Synthesizing multidimensional clinical profiles from published Kaplan-Meier images

Synthesizing multidimensional clinical profiles from published Kaplan–Meier images | medRxiv Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Synthesizing multidimensional clinical profiles from published Kaplan–Meier images View ORCID Profile Zheqing Zhu , View ORCID Profile Fangfang Shen , Yuhua Qian , Jieting Wang doi: https://doi.org/10.64898/2026.03.17.26348584 Zheqing Zhu 1 Institute of Big Data Science and Industry, Shanxi University , Taiyuan, Shanxi, China 2 Shanxi Key Laboratory of Evolutionary Science Intelligence , Taiyuan, Shanxi, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Zheqing Zhu Fangfang Shen 3 Shanxi Hospital Affiliated to Cancer Hospital, Chinese Academy of Medical Sciences , Taiyuan, Shanxi, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Fangfang Shen Yuhua Qian 1 Institute of Big Data Science and Industry, Shanxi University , Taiyuan, Shanxi, China 2 Shanxi Key Laboratory of Evolutionary Science Intelligence , Taiyuan, Shanxi, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: jinchengqyh{at}sxu.edu.cn Jieting Wang 1 Institute of Big Data Science and Industry, Shanxi University , Taiyuan, Shanxi, China 2 Shanxi Key Laboratory of Evolutionary Science Intelligence , Taiyuan, Shanxi, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF ABSTRACT Strict privacy regulations restrict access to individual-level clinical trial data, confining secondary evidence synthesis to isolated one-dimensional Kaplan-Meier summaries. Inferring unreported joint distributions from these independent marginals constitutes an ill-posed inverse problem that limits multivariable precision oncology. Conventional algorithms default to the maximum entropy (MaxEnt) prior, which forces uncalibrated conditional independence upon highly collinear variables. This dismantles inherent informational dependencies, yielding unrealistic, “off-manifold” profiles that fundamentally misspecify the data-generating process and distort multivariable credit allocation. Here, we introduce MD-JoPiGo, an open-source computational framework that overcomes this structural unidentifiability by constraining high-dimensional integration within deterministic Fréchet bounds. Drawing on partial identification principles, the algorithm employs quadratically penalized maximum entropy combined with simulated annealing to anchor point estimations within the theoretically observable Fréchet gap. Through causal topology simulations, we demonstrate that the uncalibrated MaxEnt prior suffices for weakly correlated predictors ( r ≤ 0.4) but requires structural calibration to prevent systematic error under high collinearity ( r ≥ 0.6). This topological constraint functions as a variance firewall, ensuring that structural uncertainty and prognostic misallocation remain strictly localized. Extensive Monte Carlo simulations and empirical benchmark analyses—including lung cancer ( n = 228), stage III colon cancer ( N = 929), and CheckMate 227—demonstrate that the framework successfully recovers latent intersectional treatment effects. By unifying topological inference with data generation, this method translates fragmented public evidence into realistic in silico cohorts, supplying the multivariable resolution required for cross-study benchmarking and privacy-preserving Synthetic Control Arm (SCA) construction. Competing Interest Statement The authors have declared no competing interest. Funding Statement This work was supported by the National Natural Science Foundation of China Major Project (Grant No. T2495251) and the Technology Innovation Teams of Shanxi Province (Grant No. 202304051001001). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors and their institution at no time received payment or services from a third party for any aspect of the submitted work. Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: All source data used in this study were openly available before the initiation of the work. The empirical datasets for algorithm validation were obtained from the R "survival" package (v3.5-5), specifically the NCCTG lung cancer dataset ('lung') and the Moertel colorectal cancer trial dataset ('colon'). These are accessible via the Comprehensive R Archive Network (CRAN) at: . The CheckMate 227 trial data used for the real-world application were reconstructed from Kaplan-Meier curves published in the following peer-reviewed articles: Hellmann et al. (N Engl J Med 2018; 378:1655-1667; DOI: 10.1056/NEJMoa1801946) Paz-Ares et al. (J Thorac Oncol. 2022; 17(1):97-108; DOI: 10.1016/j.jtho.2021.09.010) I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Footnotes Revised the conceptual framework to formally define the MD-JoPiGo synthesis as a deterministically bounded inverse problem. Expanded the causal topology sensitivity analysis to include mediation and collider stratification (Berkson's paradox). Furthermore, established the operational boundaries of the MaxEnt prior under varying marginal asymmetry using Frechet inequalities, demonstrating the "variance firewall" mechanism that ensures robust multivariable reconstruction. Data Availability The MD-JoPiGo framework and its validation testing datasets are openly available at https://github.com/ZheqingZhu/mdjopigo . The empirical datasets used in this study (NCCTG lung cancer and Moertel colorectal cancer) are publicly accessible via the R 'survival' package ( https://CRAN.R-project.org/package=survival ). Copyright The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license . Back to top Previous Next Posted April 02, 2026. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Synthesizing multidimensional clinical profiles from published Kaplan–Meier images Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Synthesizing multidimensional clinical profiles from published Kaplan–Meier images Zheqing Zhu , Fangfang Shen , Yuhua Qian , Jieting Wang medRxiv 2026.03.17.26348584; doi: https://doi.org/10.64898/2026.03.17.26348584 Share This Article: Copy Citation Tools Synthesizing multidimensional clinical profiles from published Kaplan–Meier images Zheqing Zhu , Fangfang Shen , Yuhua Qian , Jieting Wang medRxiv 2026.03.17.26348584; doi: https://doi.org/10.64898/2026.03.17.26348584 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Oncology Subject Areas All Articles Addiction Medicine (638) Allergy and Immunology (917) Anesthesia (349) Cardiovascular Medicine (4985) Dentistry and Oral Medicine (493) Dermatology (426) Emergency Medicine (675) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1688) Epidemiology (16236) Forensic Medicine (36) Gastroenterology (1240) Genetic and Genomic Medicine (7264) Geriatric Medicine (751) Health Economics (1093) Health Informatics (5276) Health Policy (1464) Health Systems and Quality Improvement (1829) Hematology (599) HIV/AIDS (1379) Infectious Diseases (except HIV/AIDS) (16475) Intensive Care and Critical Care Medicine (1200) Medical Education (698) Medical Ethics (156) Nephrology (749) Neurology (7571) Nursing (377) Nutrition (1124) Obstetrics and Gynecology (1295) Occupational and Environmental Health (1038) Oncology (3739) Ophthalmology (1090) Orthopedics (407) Otolaryngology (465) Pain Medicine (499) Palliative Medicine (143) Pathology (720) Pediatrics (1854) Pharmacology and Therapeutics (754) Primary Care Research (785) Psychiatry and Clinical Psychology (6113) Public and Global Health (9971) Radiology and Imaging (2522) Rehabilitation Medicine and Physical Therapy (1511) Respiratory Medicine (1268) Rheumatology (658) Sexual and Reproductive Health (801) Sports Medicine (595) Surgery (801) Toxicology (109) Transplantation (318) Urology (307)

Related documents

• Pathomolecular Analysis of Rare EGFR Mutations in Advanced NSCLC#895085
• Mitochondrial Video Assessment Source Material Study#876182
• Clinically integrated multi-modal transformer framework with cross-modal gated fusion and clinical nomogram for automated Kellgren-Lawrence grading of knee osteoarthritis on x-ray images.#297319
• MHub.ai: A Standardized Platform for Reproducible AI Research in Medical Imaging#387685
• Development of a cfDNA 5mC/5hmC-based Biomarker Panel to Predict Targeted Therapy Efficacy in mCRC#836824