Venous thromboembolism (VTE), including deep vein thrombosis and pulmonary embolism, is a life-threatening complication in patients with cervical cancer that can significantly worsen survival and treatment outcomes. Although several clinical risk assessment tools are available, they primarily rely on laboratory measurements obtained at a single time point, limiting their ability to capture the evolving biological processes that precede thrombosis. While deep learning models have demonstrated strong predictive performance, their limited interpretability and high computational demands remain barriers to routine clinical implementation.

In a recent study published in Genes & Diseases, researchers from Shanghai Jiao Tong University and Chongqing University Cancer Hospital developed TempVTE, a lightweight and interpretable artificial intelligence framework that estimates VTE risk by analyzing temporal changes in routinely measured blood biomarkers. Using longitudinal clinical data from 1,427 post-treatment cervical cancer patients, including 91 patients who developed VTE, the researchers transformed repeated laboratory measurements into temporal feature vectors representing each biomarker's initial value, final value, and overall trajectory over time.

To identify the most informative biomarkers, the team first applied a pretrained deep learning model to evaluate feature importance across coagulation, immune, and hematologic parameters. Coagulation markers, including D-dimer and fibrinogen degradation products (FDP), together with immune indicators such as CD4, CD8, and the CD4/CD8 ratio, emerged as the strongest predictors of VTE development. Rather than relying solely on static laboratory values, TempVTE incorporated the temporal behavior of these biomarkers, enabling more informative risk estimation.

When compared with established clinical scoring systems and conventional machine learning models, TempVTE achieved the highest predictive performance, with an AUC of 0.924 and an accuracy of 0.858. The model also outperformed time-aware survival models while requiring substantially less computation than sequential deep learning approaches, making it well suited for rapid clinical decision-making. Importantly, coefficient analyses demonstrated that rising D-dimer, FDP, and CD8 levels, together with declining hemoglobin concentrations, were strongly associated with increased VTE risk, highlighting the importance of monitoring biomarker dynamics rather than isolated measurements.

Overall, this study demonstrates that incorporating longitudinal biomarker trajectories can substantially improve VTE risk estimation in cervical cancer patients. By combining high predictive accuracy with model transparency and computational efficiency, TempVTE provides a practical framework for dynamic thrombosis risk assessment while offering new insights into the interplay between coagulation and immune responses during cancer-associated thrombosis.

Reference

Title of Original Paper: Artificial intelligence-based dynamic modeling for venous thromboembolism risk estimation in post-treatment cervical cancer patients
Journal: Genes & Diseases
Genes & Diseases is a journal for molecular and translational medicine. The journal primarily focuses on publishing investigations on the molecular bases and experimental therapeutics of human diseases. Publication formats include full length research article, review article, short communication, correspondence, perspectives, commentary, views on news, and research watch.
DOI: https://doi.org/10.1016/j.gendis.2026.102287

Funding Information:
None.

# # # # # #

Genes & Diseases publishes rigorously peer-reviewed and high quality original articles and authoritative reviews that focus on the molecular bases of human diseases. Emphasis is placed on hypothesis-driven, mechanistic studies relevant to pathogenesis and/or experimental therapeutics of human diseases. The journal has worldwide authorship, and a broad scope in basic and translational biomedical research of molecular biology, molecular genetics, and cell biology, including but not limited to cell proliferation and apoptosis, signal transduction, stem cell biology, developmental biology, gene regulation and epigenetics, cancer biology, immunity and infection, neuroscience, disease-specific animal models, gene and cell-based therapies, and regenerative medicine.

Scopus Cite Score: 10.4 | Impact Factor: 14.6

# # # # # #

More information: https://www.keaipublishing.com/en/journals/genes-and-diseases/
Editorial Board: https://www.keaipublishing.com/en/journals/genes-and-diseases/editorial-board/
All issues and articles in press are available online in ScienceDirect (https://www.sciencedirect.com/journal/genes-and-diseases).
Submissions to Genes & Diseases may be made using Editorial Manager (https://www.editorialmanager.com/gendis/default.aspx).
Print ISSN: 2352-4820
eISSN: 2352-3042
CN: 50-1221/R
Contact Us: editor@genesndiseases.cn
X (formerly twitter): @GenesNDiseases (https://x.com/GenesNDiseases)

# # # # # #