved imaging capabilities gave detailed access to structural and biochemical changes that occur in disease progression. Cryomesoscopy was applied to study mitochondrial redox state in preclinical models of Alzheimer’s and cancer. The optical imaging tools were utilized to visualize the livers and kidneys of mutated mice and investigate their metabolic states. The results in both investigations revealed oxidized metabolic states, a marker of oxidative stress and metabolic dysfunction. The cryomesoscopy system has proven instrumental in quantifying metabolic shifts and offers new insights into disease pathologies. Optical imaging can be applied to understanding metabolic mechanisms in many diseases.
Show less
Date Issued: 2024
PURL: http://purl.flvc.org/fau/fd/FA00014404
Subject Headings: Optical instruments, Optical Imaging, Diagnostic Techniques and Procedures
Format: Document (PDF)

Title: Bioinformatics-inspired binary image correlation: application to bio-/medical-images, microsarrays, finger-prints and signature classifications.
Creator: Pappusetty, Deepti, College of Engineering and Computer Science, Department of Computer and Electrical Engineering and Computer Science
Abstract/Description: The efforts addressed in this thesis refer to assaying the extent of local features in 2D-images for the purpose of recognition and classification. It is based on comparing a test-image against a template in binary format. It is a bioinformatics-inspired approach pursued and presented as deliverables of this thesis as summarized below: 1. By applying the so-called 'Smith-Waterman (SW) local alignment' and 'Needleman-Wunsch (NW) global alignment' approaches of bioinformatics, a test 2D-image...
Show moreThe efforts addressed in this thesis refer to assaying the extent of local features in 2D-images for the purpose of recognition and classification. It is based on comparing a test-image against a template in binary format. It is a bioinformatics-inspired approach pursued and presented as deliverables of this thesis as summarized below: 1. By applying the so-called 'Smith-Waterman (SW) local alignment' and 'Needleman-Wunsch (NW) global alignment' approaches of bioinformatics, a test 2D-image in binary format is compared against a reference image so as to recognize the differential features that reside locally in the images being compared 2. SW and NW algorithms based binary comparison involves conversion of one-dimensional sequence alignment procedure (indicated traditionally for molecular sequence comparison adopted in bioinformatics) to 2D-image matrix 3. Relevant algorithms specific to computations are implemented as MatLabTM codes 4. Test-images considered are: Real-world bio-/medical-images, synthetic images, microarrays, biometric finger prints (thumb-impressions) and handwritten signatures. Based on the results, conclusions are enumerated and inferences are made with directions for future studies.
Show less
Date Issued: 2011
PURL: http://purl.flvc.org/FAU/3333052
Subject Headings: Bioinformatics, Statistical methods, Diagnostic imaging, Digital techniques, Image processing, Digital techniques, Pattern perception, Data processing, DNA microarrays
Format: Document (PDF)

Title: Comparison of treatment plans calculated using ray tracing and Monte Carlo algorithms for lung cancer patients having undergone radiotherapy with cyberknife.
Creator: Pennington, Andreea, Selvaraj, Raj, Florida Atlantic University, Charles E. Schmidt College of Science, Department of Physics
Abstract/Description: The purpose of this research is to determine the feasibility of introducing the Monte Carlo (MC) dose calculation algorithm into the clinical practice. Unlike the Ray Tracing (RT) algorithm, the MC algorithm is not affected by the tissue inhomogeneities, which are significant inside the chest cavity. A retrospective study was completed for 102 plans calculated using both the RT and MC algorithms. The D95 of the PTV was 26% lower for the MC calculation. The first parameter of conformality, as...
Show moreThe purpose of this research is to determine the feasibility of introducing the Monte Carlo (MC) dose calculation algorithm into the clinical practice. Unlike the Ray Tracing (RT) algorithm, the MC algorithm is not affected by the tissue inhomogeneities, which are significant inside the chest cavity. A retrospective study was completed for 102 plans calculated using both the RT and MC algorithms. The D95 of the PTV was 26% lower for the MC calculation. The first parameter of conformality, as defined as the ratio of the Prescription Isodose Volume to the PTV Volume was on average 1.27 for RT and 0.67 for MC. The results confirm that the RT algorithm significantly overestimates the dosages delivered confirming previous analyses. Correlations indicate that these overestimates are largest for small PTV and/or when the ratio of the volume of lung tissue to the PTV approaches 1.
Show less
Date Issued: 2014
PURL: http://purl.flvc.org/fau/fd/FA00004316
Subject Headings: Computer graphics, Diagnostic imaging, Image guided radiation therapy, Lung cancer -- Treatment, Lungs -- Cancer -- Radiotherapy, Monte Carlo method
Format: Document (PDF)

