Dataset Information

A quantitative performance study of two automatic methods for the diagnosis of ovarian cancer.

ABSTRACT: We present a quantitative study of the performance of two automatic methods for the early detection of ovarian cancer that can exploit longitudinal measurements of multiple biomarkers. The study is carried out for a subset of the data collected in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). We use statistical analysis techniques, such as the area under the Receiver Operating Characteristic (ROC) curve, for evaluating the performance of two techniques that aim at the classification of subjects as either healthy or suffering from the disease using time-series of multiple biomarkers as inputs. The first method relies on a Bayesian hierarchical model that establishes connections within a set of clinically interpretable parameters. The second technique is a purely discriminative method that employs a recurrent neural network (RNN) for the binary classification of the inputs. For the available dataset, the performance of the two detection schemes is similar (the area under ROC curve is 0.98 for the combination of three biomarkers) and the Bayesian approach has the advantage that its outputs (parameters estimates and their uncertainty) can be further analysed by a clinical expert.

SUBMITTER: Vazquez MA

PROVIDER: S-EPMC6146655 | biostudies-literature | 2018 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A quantitative performance study of two automatic methods for the diagnosis of ovarian cancer.

Vázquez Manuel A MA Mariño Inés P IP Blyuss Oleg O Ryan Andy A Gentry-Maharaj Aleksandra A Kalsi Jatinderpal J Manchanda Ranjit R Jacobs Ian I Menon Usha U Zaikin Alexey A

Biomedical signal processing and control 20180901

We present a quantitative study of the performance of two automatic methods for the early detection of ovarian cancer that can exploit longitudinal measurements of multiple biomarkers. The study is carried out for a subset of the data collected in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). We use statistical analysis techniques, such as the area under the Receiver Operating Characteristic (ROC) curve, for evaluating the performance of two techniques that aim at the classif ...[more]

PMID: 30245736

Similar Datasets

Project description:BackgroundOvarian cancer is often diagnosed at a late stage, when survival is poor. Qualitative narratives of patients' pathways to ovarian cancer diagnoses may identify opportunities for earlier cancer detection and, consequently, earlier stage at diagnosis.MethodsWe conducted semi-structured interviews of ovarian cancer patients and survivors (n = 14) and healthcare providers (n = 11) between 10/2019 and 10/2021. Interviews focused on the time leading up to an ovarian cancer diagnosis. Thematic analysis was conducted by two independent reviewers using a two-phase deductive and inductive coding approach. Deductive coding used a priori time intervals from the validated Model of Pathways to Treatment (MPT), including self-appraisal and management of symptoms, medical help-seeking, diagnosis, and pre-treatment. Inductive coding identified common themes within each stage of the MPT across patient and provider interviews.ResultsThe median age at ovarian cancer diagnosis was 61.5 years (range, 29-78 years), and the majority of participants (11/14) were diagnosed with advanced-stage disease. The median time from first symptom to initiation of treatment was 2.8 months (range, 19 days to 4.7 years). The appraisal and help-seeking intervals contributed the greatest delays in time-to-diagnosis for ovarian cancer. Nonspecific symptoms, perceptions of health and aging, avoidant coping strategies, symptom embarrassment, and concerns about potential judgment from providers prolonged the appraisal and help-seeking intervals. Patients and providers also emphasized access to care, including financial access, as critical to a timely diagnosis.ConclusionInterventions are urgently needed to reduce ovarian cancer morbidity and mortality. Population-level screening remains unlikely to improve ovarian cancer survival, but findings from our study suggest that developing interventions to improve self-appraisal of symptoms and reduce barriers to help-seeking could reduce time-to-diagnosis for ovarian cancer. Affordability of care and insurance may be particularly important for ovarian cancer patients diagnosed in the United States.

Dataset Information

A quantitative performance study of two automatic methods for the diagnosis of ovarian cancer.

Publications

A quantitative performance study of two automatic methods for the diagnosis of ovarian cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets