Obstetrics and Gynaecology - Research Publications
Permanent URI for this collection
Search Results
1 - 3 of 3
-
ItemTransfer Learning to Enhance Amenorrhea Status Prediction in Cancer and Fertility Data with Missing Values(Productivity Press (Taylor & Francis), 2020)
-
ItemImputation techniques on missing values in breast cancer treatment and fertility data.(BioMed Central, 2019-12-01)Clinical decision support using data mining techniques offers more intelligent way to reduce the decision error in the last few years. However, clinical datasets often suffer from high missingness, which adversely impacts the quality of modelling if handled improperly. Imputing missing values provides an opportunity to resolve the issue. Conventional imputation methods adopt simple statistical analysis, such as mean imputation or discarding missing cases, which have many limitations and thus degrade the performance of learning. This study examines a series of machine learning based imputation methods and suggests an efficient approach to in preparing a good quality breast cancer (BC) dataset, to find the relationship between BC treatment and chemotherapy-related amenorrhoea, where the performance is evaluated with the accuracy of the prediction. To this end, the reliability and robustness of six well-known imputation methods are evaluated. Our results show that imputation leads to a significant boost in the classification performance compared to the model prediction based on listwise deletion. Furthermore, the results reveal that most methods gain strong robustness and discriminant power even the dataset experiences high missing rate (> 50%).
-
ItemExploring the facilitators and barriers to using an online infertility risk prediction tool (FoRECAsT) for young women with breast cancer: a qualitative study protocol.(BMJ Journals, 2020-02-10)INTRODUCTION: As cancer treatments may impact on fertility, a high priority for young patients with breast cancer is access to evidence-based, personalised information for them and their healthcare providers to guide treatment and fertility-related decisions prior to cancer treatment. Current tools to predict fertility outcomes after breast cancer treatments are imprecise and do not offer individualised prediction. To address the gap, we are developing a novel personalised infertility risk prediction tool (FoRECAsT) for premenopausal patients with breast cancer that considers current reproductive status, planned chemotherapy and adjuvant endocrine therapy to determine likely post-treatment infertility. The aim of this study is to explore the feasibility of implementing this FoRECAsT tool into clinical practice by exploring the barriers and facilitators of its use among patients and healthcare providers. METHODS AND ANALYSIS: A cross-sectional exploratory study is being conducted using semistructured in-depth telephone interviews with 15-20 participants each from the following groups: (1) premenopausal patients with breast cancer younger than 40, diagnosed within last 5 years, (2) breast surgeons, (3) breast medical oncologists, (4) breast care nurses (5) fertility specialists and (6) fertility preservation nurses. Patients with breast cancer are being recruited from the joint Breast Service of three affiliated institutions of Victorian Comprehensive Cancer Centre in Melbourne, Australia-Peter MacCallum Cancer Centre, Royal Melbourne Hospital and Royal Women's Hospital, and clinicians are being recruited from across Australia. Interviews are being audio recorded, transcribed verbatim and imported into qualitative data analysis software to facilitate data management and analyses. ETHICS AND DISSEMINATION: The study protocol has been approved by Melbourne Health Human Research Ethics Committee, Australia (HREC number: 2017.163). Confidentiality and privacy are maintained at every stage of the study. Findings will be disseminated through peer-reviewed scholarly and scientific journals, national and international conference presentations, social media, broadcast media, print media, internet and various community/stakeholder engagement activities.