Title: Statistical and Entropy Considerations for Ultrasound Tissue Characterization.
Creator: Navumenka, Khrystsina, Aalo, Valentine A., Florida Atlantic University, College of Engineering and Computer Science, Department of Computer and Electrical Engineering and Computer Science
Abstract/Description: Modern cancerous tumor diagnostics is nearly impossible without invasive methods, such as biopsy, that may require involved surgical procedures. In recent years some work has been done to develop alternative non-invasive methods of medical diagnostics. For this purpose, the data obtained from an ultrasound image of the body crosssection, has been analyzed using statistical models, including Rayleigh, Rice, Nakagami, and K statistical distributions. The homodyned-K (H-K) distribution has been...
Show moreModern cancerous tumor diagnostics is nearly impossible without invasive methods, such as biopsy, that may require involved surgical procedures. In recent years some work has been done to develop alternative non-invasive methods of medical diagnostics. For this purpose, the data obtained from an ultrasound image of the body crosssection, has been analyzed using statistical models, including Rayleigh, Rice, Nakagami, and K statistical distributions. The homodyned-K (H-K) distribution has been found to be a good statistical tool to analyze the envelope and/or the intensity of backscattered signal in ultrasound tissue characterization. However, its use has usually been limited due to the fact that its probability density function (PDF) is not available in closed-form. In this work we present a novel closed-form representation for the H-K distribution. In addition, we propose using the first order approximation of the H-K distribution, the I-K distribution that has a closed-form, for the ultrasound tissue characterization applications. More specifically, we show that some tissue conditions that cause the backscattered signal to have low effective density values, can be successfully modeled by the I-K PDF. We introduce the concept of using H-K PDF-based and I-K PDF-based entropies as additional tools for characterization of ultrasonic breast tissue images. The entropy may be used as a goodness of fit measure that allows to select a better-fitting statistical model for a specific data set. In addition, the values of the entropies as well as the values of the statistical distribution parameters, allow for more accurate classification of tumors.
Show less
Date Issued: 2017
PURL: http://purl.flvc.org/fau/fd/FA00004922, http://purl.flvc.org/fau/fd/FA00004922
Subject Headings: Ultrasonics in medicine., Artificial intelligence., Computer vision in medicine., Diagnostic ultrasonic imaging., Bioinformatics.
Format: Document (PDF)

Title: Sparse Modeling Applied to Patient Identification for Safety in Medical Physics Applications.
Creator: Lewkowitz, Stephanie, Kalantzis, Georgios, Florida Atlantic University, Charles E. Schmidt College of Science, Department of Physics
Abstract/Description: Every scheduled treatment at a radiation therapy clinic involves a series of safety protocol to ensure the utmost patient care. Despite safety protocol, on a rare occasion an entirely preventable medical event, an accident, may occur. Delivering a treatment plan to the wrong patient is preventable, yet still is a clinically documented error. This research describes a computational method to identify patients with a novel machine learning technique to combat misadministration.The patient...
Show moreEvery scheduled treatment at a radiation therapy clinic involves a series of safety protocol to ensure the utmost patient care. Despite safety protocol, on a rare occasion an entirely preventable medical event, an accident, may occur. Delivering a treatment plan to the wrong patient is preventable, yet still is a clinically documented error. This research describes a computational method to identify patients with a novel machine learning technique to combat misadministration.The patient identification program stores face and fingerprint data for each patient. New, unlabeled data from those patients are categorized according to the library. The categorization of data by this face-fingerprint detector is accomplished with new machine learning algorithms based on Sparse Modeling that have already begun transforming the foundation of Computer Vision. Previous patient recognition software required special subroutines for faces and di↵erent tailored subroutines for fingerprints. In this research, the same exact model is used for both fingerprints and faces, without any additional subroutines and even without adjusting the two hyperparameters. Sparse modeling is a powerful tool, already shown utility in the areas of super-resolution, denoising, inpainting, demosaicing, and sub-nyquist sampling, i.e. compressed sensing. Sparse Modeling is possible because natural images are inherrently sparse in some bases, due to their inherrant structure. This research chooses datasets of face and fingerprint images to test the patient identification model. The model stores the images of each dataset as a basis (library). One image at a time is removed from the library, and is classified by a sparse code in terms of the remaining library. The Locally Competetive Algorithm, a truly neural inspired Artificial Neural Network, solves the computationally difficult task of finding the sparse code for the test image. The components of the sparse representation vector are summed by `1 pooling, and correct patient identification is consistently achieved 100% over 1000 trials, when either the face data or fingerprint data are implemented as a classification basis. The algorithm gets 100% classification when faces and fingerprints are concatenated into multimodal datasets. This suggests that 100% patient identification will be achievable in the clinal setting.
Show less
Date Issued: 2016
PURL: http://purl.flvc.org/fau/fd/FA00004721, http://purl.flvc.org/fau/fd/FA00004721
Subject Headings: Computer vision in medicine, Diagnostic imaging -- Data processing, Mathematical models, Medical errors -- Prevention, Medical physics, Sampling (Statistics)
Format: Document (PDF)

