Department of Organ Surgery and Transplantation, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark.
Center for Surgical Translational and Artificial Intelligence Research (CSTAR), Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark.
PLoS One. 2024 Jul 15;19(7):e0294368. doi: 10.1371/journal.pone.0294368. eCollection 2024.
Postoperative complications affect up to 15% of surgical patients constituting a major part of the overall disease burden in a modern healthcare system. While several surgical risk calculators have been developed, none have so far been shown to decrease the associated mortality and morbidity. Combining deep neural networks and genomics with the already established clinical predictors may hold promise for improvement.
The UK Biobank was utilized to build linear and deep learning models for the prediction of surgery relevant outcomes. An initial GWAS for the relevant outcomes was initially conducted to select the Single Nucleotide Polymorphisms for inclusion in the models. Model performance was assessed with Receiver Operator Characteristics of the Area Under the Curve and optimum precision and recall. Feature importance was assessed with SHapley Additive exPlanations.
Models were generated for atrial fibrillation, venous thromboembolism and pneumonia as genetics only, clinical features only and a combined model. For venous thromboembolism, the ROC-AUCs were 60.1% [59.6%-60.4%], 63.4% [63.2%-63.4%] and 66.6% [66.2%-66.9%] for the linear models and 51.5% [49.4%-53.4%], 63.2% [61.2%-65.0%] and 62.6% [60.7%-64.5%] for the deep learning SNP, clinical and combined models, respectively. For atrial fibrillation, the ROC-AUCs were 60.3% [60.0%-60.4%], 78.7% [78.7%-78.7%] and 80.0% [79.9%-80.0%] for the linear models and 59.4% [58.2%-60.9%], 78.8% [77.8%-79.8%] and 79.8% [78.8%-80.9%] for the deep learning SNP, clinical and combined models, respectively. For pneumonia, the ROC-AUCs were 50.1% [49.6%-50.6%], 69.2% [69.1%-69.2%] and 68.4% [68.0%-68.5%] for the linear models and 51.0% [49.7%-52.4%], 69.7% [.5%-70.8%] and 69.7% [68.6%-70.8%] for the deep learning SNP, clinical and combined models, respectively.
In this report we presented linear and deep learning predictive models for surgery relevant outcomes. Overall, predictability was similar between linear and deep learning models and inclusion of genetics seemed to improve accuracy.
术后并发症影响多达 15%的手术患者，构成现代医疗体系中整体疾病负担的主要部分。虽然已经开发出几种手术风险计算器，但迄今为止，没有一种能够降低相关死亡率和发病率。将深度学习神经网络与基因组学相结合，并结合已经建立的临床预测因素，可能有希望得到改善。
利用英国生物库建立线性和深度学习模型，以预测与手术相关的结果。最初进行了全基因组关联研究（GWAS），以选择与相关结果相关的单核苷酸多态性（SNP）纳入模型。使用曲线下面积的接收者操作特性（ROC-AUC）和最佳精度和召回率评估模型性能。使用 Shapley Additive exPlanations 评估特征重要性。
为心房颤动、静脉血栓栓塞和肺炎生成了仅基于遗传学、仅基于临床特征和综合模型的模型。对于静脉血栓栓塞，线性模型的 ROC-AUC 分别为 60.1%[59.6%-60.4%]、63.4%[63.2%-63.4%]和 66.6%[66.2%-66.9%]，深度学习 SNP、临床和综合模型的 ROC-AUC 分别为 51.5%[49.4%-53.4%]、63.2%[61.2%-65.0%]和 62.6%[60.7%-64.5%]。对于心房颤动，线性模型的 ROC-AUC 分别为 60.3%[60.0%-60.4%]、78.7%[78.7%-78.7%]和 80.0%[79.9%-80.0%]，深度学习 SNP、临床和综合模型的 ROC-AUC 分别为 59.4%[58.2%-60.9%]、78.8%[77.8%-79.8%]和 79.8%[78.8%-80.9%]。对于肺炎，线性模型的 ROC-AUC 分别为 50.1%[49.6%-50.6%]、69.2%[69.1%-69.2%]和 68.4%[68.0%-68.5%]，深度学习 SNP、临床和综合模型的 ROC-AUC 分别为 51.0%[49.7%-52.4%]、69.7%[.5%-70.8%]和 69.7%[68.6%-70.8%]。
在本报告中，我们提出了与手术相关结果的线性和深度学习预测模型。总体而言，线性和深度学习模型的可预测性相似，并且纳入遗传学似乎可以提高准确性。
