AI 用於風險分層：多模態 DL 模型為肺栓塞提供增強預後

連結表格

摘要

1. 引言
2. 方法
3. 結果
4. 討論
5. 結論、致謝和參考文獻

5. 結論

基於結合 CTPA 特徵和臨床變數的多組學深度學習模型在肺栓塞死亡率預測方面表現優於單獨使用 PESI 評分。將 PESI 添加到多模態模型中僅顯示出微小的性能改善，說明基於人工智慧的模型已足以進行生存預測。多模態模型在 30 天死亡風險估計中同樣改善了單獨使用 PESI 的表現。通過 NRI 分析，臨床和影像數據被獨立證明都有助於提高多模態模型的性能。這些發現展示了多模態深度學習模型相較於當前臨床標準 PESI 的優勢，將預後轉變為一個整合更多臨床和影像信息的智能過程。此外，我們證明了我們的模型與臨床死亡率指標（如右心室功能障礙）的一致性。進一步分析可以更清楚地揭示各種風險因素與肺栓塞患者死亡率之間的聯繫，以及如何利用這些信息進行生存預測模型開發。然而，我們模型的益處只能通過在更大更多樣化的數據集上進行額外驗證，以及對開發模型進行前瞻性測試來確認。

\ 我們的研究強調了基於深度學習模型在肺栓塞患者預後和風險分層中的實用性。人工智慧有潛力通過提供快速準確的診斷和預後信息來改善放射科醫師和臨床醫生的臨床工作流程。通過為肺栓塞患者提供及時且準確的風險分層，人工智慧可能通過指導臨床決策為患者和醫療提供者帶來實質性益處，潛在地改善患者預後。

致謝

無。

參考文獻

1. Beckman MG, Hooper WC, Critchley SE, Ortel TL. Venous thromboembolism: a public health concern. American journal of preventive medicine. 2010;38(4):S495-S501.

   \

2. Lewis AE, Gerstein NS, Venkataramani R, Ramakrishna H. Evolving management trends and outcomes in catheter management of acute pulmonary embolism. Journal of Cardiothoracic and Vascular Anesthesia. 2022;36(8):3344-3356.

   \

3. Goldhaber SZ, Bounameaux H. Pulmonary embolism and deep vein thrombosis. The Lancet. 2012;379(9828):1835-1846.

   \

4. Piazza G, Goldhaber SZ. Acute pulmonary embolism: part I: epidemiology and diagnosis. Circulation. 2006;114(2):e28-e32.

   \

5. Aujesky D, Obrosky DS, Stone RA, et al. Derivation and validation of a prognostic model for pulmonary embolism. American journal of respiratory and critical care medicine. 2005;172(8):1041-1046.

   \

6. Donzé J, Le Gal G, Fine MJ, et al. Prospective validation of the pulmonary embolism severity index. Thrombosis and haemostasis. 2008;100(05):943-948.

   \

7. Ishwaran H, Kogalur UB, Blackstone EH, Lauer MS. Random survival forests. 2008;

   \

8. Fox J, Weisberg S. Cox proportional-hazards regression for survival data. An R and S-PLUS companion to applied regression. 2002;2002

   \

9. Katzman JL, Shaham U, Cloninger A, Bates J, Jiang T, Kluger Y. DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC medical research methodology. 2018;18(1):1-12.

   \

10. Oren O, Gersh BJ, Bhatt DL. Artificial intelligence in medical imaging: switching from radiographic pathological data to clinically meaningful endpoints. The Lancet Digital Health. 2020;2(9):e486-e488.

    \

11. Weikert T, Winkel DJ, Bremerich J, et al. Automated detection of pulmonary embolism in CT pulmonary angiograms using an AI-powered algorithm. European radiology. 2020;30:6545-6553.

    \

12. Huang S-C, Kothari T, Banerjee I, et al. PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging. NPJ digital medicine. 2020;3(1):61.

    \

13. Huang S-C, Pareek A, Zamanian R, Banerjee I, Lungren MP. Multimodal fusion with deep neural networks for leveraging CT imaging and electronic health record: a case-study in pulmonary embolism detection. Scientific reports. 2020;10(1):22147.

    \

14. Liu W, Liu M, Guo X, et al. Evaluation of acute pulmonary embolism and clot burden on CTPA with deep learning. European radiology. 2020;30:3567-3575.

    \

15. Yao J, Zhu X, Zhu F, Huang J. Deep correlational learning for survival prediction from multi-modality data. Springer; 2017:406-414.

    \

16. Vale-Silva LA, Rohr K. Long-term cancer survival prediction using multimodal deep learning. Scientific Reports. 2021;11(1):13505.

    \

17. Meyer G, Vicaut E, Danays T, et al. Fibrinolysis for patients with intermediate-risk pulmonary embolism. New England Journal of Medicine. 2014;370(15):1402-1411.

    \

18. Hofmanninger J, Prayer F, Pan J, Röhrich S, Prosch H, Langs G. Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem. European Radiology Experimental. 2020;4(1):1-13.

    \

19. Harrell Jr FE, Lee KL, Califf RM, Pryor DB, Rosati RA. Regression modelling strategies for improved prognostic prediction. Statistics in medicine. 1984;3(2):143-152.

    \

20. Leening MJ, Vedder MM, Witteman JC, Pencina MJ, Steyerberg EW. Net reclassification improvement: computation, interpretation, and controversies: a literature review and clinician's guide. Annals of internal medicine. 2014;160(2):122-131.

    \

21. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. Journal of the American statistical association. 1958;53(282):457-481.

    \

22. Grifoni S, Olivotto I, Cecchini P, et al. Short-term clinical outcome of patients with acute pulmonary embolism, normal blood pressure, and echocardiographic right ventricular dysfunction. Circulation. 2000;101(24):2817-2822.

    \

23. Coutance G, Cauderlier E, Ehtisham J, Hamon M, Hamon M. The prognostic value of markers of right ventricular dysfunction in pulmonary embolism: a meta-analysis. Critical care. 2011;15:1-10.

    \

24. Chaudhary K, Poirion OB, Lu L, Garmire LX. Deep learning–based multi-omics integration robustly predicts survival in liver cancer. Clinical Cancer Research. 2018;24(6):1248-1259.

    \

25. Somani SS, Honarvar H, Narula S, et al. Development of a machine learning model using electrocardiogram signals to improve acute pulmonary embolism screening. European Heart Journal-Digital Health. 2022;3(1):56-66.

    \

26. Tourassi GD, Floyd CE, Sostman HD, Coleman RE. Acute pulmonary embolism: artificial neural network approach for diagnosis. Radiology. 1993;189(2):555-558.

    \

27. Soffer S, Klang E, Shimon O, et al. Deep learning for pulmonary embolism detection on computed tomography pulmonary angiogram: a systematic review and meta-analysis. Scientific reports. 2021;11(1):15814.

    \

28. Elias A, Mallett S, Daoud-Elias M, Poggi J-N, Clarke M. Prognostic models in acute pulmonary embolism: a systematic review and meta-analysis. BMJ open. 2016;6(4):e010324.

    \

29. Cahan N, Klang E, Marom EM, et al. Multimodal fusion models for pulmonary embolism mortality prediction. Scientific Reports. 2023;13(1):1-15.

圖表

\

\

\

\

\

\

\ 用於計算每位患者PESI評分的PESI臨床變數的詳細患者特徵。

\ 所有連續變數均報告為中位數(四分位距)，所有分類變數均報告為數量(%)。統計顯著的p值以粗體顯示(p < 0.05)。死亡狀態不是PESI臨床變數。

\ BP = 血壓。PESI = 肺