Title: Suffering in the midst of technology: the lived experience of an abnormal prenatal ultrasound.
Creator: Gottlieb, Jeanne C., Christine E. Lynn College of Nursing
Abstract/Description: The purpose of this hermeneutic phenomenological study was to understand the essence of the lived experience of women after having an abnormal prenatal ultrasound. One hundred years ago, health disciplines had limited therapies for prenatal and neonatal disorders. During this period, the eugenics movement influenced leaders to involuntarily sterilize individuals who were sought to be "unfit" to prevent disorders in offspring. ... One of these contemporary reproductive genetic technologies is...
Show moreThe purpose of this hermeneutic phenomenological study was to understand the essence of the lived experience of women after having an abnormal prenatal ultrasound. One hundred years ago, health disciplines had limited therapies for prenatal and neonatal disorders. During this period, the eugenics movement influenced leaders to involuntarily sterilize individuals who were sought to be "unfit" to prevent disorders in offspring. ... One of these contemporary reproductive genetic technologies is the use of ultrasound and serum bio-medical markers for detection of congenital, chromosome, and genetic disorders. When ultrasounds reveal abnormal findings, the perceived perfect pregnancy vanishes and gives way to feelings of shock, disbelief, fear, guilt, loss, and threats to self and their unborn baby. Twelve women who had an abnormal ultrasound were interviewed within the context of their cultural values and beliefs. The method of van Manen's hermeneutic phenomenology illuminated the meaning for these women in their life worlds. ... They endured this experience through their own coping mechanisms, but often felt uncertainty and emotional turmoil until the birth. The women also sought comfort through their cultural values, beliefs, and traditions. In coping with the risks found on this abnormal ultrasound, women often selected silence or blocking perceived threats. With these coping methods, they were alone in their suffering. ... Health providers, in not recognizing these women's misunderstandings and emotional fears, abandoned them in their psychosocial and cultural needs. The significance reveals that nurses and health providers need to infuse human caring ways of being, knowing, and doing within advanced technological environments.
Show less
Date Issued: 2013
PURL: http://purl.flvc.org/fcla/dt/3362381
Subject Headings: Medical genetics, Medical care, Decision-making, Health services accessibility, Abortion, Moral and ethical aspects, Pregnancy, Complications, Diagnostic ultrasonic imaging, Communication in medicine, Genetic counseling, Genetic disorders, Nursing, Standards
Format: Document (PDF)

FAU Theses and Dissertations (11)	+ -
FAU Graduate Student Research (1)	+ -

Diagnostic imaging (4)	+ -
Artificial intelligence (2)	+ -
Deep learning (Machine learning) (2)	+ -
Diagnostic ultrasonic imaging (2)	+ -
Machine learning (2)	+ -

Abortion (1)	+ -
Artificial intelligence. (1)	+ -
Bioinformatics (1)	+ -
Bioinformatics. (1)	+ -
Communication in medicine (1)	+ -
Complications (1)	+ -
Computational intelligence (1)	+ -
Computer graphics (1)	+ -
Computer vision in medicine (1)	+ -
Computer vision in medicine. (1)	+ -
DNA microarrays (1)	+ -
Data processing (1)	+ -
Decision-making (1)	+ -
Dermatology (1)	+ -
Diagnostic Imaging (1)	+ -
Diagnostic Imaging--classification (1)	+ -
Diagnostic Techniques and Procedures (1)	+ -
Diagnostic imaging -- Data processing (1)	+ -
Diagnostic imaging--Data processing (1)	+ -
Diagnostic ultrasonic imaging. (1)	+ -
Digital techniques (1)	+ -
Electrical impedance tomography (1)	+ -
Fetus--Abnormalities (1)	+ -
Genetic counseling (1)	+ -
Genetic disorders (1)	+ -

Electronic Thesis or Dissertation (11)	+ -
abstract (1)	+ -

Florida Atlantic University (9)	+ -
College of Engineering and Computer Science (8)	+ -
Department of Computer and Electrical Engineering and Computer Science (8)	+ -
Charles E. Schmidt College of Science (2)	+ -
Department of Physics (2)	+ -

Search In

Sort Results By:

Narrow results by:

Collection

Topical Subject

Genre Subject

Genre

Type of Resource

Language

Creator

Format

Owner