Development and Validation of Image-Based Deep Learning Models to Predict Surgical Complexity and Complications in Abdominal Wall Reconstruction

Abstract

Image-based deep learning models (DLMs) have been used in other disciplines, but this method has yet to be used to predict surgical outcomes.OBJECTIVE To apply image-based deep learning to predict complexity, defined as need for component separation, and pulmonary and wound complications after abdominal wall reconstruction (AWR). DESIGN, SETTING, AND PARTICIPANTSThis quality improvement study was performed at an 874-bed hospital and tertiary hernia referral center from September 2019 to January 2020. A prospective database was queried for patients with ventral hernias who underwent open AWR by experienced surgeons and had preoperative computed tomography images containing the entire hernia defect. An 8-layer convolutional neural network was generated to analyze image characteristics. Images were batched into training (approximately 80%) or test sets (approximately 20%) to analyze model output. Test sets were blinded from the convolutional neural network until training was completed. For the surgical complexity model, a separate validation set of computed tomography images was evaluated by a blinded panel of 6 expert AWR surgeons and the surgical complexity DLM. Analysis started February 2020. EXPOSURES Image-based DLM. MAIN OUTCOMES AND MEASURESThe primary outcome was model performance as measured by area under the curve in the receiver operating curve (ROC) calculated for each model; accuracy with accompanying sensitivity and specificity were also calculated. Measures were DLM prediction of surgical complexity using need for component separation techniques as a surrogate and prediction of postoperative surgical site infection and pulmonary failure. The DLM for predicting surgical complexity was compared against the prediction of 6 expert AWR surgeons.RESULTS A total of 369 patients and 9303 computed tomography images were used. The mean (SD) age of patients was 57.9 (12.6) years, 232 (62.9%) were female, and 323 (87.5%) were White. The surgical complexity DLM performed well (ROC = 0.744; P < .001) and, when compared with surgeon prediction on the validation set, performed better with an accuracy of 81.3% compared with 65.0% (P < .001). Surgical site infection was predicted successfully with an ROC of 0.898 (P < .001). However, the DLM for predicting pulmonary failure was less effective with an ROC of 0.545 (P = .03).CONCLUSIONS AND RELEVANCE Image-based DLM using routine, preoperative computed tomography images was successful in predicting surgical complexity and more accurate than expert surgeon judgment. An additional DLM accurately predicted the development of surgical site infection.

The Development of Artificial Intelligence in Hernia Surgery: A Scoping Review

“…In another study, López-Cano et al argued that the incorporation of artificial intelligence techniques could improve the delivery of patient care by combining and analyzing thousands of abdominal hernia cases and classifying them based on severity and prioritization ( 20 ). Artificial intelligence in abdominal hernia surgery is evident via the research conducted by Elhage et al, which aimed to apply three deep learning models to predict complexity and wound infections after an abdominal hernia procedure ( 21 ). The study outcomes demonstrated that the three image-based models, through computed tomography images, were effective in predicting surgical complexity and more accurate than expert surgeon judgment ( 21 ).…”

“…Artificial intelligence in abdominal hernia surgery is evident via the research conducted by Elhage et al, which aimed to apply three deep learning models to predict complexity and wound infections after an abdominal hernia procedure ( 21 ). The study outcomes demonstrated that the three image-based models, through computed tomography images, were effective in predicting surgical complexity and more accurate than expert surgeon judgment ( 21 ). These arguments imply that AI has the potential of enhancing multiple processes involved in abdominal hernia surgery.…”

Telemedicine-based new patient consultations for hernia repair and advanced abdominal wall reconstruction

“…The decision to provide audio-only evaluations was further bolstered by a large volume of data related to the predictive nature of cross-sectional imaging for abdominal wall reconstruction [ 20 – 22 ]. In particular, recent work by Elhage et al have demonstrated that findings on cross-sectional imaging are predictive of need for advanced reconstruction techniques (e.g., myofascial advancement flaps), as well as postoperative outcomes related to surgical site infections [ 23 ]. As such, we have broadened our options for preoperative telemedicine evaluations to include audio-only options for patients presenting for ventral or inguinal hernia repair consultations that imaging-based confirmation available for review at the time of consultation.…”

Spotlight in Plastic Surgery: April 2022

“…2021;156:933-940. 5 The authors applied image-based deep learning to predict complexity, defined as the need for component separation, and pulmonary and wound complications after abdominal wall reconstruction. They concluded that image-based deep-learning models using routine, preoperative computed tomography images were successful in predicting surgical complexity and more accurate than expert surgeon judgment.…”