Development and assessment of novel assist device for cardiac cannulation.

PURPOSE: Congenital heart defects are the most common birth defects in the USA and in 25% of cases need to be treated with cardiovascular interventions. One of such interventions is the postoperative use of an extracorporeal membrane oxygenation (ECMO) machine for the treatment of cardiorespiratory failure. The process of placing the patient on the ECMO is extremely time-critical and requires the use of cardiac cannulation. For the first time, our team developed and evaluated a new quick-connect cannulation system that allows for rapid, easy, and safe ECMO cannulation in the pediatric population. The design should eliminate the need for purse-string sutures that are currently used to secure cannulas, as the cannulas will be inserted through a port that is glued to the cardiovascular tissue. METHODS: The rapid cannulation assistance device was designed on the SolidWorks computer-aided design software using the dimensions of the commercially available arterial and venous catheters. These designs were then 3D printed, and tensile testing was performed. Then, a flow loop was developed, and cannulation was performed and analyzed on both 3D-printed hearts and porcine hearts. RESULTS: The rapid cannulation assistance device was designed and 3D printed. Tensile testing found that the parts were strong enough to withstand forces that may be introduced in studies. 3D-printed and porcine heart tests with a flow loop found no leakage with the 3D-printed hearts but minimal leaking with the porcine hearts. However, this leakage was observed at the junction between the device and the heart, leading us to believe that a glue better suited to attach the device to the heart would prevent leakage in the future. CONCLUSIONS: This project successfully demonstrated how a rapid cannulation assistance device could be developed and tested. Future studies will be conducted that address device adhesion to the cardiovascular tissue so that accurate pressure and flow rates can be measured. Future studies will also include testing the device in a fluid environment to more effectively analyze the device success and comparing the time required to cannulate using our device compared to the standard of care.

A methodology for the annotation of surgical videos for supervised machine learning applications.

PURPOSE: Surgical data science is an emerging field focused on quantitative analysis of pre-, intra-, and postoperative patient data (Maier-Hein et al. in Med Image Anal 76: 102306, 2022). Data science approaches can decompose complex procedures, train surgical novices, assess outcomes of actions, and create predictive models of surgical outcomes (Marcus et al. in Pituitary 24: 839-853, 2021; Røadsch et al. in Nat Mach Intell, 2022). Surgical videos contain powerful signals of events that may impact patient outcomes. A necessary step before the deployment of supervised machine learning methods is the development of labels for objects and anatomy. We describe a complete method for annotating videos of transsphenoidal surgery. METHODS: Endoscopic video recordings of transsphenoidal pituitary tumor removal surgeries were collected from a multicenter research collaborative. These videos were anonymized and stored in a cloud-based platform. Videos were uploaded to an online annotation platform. Annotation framework was developed based on a literature review and surgical observations to ensure proper understanding of the tools, anatomy, and steps present. A user guide was developed to trained annotators to ensure standardization. RESULTS: A fully annotated video of a transsphenoidal pituitary tumor removal surgery was produced. This annotated video included over 129,826 frames. To prevent any missing annotations, all frames were later reviewed by highly experienced annotators and a surgeon reviewer. Iterations to annotated videos allowed for the creation of an annotated video complete with labeled surgical tools, anatomy, and phases. In addition, a user guide was developed for the training of novice annotators, which provides information about the annotation software to ensure the production of standardized annotations. CONCLUSIONS: A standardized and reproducible workflow for managing surgical video data is a necessary prerequisite to surgical data science applications. We developed a standard methodology for annotating surgical videos that may facilitate the quantitative analysis of videos using machine learning applications. Future work will demonstrate the clinical relevance and impact of this workflow by developing process modeling and outcome predictors.