Medicine and Health Sciences Commons^™

A Novel Fuzzy Relative-Position-Coding Transformer For Breast Cancer Diagnosis Using Ultrasonography, Yanhui Guo, Ruquan Jiang, Xin Gu, Heng-Da Cheng, Harish Garg

Computer Science Faculty and Staff Publications

Breast cancer is a leading cause of death in women worldwide, and early detection is crucial for successful treatment. Computer-aided diagnosis (CAD) systems have been developed to assist doctors in identifying breast cancer on ultrasound images. In this paper, we propose a novel fuzzy relative-position-coding (FRPC) Transformer to classify breast ultrasound (BUS) images for breast cancer diagnosis. The proposed FRPC Transformer utilizes the self-attention mechanism of Transformer networks combined with fuzzy relative-position-coding to capture global and local features of the BUS images. The performance of the proposed method is evaluated on one benchmark dataset and compared with those obtained by …

Leveraging Telehealth Supportive Oncology Services To Combat Covid-19 Isolation In Breast Cancer Patients: A Cancer Center’S Perspective, Jake R. Erickson, Maranda K. Pahlkotter, Paul D. Thielking, Jane M. Porretta, Regina Rosenthal, Cindy B. Matsen, Kirstyn E. Brownson

Patient Experience Journal

During the COVID-19 pandemic, breast cancer patient in-person visits were converted to telehealth visits. Given our concerns about increased isolation amongst breast cancer patients during the pandemic, and the deleterious effects of such isolation on patient outcomes, we investigated utilization rates of psycho-social services amongst newly diagnosed breast cancer patients at our institution. We explored visit platforms (telehealth versus in-person) and time points prior to and encompassing the early pandemic. Despite decreased new breast cancer visits, there was a greater than 2-fold increase in supportive oncology service encounters in breast cancer patients during COVID-19 suggesting increased need for psycho-oncology resources. …

Perspectives Of People With Cancer Or Hereditary Cancer Risk On The Use And Value Of Online Peer Support, Jill Holdren, Karl Surkan, Andrea Downing

Journal of Patient-Centered Research and Reviews

Purpose: People with cancer routinely seek information and support in peer groups online. While peer communities constitute a major component of the health care landscape, they exist in isolation from clinical and research institutions. This study aimed to explore how and why cancer patients utilize online peer support groups and how they might be improved.

Methods: A convenience sample of members of 6 closed Facebook cancer peer support groups (n = 291) participated in an online needs assessment survey. We further conducted semi-structured interviews with 14 members and 6 moderator-patients, hand-coding the free-text responses and interview transcripts.

Results: Group …

Evaluation Of A Remote Symptom Assessment And Management (Sam) System For People Receiving Adjuvant Chemotherapy For Breast Or Colorectal Cancer: Mixed Methods Study, Lisa Whitehead, Laura Emery, Deborah Kirk, Diane Twigg, Deborah Brown, Joanna Dewar

Research outputs 2014 to 2021

©Lisa Whitehead, Laura Emery, Deborah Kirk, Diane Twigg, Deborah Brown, Joanna Dewar. Background: The Symptom Assessment and Management (SAM) program is a structured, online, nurse-supported intervention to support symptom self-management in people receiving adjuvant chemotherapy post surgery for breast or colorectal cancer. Objective: The objective of this study was to describe the development, implementation strategy, and evaluation of the SAM system. Methods: The development of the SAM program involved 3 phases. In phase 1, the web app was developed through consultation with consumers and clinicians and of the literature to ensure that the system was evidence-based and reflected the realities …

Markov Decision Process Approach To Strategize National Breast Cancer Screening Policy In Data-Limited Settings, Vijeta Deshpande

Masters Theses

Early diagnosis is a promising strategy to reduce premature mortalities and for optimal use of resources. But the absence of mathematical models specific to the data settings in LMIC’s impedes the construction of economic analysis necessary for decision-makers in the development of cancer control programs. This thesis presents a new methodology for parameterizing the natural history model of breast cancer based on data availabilities in low and middle income countries, and formulation of a control optimization problem to find the optimal screening schedule for mammography screening, solved using dynamic programming. As harms and benefits are known to increase with the …

Stay Beautiful -- Stay Alive: Assessing The Receptivity Of African American Beauty Salon Owners To The Integration Of Breast Cancer Intervention Programs Into Salon Operations, Funmi Apantaku-Onayemi

Ed.D. Dissertations

The lower incidence rate of breast cancer in African American women is dwarfed by the excessive number of deaths due to late diagnosis and treatment. Lack of screening, socioeconomic factors, fatalistic beliefs and inequality of care are major contributing factors. Studies have suggested that those who had more knowledge about breast cancer are more likely to have reduced fatalistic attitudes and engage in screening behaviors. This study investigated beauty salons as sustainable and viable venues to reach women with health intervention programs because they fit the prescriptions of the principles of adult learning. In a mixed-method, descriptive study involving 115 …

Artificial Intelligence – I: A New Hybrid Agent-Based Modeling & Simulation Decision Support System For Breast Cancer Data Analysis, Amnah Siddiqa, Muaz Niazi, Farah Mustafa, Habib Bokhari, Amir Hussain, Noreen Akram, Shabnum Shaheen, Fouzia Ahmed, Sarah Iqbal

International Conference on Information and Communication Technologies

In this paper, we present a novel technique of building hybrid decision support systems which integrates traditional decision support systems with agent based models for use in breast cancer analysis for better prediction and recommendation. Our system is based on using queries from data (converted to a standardized electronic template) to provide for simulation variables in an agent-based model. The goal is to develop an ICT tool to assist non-specialist biologist researcher users in performing analysis of large amounts of data by applying simple simulation techniques. To demonstrate the effectiveness of this novel decision support system, an extensive breast cancer …

Nonparametric Techniques To Extract Fuzzy Rules For Breast Cancer Diagnosis Problem, Manish Sarkar, Tze-Yun Leong

Research Collection School Of Computing and Information Systems

This paper addresses breast cancer diagnosis problem as a pattern classification problem. Specifically, the problem is studied using Wisconsin-Madison breast cancer data set. Fuzzy rules are generated from the input-output relationship so that the diagnosis becomes easier and transparent for both patients and physicians. For each class, at least one training pattern is chosen as the prototype, provided (a) the maximum membership of the training pattern is in the given class, and (b) among all the training patterns, the neighborhood of this training pattern has the least fuzzy-rough uncertainty in the given class. Using the fuzzy-rough uncertainty, a cluster is …