Dataset Information

Stratifying Ovarian Cancer Risk Using Personal Health Data.

ABSTRACT: Purpose: Screening the general population for ovarian cancer is not recommended by every major medical or public health organization because the harms from screening outweigh the benefit it provides. To improve ovarian cancer detection and survival many are looking at high-risk populations who would benefit from screening. Methods: We train a neural network on readily available personal health data to predict and stratify ovarian cancer risk. We use two different datasets to train our network: The National Health Interview Survey and Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Results: Our model has an area under the receiver operating characteristic curve of 0.71. We further demonstrate how the model could be used to stratify patients into different risk categories. A simple 3-tier scheme classifies 23.8% of those with cancer and 1.0% of those without as high-risk similar to genetic testing, and 1.1% of those with cancer and 24.4% of those without as low risk. Conclusion: The developed neural network offers a cost-effective and non-invasive way to identify those who could benefit from targeted screening.

SUBMITTER: Hart GR

PROVIDER: S-EPMC7931902 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Stratifying Ovarian Cancer Risk Using Personal Health Data.

Hart Gregory R GR Nartowt Bradley J BJ Muhammad Wazir W Liang Ying Y Huang Gloria S GS Deng Jun J

Frontiers in big data 20190702

<b>Purpose:</b> Screening the general population for ovarian cancer is not recommended by every major medical or public health organization because the harms from screening outweigh the benefit it provides. To improve ovarian cancer detection and survival many are looking at high-risk populations who would benefit from screening. <b>Methods:</b> We train a neural network on readily available personal health data to predict and stratify ovarian cancer risk. We use two different datasets to train ...[more]

PMID: 33693347

Similar Datasets

Project description:BackgroundAs the need for sharing genomic data grows, privacy issues and concerns, such as the ethics surrounding data sharing and disclosure of personal information, are raised.ObjectiveThe main purpose of this study was to verify whether genomic data is sufficient to predict a patient's personal information.MethodsRNA expression data and matched patient personal information were collected from 9538 patients in The Cancer Genome Atlas program. Five personal information variables (age, gender, race, cancer type, and cancer stage) were recorded for each patient. Four different machine learning algorithms (support vector machine, decision tree, random forest, and artificial neural network) were used to determine whether a patient's personal information could be accurately predicted from RNA expression data. Performance measurement of the prediction models was based on the accuracy and area under the receiver operating characteristic curve. We selected five cancer types (breast carcinoma, kidney renal clear cell carcinoma, head and neck squamous cell carcinoma, low-grade glioma, and lung adenocarcinoma) with large samples sizes to verify whether predictive accuracy would differ between them. We also validated the efficacy of our four machine learning models in analyzing normal samples from 593 cancer patients.ResultsIn most samples, personal information with high genetic relevance, such as gender and cancer type, could be predicted from RNA expression data alone. The prediction accuracies for gender and cancer type, which were the best models, were 0.93-0.99 and 0.78-0.94, respectively. Other aspects of personal information, such as age, race, and cancer stage, were difficult to predict from RNA expression data, with accuracies ranging from 0.0026-0.29, 0.76-0.96, and 0.45-0.79, respectively. Among the tested machine learning methods, the highest predictive accuracy was obtained using the support vector machine algorithm (mean accuracy 0.77), while the lowest accuracy was obtained using the random forest method (mean accuracy 0.65). Gender and race were predicted more accurately than other variables in the samples. On average, the accuracy of cancer stage prediction ranged between 0.71-0.67, while the age prediction accuracy ranged between 0.18-0.23 for the five cancer types.ConclusionsWe attempted to predict patient information using RNA expression data. We found that some identifiers could be predicted, but most others could not. This study showed that personal information available from RNA expression data is limited and this information cannot be used to identify specific patients.

Project description:The data presented here were obtained for a New Zealand nationwide population-based case-control analysis undertaken to assess the association between ovarian cancer and depot medroxyprogesterone acetate (DMPA), intrauterine contraceptive devices (IUDs), and vasectomy of a woman's sexual partner (Chesang et al., 2021). The research involved women aged 35 to 69 years. Controls were randomly selected from the New Zealand electoral roll. Cases were women with a diagnosis of incident ovarian cancer recruited from the New Zealand Cancer Registry and had to be listed on the electoral roll. Data collection was conducted between 1st May 2013 and 31st October, 2015. A structured postal questionnaire was used to gather information. Data were analysed using IBM Statistical Package for the Social Sciences (IBM SPSS statistics 22). Odds ratios adjusted for age were calculated using the method of Mantel and Haensze (Rosner et al., 2007). For multivariable analyses, binary logistic regression was used. Description of study participants and age-adjusted and multivariable analyses of the association between ever-use and specifics of use of DMPA, IUDs, and vasectomy were presented in a journal article (Chesang et al., 2021). Here, we present data from analyses of the risk of ovarian cancer by histological type associated with the use of DMPA, IUDs and ever having had a vasectomised partner. In addition, analyses assessing the association between ovarian cancer and these contraceptives restricted to ever-users and never-users of hormonal contraceptives (defined as oral contraceptives or DMPA) are presented. Data from analyses of the association between history of tubal ligation and the risk of ovarian cancer are also presented. These data, including the findings of a related study (Chesang et al., 2021) and the raw data, can be included in a collaborative analysis of existing studies undertaken to assess the association between IUDs, long-acting progestogen-based contraceptives, and partner vasectomy and the risk of ovarian cancer.

Project description:Rationale: NT-proBNP (N-terminal pro-brain natriuretic peptide), a biomarker of cardiac origin, is used to risk stratify patients with pulmonary arterial hypertension (PAH). Its limitations include poor sensitivity to early vascular pathology. Other biomarkers of vascular or systemic origin may also be useful in the management of PAH. Objectives: Identify prognostic proteins in PAH that complement NT-proBNP and clinical risk scores. Methods: An aptamer-based assay (SomaScan version 4) targeting 4,152 proteins was used to measure plasma proteins in patients with idiopathic, heritable, or drug-induced PAH from the UK National Cohort of PAH (n = 357) and the French EFORT (Evaluation of Prognostic Factors and Therapeutic Targets in PAH) study (n = 79). Prognostic proteins were identified in discovery-replication analyses of UK samples. Proteins independent of 6-minute-walk distance and NT-proBNP entered least absolute shrinkage and selection operator modeling, and the best combination in a single score was evaluated against clinical targets in EFORT. Measurements and Main Results: Thirty-one proteins robustly informed prognosis independent of NT-proBNP and 6-minute-walk distance in the UK cohort. A weighted combination score of six proteins was validated at baseline (5-yr mortality; area under the curve [AUC], 0.73; 95% confidence interval [CI], 0.63-0.85) and follow-up in EFORT (AUC, 0.84; 95% CI, 0.75-0.94; P = 9.96 × 10-6). The protein score risk stratified patients independent of established clinical targets and risk equations. The addition of the six-protein model score to NT-proBNP improved prediction of 5-year outcomes from AUC 0.762 (0.702-0.821) to 0.818 (0.767-0.869) by receiver operating characteristic analysis (P = 0.00426 for difference in AUC) in the UK replication and French samples combined. Conclusions: The plasma proteome informs prognosis beyond established factors in PAH and may provide a more sensitive measure of therapeutic response.

Dataset Information

Stratifying Ovarian Cancer Risk Using Personal Health Data.

Publications

Stratifying Ovarian Cancer Risk Using Personal Health Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets