What question did this study set out to answer?

Develop and validate a time-inhomogeneous Markov model using real-world electronic medical record data to estimate 10-year disease progression and mortality in patients with prostate cancer.

September 18, 2026iScienceOpen Access

Predicting disease progression and mortality in prostate cancer using real-world data-driven time-inhomogeneous Markov models

View Full Paper

Ask AI

Bookmark

Authors

JWJiaqi WangHebei University of Engineering YJYuanshi JiaoChinese University of Hong Kong DCDawn CraigNIHR Newcastle Biomedical Research Centre

Discussion

Member takes

Implication

Cohort study demonstrates accurate 10-year progression and mortality predictions in prostate cancer, highlighting the value of dynamic Markov models for long-term risk assessment.

Key Points

Develop and validate a time-inhomogeneous Markov model using real-world electronic medical record data to estimate 10-year disease progression and mortality in patients with prostate cancer.
Trained a time-inhomogeneous Markov model using electronic medical records from 3,274 newly diagnosed prostate cancer patients (2010–2012) in Hong Kong.
Defined dynamic health states using baseline covariates, including age, Charlson Comorbidity Index, and prostate-specific antigen levels.
Validated predictive performance against Cox proportional hazards and random survival forest models using an independent 2013 patient cohort.
Estimated 5-year and 10-year mortality rates of 16.9% and 28.0%, respectively, for mild cases defined by age ≤65, CCI=0, and PSA ≤4 ng/mL.
Demonstrated strong discriminative ability in the independent validation cohort for metastasis-free survival (C-index: 0.780; 95% CI: 0.723–0.837) and overall survival (C-index: 0.787; 95% CI: 0.731–0.843), outperforming benchmark models.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/6aad0bb6de0393d728b8a214 https://doi.org/10.1016/j.isci.2026.117536

View Full Paper

Ask AI

Bookmark

Implication

Cohort study demonstrates accurate 10-year progression and mortality predictions in prostate cancer, highlighting the value of dynamic Markov models for long-term risk assessment.

Predicting disease progression and mortality in prostate cancer using real-world data-driven time-inhomogeneous Markov models

Discussion

Member takes

Implication

Key Points

Cite This Study

Predicting disease progression and mortality in prostate cancer using real-world data-driven time-inhomogeneous Markov models

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider