SurvMetrics: An R package for Predictive Evaluation Metrics in Survival Analysis

Zhou, Hanpu; Wang, Hong; Wang, Sizheng; Zou, Yi

doi:10.32614/rj-2023-009

Cited by 5 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two types of prediction outcomes were used: time-to-T2D diagnosis and a binary outcome defined by whether a T2D diagnosis was recorded within 10 years from baseline. For the time-to-T2D outcome, C-index and Brier scores were calculated using the SurvMetrics R package (version 0.5.0) [37]. C-index gives a measure of discrimination for a model, defined as proportion of concordant pairs of individuals predicted by the model.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Feature pre-selection for the development of epigenetic biomarkers

Cheng,

Gieger,

Campbell

et al. 2024

Preprint

View full text Add to dashboard Cite

Over the last decade, a plethora of blood-based DNA methylation biomarkers have been developed to track differences in ageing, lifestyle, health, and biological outcomes. Typically, penalised regression models are used to generate these predictors, with hundreds or thousands of CpGs included as potential features. However, in such ultra high-dimensional settings, the effectiveness of these methods may be reduced. Here, we introduce Related Trait-based Feature Screening (RTFS), a method for performing CpG pre-selection for incident disease prediction models by utilising associations between CpGs and health-related continuous traits. In a comparison with commonly used CpG pre-selection methods, we evaluate resulting downstream Cox proportional-hazards prediction models for 10-year type 2 diabetes (T2D) onset risk in Generation Scotland (n=18,414). The top performing models utilised incident T2D EWAS (AUC=0.881, PRAUC=0.279) and RTFS (AUC=0.877, PRAUC=0.277). The resulting models also improve prediction over a model using standard risk factors only (AUC=0.841, PRAUC=0.194) and replication was observed in the German-based KORA study (n=4,261) RTFS is a flexible and generalisable framework that can help to refine biomarker development for incident disease outcomes.

show abstract

Section: Methodsmentioning

confidence: 99%

“…For the time-to-T2D outcome, C-index and Brier scores were calculated using the SurvMetrics R package (version 0.5.0) [37]. C-index gives a measure of discrimination for a model, defined as proportion of concordant pairs of individuals predicted by the model.…”

mentioning

confidence: 99%

Feature pre-selection for the development of epigenetic biomarkers

Cheng,

Gieger,

Campbell

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…All P values were from 2-sided tests and results were deemed statistically significant at P < .05. All analyses were estimated using the survival and the SurvMetrics packages in R, version 4.2.3 . To maintain open science, the R script used to clean the data and estimate all of the statistical analyses is provided in eAppendix 4 in Supplement 1.…”

Section: Methodsmentioning

confidence: 99%

Incarceration of Youths in an Adult Correctional Facility and Risk of Premature Death

2023

View full text Add to dashboard Cite

ImportanceYouths incarcerated in adult correctional facilities are exposed to a variety of adverse circumstances that could diminish psychological and physical health, potentially leading to early mortality.ObjectiveTo evaluate whether being incarcerated in an adult correctional facility as a youth was associated with mortality between 18 and 39 years of age.Design, Setting, and ParticipantsThis cohort study relied on longitudinal data collected from 1997 to 2019 as part of the National Longitudinal Survey of Youth–1997, a nationally representative sample of 8984 individuals born in the United States between January 1, 1980, and December 1, 1984. The data analyzed for the current study were derived from annual interviews between 1997 and 2011 and interviews every other year from 2013 to 2019 (19 interviews in total). Participants were limited to respondents aged 17 years or younger during the 1997 interview and alive during their 18th birthday (8951 individuals; &gt;99% of the original sample). Statistical analysis was performed from November 2022 to May 2023.InterventionIncarceration in an adult correctional facility before the age of 18 years compared with being arrested before the age of 18 years or never arrested or incarcerated before the age of 18 years.Main Outcomes and MeasuresThe main outcome for the study was age at mortality between 18 and 39 years of age.ResultsThe sample of 8951 individuals included 4582 male participants (51%), 61 American Indian or Alaska Native participants (1%), 157 Asian participants (2%), 2438 Black participants (27%), 1895 Hispanic participants (21%), 1065 participants of other race (12%), and 5233 White participants (59%). A total of 225 participants (3%) died during the study period, with a mean (SD) age at death of 27.7 (5.9) years. Incarceration in an adult correctional facility before the age of 18 years was associated with an increased risk of earlier mortality between 18 and 39 years of age compared with individuals who were never arrested or incarcerated before the age of 18 years (time ratio, 0.67; 95% CI, 0.47-0.95). Being arrested before the age of 18 years was associated with an increased risk of earlier mortality between 18 and 39 years of age when compared with individuals who were never arrested or incarcerated before the age of 18 years (time ratio, 0.82; 95% CI, 0.73-0.93).Conclusions and RelevanceIn this cohort study of 8951 youths, the survival model suggested that being incarcerated in an adult correctional facility may be associated with an increased risk of early mortality between 18 and 39 years of age.

show abstract

“…Typically, high values of C > 0.8 are needed to prove the validity of a new clinical biomarker [9]. When it comes to the comparison of algorithms with close C-indices or under special conditions, other metrics are used [10]. For model comparison regarding their utility in medical decision making, calibration is included in the list of quality check-ups in central journals.…”

Section: The Cox Modelmentioning

confidence: 99%

Review: Deep Learning-Based Survival Analysis of Omics and Clinicopathological Data

Sidorova,

Lozano

2024

Inventions

View full text Add to dashboard Cite

The 2017–2024 period has been prolific in the area of the algorithms for deep-based survival analysis. We have searched the answers to the following three questions. (1) Is there a new “gold standard” already in clinical data analysis? (2) Does the DL component lead to a notably improved performance? (3) Are there tangible benefits of deep-based survival that are not directly attainable with non-deep methods? We have analyzed and compared the selected influential algorithms devised for two types of input: clinicopathological (a small set of numeric, binary and categorical) and omics data (numeric and extremely high dimensional with a pronounced p >> n complication).

show abstract

SurvMetrics: An R package for Predictive Evaluation Metrics in Survival Analysis

Cited by 5 publications

References 39 publications

Feature pre-selection for the development of epigenetic biomarkers

Feature pre-selection for the development of epigenetic biomarkers

Incarceration of Youths in an Adult Correctional Facility and Risk of Premature Death

Review: Deep Learning-Based Survival Analysis of Omics and Clinicopathological Data

Contact Info

Product

Resources

About