ClipAssistNet: bringing real-time safety feedback to operating rooms

Abstract

Purpose Cholecystectomy is one of the most common laparoscopic procedures. A critical phase of laparoscopic cholecystectomy consists in clipping the cystic duct and artery before cutting them. Surgeons can improve the clipping safety by ensuring full visibility of the clipper, while enclosing the artery or the duct with the clip applier jaws. This can prevent unintentional interaction with neighboring tissues or clip misplacement. In this article, we present a novel real-time feedback to ensure safe visibility of the instrument during this critical phase. This feedback incites surgeons to keep the tip of their clip applier visible while operating. Methods We present a new dataset of 300 laparoscopic cholecystectomy videos with frame-wise annotation of clipper tip visibility. We further present ClipAssistNet, a neural network-based image classifier which detects the clipper tip visibility in single frames. ClipAssistNet ensembles predictions from 5 neural networks trained on different subsets of the dataset. Results Our model learns to classify the clipper tip visibility by detecting its presence in the image. Measured on a separate test set, ClipAssistNet classifies the clipper tip visibility with an AUROC of 0.9107, and 66.15% specificity at 95% sensitivity. Additionally, it can perform real-time inference (16 FPS) on an embedded computing board; this enables its deployment in operating room settings. Conclusion This work presents a new application of computer-assisted surgery for laparoscopic cholecystectomy, namely real-time feedback on adequate visibility of the clip applier. We believe this feedback can increase surgeons’ attentiveness when departing from safe visibility during the critical clipping of the cystic duct and artery.

Computer vision in surgery: from potential to clinical value

“…While automated confirmation of the CVS can provide the surgeon with additional assurance of anatomy, other CV tools can ensure that clips are well placed, and no other structures are inadvertently being clipped. To provide such assistance, Aspart et al recently proposed ClipAssistNet, a neural network trained to detect the tips of a clip applier during LC 46 . If experienced surgeons may find such assistance unnecessary and even trivial, trainees and early career surgeons may benefit from the reassurance that can be provided by real-time decision-support algorithms such as GoNoGoNet, DeepCVS, and ClipAssistNet.…”

Der digitale OP-Saal

“…Kürzlich zeigte eine internationale Studie außerdem, dass KI und maschinelles Lernen eine große Hilfestellung bei der Laparoskopie bieten könnten. "ClipAssist-Net" wurde hier als Echtzeit-Feedback-Mechanismus vorgestellt [16]. Das Programm kann während laparoskopischer Cholezystektomien erkennen, ob der Clip vor Absetzen des Gallengangs unter Sicht gesetzt wurde.…”

“…Das Programm kann während laparoskopischer Cholezystektomien erkennen, ob der Clip vor Absetzen des Gallengangs unter Sicht gesetzt wurde. Waren die Branchenenden der Clipzange nicht sichtbar, so konnte das Programm den Chirurgen darauf hinweisen [16].…”

“…Durch ein Verknüpfen mit kontinuierlicher Auswertung von erhobenen Videodaten und das Verlinken dieser Daten mit dem intra-und postoperativen Verlauf gelingt es, mit Maschinellem Lernen Algorithmen zu entwickeln, die bereits bei der eigentlichen Operation potenzielle Fehler automatisch erkennen und Warnsignale senden, bevor sie überhaupt entstehen. Erste vielversprechende Daten hierfür lieferten Aspart und Kollegen in der bereits beschriebenen ClipAssistNet-Studie [16].…”

Quantification of Robotic Surgeries with Vision-Based Deep Learning

“…Another line of research has instead focused on exclusively exploiting live surgical videos from endoscopic cameras to classify surgical activity 16 , gestures 17,18 , and skills 19 , among other tasks [20][21][22] . Most recently, attention-based neural networks such as Transformers 23 have been used to distinguish between distinct surgical steps within a procedure [24][25][26][27] .…”