StochIKM: Automated High-Fidelity Reconstruction of Individual Patient Data from Published Kaplan-Meier Curves
Abstract
Individual patient data are the gold standard for secondary analysis of time-to-event evidence, but they are rarely released, and reconstruction from a published Kaplan-Meier figure could help. Existing methods all claim high fidelity, each judged by whichever agreements it happens to achieve, and each returns a single dataset although the map from figure to data is not bijective. We define fidelity explicitly: a dataset is admissible when it reproduces the risk table exactly, follows the digitized curve to within the image resolution, and respects the printed median, interval and event count. We propose StochIKM, a stochastic algorithm that digitizes the figure with a vision-capable language model and uses the high-fidelity definition as constraints, returning an admissible set of individual patient datasets. A solution is guaranteed to exist whenever the figure has been read correctly, so a failure indicts either the published figure or its digitization. In simulation and in 79 arms from published oncology trials, StochIKM reproduces the published risk tables and medians more closely than LLM-improved IPDfromKM, a high-impact alternative.
Related
- Mask Reconstruction via Kernel-Based Surrogate Modeling for Automated Circulator Design
- DSReg: Provably Recovering Individual World Latents without Reconstruction
- A partial post-review by a peer of the peer-reviewed article [1] "Modeling COVID-19 transmission: effects of age structure and vaccination, published in 2025 in the Springer Nature journal \it{\bf Scientific Reports}
- Undocumented Behavior in the gsynth R package and its Consequences for Three Published Studies
- Variability among large language models in assessing CONSORT compliance of published randomized clinical trials
BibTeX
@misc{ke2026stochikm,
title = {{StochIKM: Automated High-Fidelity Reconstruction of Individual Patient Data from Published Kaplan-Meier Curves}},
author = {Liangze Ke and Daniel Zelterman and Wei Wei and Guannan Gong},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.29122},
eprint = {2610.29122},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.29122},
}
RIS
TY - UNPB
TI - StochIKM: Automated High-Fidelity Reconstruction of Individual Patient Data from Published Kaplan-Meier Curves
AU - Liangze Ke
AU - Daniel Zelterman
AU - Wei Wei
AU - Guannan Gong
PY - 2026
DA - 2026/10/22
PB - arXiv
DO - 10.48550/arxiv.2610.29122
UR - https://arxiv.org/abs/2610.29122
AB - Individual patient data are the gold standard for secondary analysis of time-to-event evidence, but they are rarely released, and reconstruction from a published Kaplan-Meier figure could help. Existing methods all claim high fidelity, each judged by whichever agreements it happens to achieve, and each returns a single dataset although the map from figure to data is not bijective. We define fidelity explicitly: a dataset is admissible when it reproduces the risk table exactly, follows the digitized curve to within the image resolution, and respects the printed median, interval and event count. We propose StochIKM, a stochastic algorithm that digitizes the figure with a vision-capable language model and uses the high-fidelity definition as constraints, returning an admissible set of individual patient datasets. A solution is guaranteed to exist whenever the figure has been read correctly, so a failure indicts either the published figure or its digitization. In simulation and in 79 arms from published oncology trials, StochIKM reproduces the published risk tables and medians more closely than LLM-improved IPDfromKM, a high-impact alternative.
ER -