Professor and Chair of Surgical Disciplines Central Michigan University College of Medicine Saginaw, Michigan
Objectives: Complications after EVAR can be fatal. Patient follow-up for surveillance imaging is becoming more challenging as fewer patients are seen, particularly after the first year. The aim of this study was to develop an artificial intelligence model to predict the complication probability of individual patients to better identify those needing more intensive surveillance. Complications after EVAR can be fatal. Patient follow-up for surveillance imaging is becoming more challenging as fewer patients are seen, particularly after the first year. The aim of this study was to develop an artificial intelligence model to predict the complication probability of individual patients to better identify those needing more intensive surveillance.
Methods: Pre-operative CTA 3D reconstruction images of AAA from 273 patients who underwent EVAR from 2011-2020 were collected (Table). Of these, 48 patients had postoperative complications including endoleak, AAA rupture, graft limb occlusion, renal artery occlusion, neck dilation, pelvic ischemia, and graft migration. A deep convolutional neural network model utilized the 3D CT images to predict the risk of complications after EVAR (Figure). The model was built with Tensorflow software and run on the Google Colab Platform. A training subset of randomly selected 40 patients with complications and 189 without were used to train the AI model and the remaining 8 positive and 36 negative cases tested its performance and prediction accuracy. Data down-sampling was used to alleviate data imbalance and data augmentation methodology to further boost the model performance.
Results: Successful training was completed on the 229 cases in the training set and then applied to predict the complication probability of each individual in the held-out performance testing cases. The model provided a complication sensitivity of 100% and identified all the patients who later developed complications after EVAR. Of 36 patients without complications, 16 (44%) were falsely predicted to develop complications. The results, therefore, demonstrated excellent sensitivity for identifying patients who would benefit from more stringent surveillance and also alleviate the need for surveillance in 56% of patients unlike to develop complications.
Conclusions: AI models can be developed to predict the risk of post-op complications with high accuracy. Compared to existing methods, the model developed in this study did not require any expert-annotated data but only the AAA CTA images as inputs. This model can play an assistive role in identifying all patients at high risk for post-EVAR complications and the need for greater compliance in surveillance.
