Predicting the quality of surgical exposure using spatial and procedural features from laparoscopic videos

Abstract

Evaluating the quality of surgical procedures is a major concern in minimal invasive surgeries. We propose a bottom-up approach based on the study of Sleeve Gastrectomy procedures, for which we analyze what we assume to be an important indicator of the surgical expertise: the exposure of the surgical scene. We first aim at predicting this indicator with features extracted from the laparoscopic video feed. Second, to analyze how the extracted features describing the surgical practice influence this indicator. Method 29 patients underwent Sleeve Gastrectomy performed by two confirmed surgeons in a monocentric study. Features were extracted from spatial and procedural annotations of the videos, and an expert surgeon evaluated the quality of the surgical exposure at specific instants. The features were used as input of a classifier (linear discriminant analysis followed by a support vector machine) to predict the expertise indicator. Features selected in different configurations of the algorithm were compared to understand their relationships with the surgical exposure and the surgeon's practice. Results The optimized algorithm giving the best performance used spatial features as input (Acc = 0.68, Sn = 0.72, Sp = 0.7). It also predicted equally the two classes of the indicator, despite their strong imbalance. Analyzing the selection of input features in the algorithm allowed a comparison of different configurations of the algorithm and showed a link between the surgical exposure and the surgeon's practice.

Current and Potential Applications of Artificial Intelligence in Metabolic Bariatric Surgery

“…AI could extract quantitative data from video with an accuracy of >85%, a feature that allows quantification of operative capacity and objective evaluation for the purposes of both training and self-development. Similarly, Derathé et al utilized annotated spatial and procedural data and processed them with SVM in order to predict surgical exposure [99].…”

Explaining a model predicting quality of surgical practice: a first presentation to and review by clinical experts

“…In [1], we predicted two facets of surgical practice during laparoscopic sleeve gastrectomy. The first was the Exposure of the Surgical Scene (ESS ), defined as the accessibility and visibility of the surgical target during the intervention.…”

“…The prediction performance of our model informs us on the level of confidence we can accord to the estimation of the importance vector I V . Table 2 presents the prediction performance for the complete clusters C ESS and C SP P (lines 1&2) taken from [1], and for each sub-cluster defined in Section 3.1. The number of input samples in each cluster is also provided.…”

“…In a previous study [1], we developed a prediction model, which was used to investigate two specific indicators of surgical practice: how well the surgeon exposes the surgical target in the scene (ESS), and the profile of practice (SPP). Based on this work (briefly summarized in Section 2), we now wished to address the explicability question by proposing a proof-of-concept framework (3.2) through which to interpret the model's prediction results.…”

Multiple instance convolutional neural network for gallbladder assessment from laparoscopic images

“…13 In 14 a computer vision technique was applied to surgical images for performance evaluation on the dissection task. In, 15 visual features extracted from manual annotations of still video frames were employed to assess surgical exposure.…”