Deep learning for surgical phase recognition using endoscopic videos.

BACKGROUND: Operating room planning is a complex task as pre-operative estimations of procedure duration have a limited accuracy. This is due to large variations in the course of procedures. Therefore, information about the progress of procedures is essential to adapt the daily operating room schedule accordingly. This information should ideally be objective, automatically retrievable and in real-time. Recordings made during endoscopic surgeries are a potential source of progress information. A trained observer is able to recognize the ongoing surgical phase from watching these videos. The introduction of deep learning techniques brought up opportunities to automatically retrieve information from surgical videos. The aim of this study was to apply state-of-the art deep learning techniques on a new set of endoscopic videos to automatically recognize the progress of a procedure, and to assess the feasibility of the approach in terms of performance, scalability and practical considerations. METHODS: A dataset of 33 laparoscopic cholecystectomies (LC) and 35 total laparoscopic hysterectomies (TLH) was used. The surgical tools that were used and the ongoing surgical phases were annotated in the recordings. Neural networks were trained on a subset of annotated videos. The automatic recognition of surgical tools and phases was then assessed on another subset. The scalability of the networks was tested and practical considerations were kept up. RESULTS: The performance of the surgical tools and phase recognition reached an average precision and recall between 0.77 and 0.89. The scalability tests showed diverging results. Legal considerations had to be taken into account and a considerable amount of time was needed to annotate the datasets. CONCLUSION: This study shows the potential of deep learning to automatically recognize information contained in surgical videos. This study also provides insights in the applicability of such a technique to support operating room planning.

The Art of Electrosurgery: Trainees and Experts.

The benefits of electrosurgery have been acknowledged since the early 1920s, and nowadays more than 80% of surgical procedures involve devices that apply energy to tissues. Despite its widespread use, it is currently unknown how the operator's choices with regard to instrument selection and application technique are related to complications. As such, the manner in which electrosurgery is applied can have a serious influence on the outcome of the procedure and the well-being of patients. The aim of this study is to investigate the variety of differences in usage of electrosurgical devices. Our approach is to measure these parameters to provide insight into application techniques. A sensor was developed that records the magnitude of electric current delivered to an electrosurgical device at a frequency of 10 Hz. The sensor is able to detect device activation times and a reliable estimate of the power-level settings. Data were recorded for 91 laparoscopic cholecystectomies performed by different surgeons and residents. Results of the current measurement data show differences in the way electrosurgery is applied by surgeons and residents during a laparoscopic cholecystectomy. Variations are seen in the number of activations, the activation time, and the approach for removal of the gallbladder. Analysis showed that experienced surgeons have a longer activation time than residents (3.01 vs 1.41 seconds, P < .001) and a lower number of activations (102 vs 123). This method offers the opportunity to relate application techniques to clinical outcome and to provide input for the development of a best practice model.

'It is Time to Prepare the Next patient' Real-Time Prediction of Procedure Duration in Laparoscopic Cholecystectomies.

Operating Room (OR) scheduling is crucial to allow efficient use of ORs. Currently, the predicted durations of surgical procedures are unreliable and the OR schedulers have to follow the progress of the procedures in order to update the daily planning accordingly. The OR schedulers often acquire the needed information through verbal communication with the OR staff, which causes undesired interruptions of the surgical process. The aim of this study was to develop a system that predicts in real-time the remaining procedure duration and to test this prediction system for reliability and usability in an OR. The prediction system was based on the activation pattern of one single piece of equipment, the electrosurgical device. The prediction system was tested during 21 laparoscopic cholecystectomies, in which the activation of the electrosurgical device was recorded and processed in real-time using pattern recognition methods. The remaining surgical procedure duration was estimated and the optimal timing to prepare the next patient for surgery was communicated to the OR staff. The mean absolute error was smaller for the prediction system (14 min) than for the OR staff (19 min). The OR staff doubted whether the prediction system could take all relevant factors into account but were positive about its potential to shorten waiting times for patients. The prediction system is a promising tool to automatically and objectively predict the remaining procedure duration, and thereby achieve optimal OR scheduling and streamline the patient flow from the nursing department to the OR.

Where are my instruments? Hazards in delivery of surgical instruments.

BACKGROUND: Unavailability of instruments is recognised to cause delays and stress in the operating room, which can lead to additional risks for the patients. The aim was to provide an overview of the hazards in the entire delivery process of surgical instruments and to provide insight into how Information Technology (IT) could support this process in terms of information availability and exchange. METHODS: The process of delivery was described according to the Healthcare Failure Mode and Effects Analysis methodology for two hospitals. The different means of information exchange and availability were listed. Then, hazards were identified and further analysed for each step of the process. RESULTS: For the first hospital, 172 hazards were identified, and 23 of hazards were classified as high risk. Only one hazard was considered as 'controlled' (when actions were taken to remove the hazard later in the process). Twenty-two hazards were 'tolerated' (when no actions were taken, and it was therefore accepted that adverse events may occur). For the second hospital, 158 hazards were identified, and 49 of hazards were classified as high risk. Eight hazards were 'controlled' and 41 were 'tolerated'. The means for information exchange and information systems were numerous for both cases, while there was not one system that provided an overview of all relevant information. CONCLUSIONS: The majority of the high-risk hazards are expected to be controlled by the use of IT support. Centralised information and information availability for different parties reduce risks related to unavailability of instruments in the operating room.

Safety status system for operating room devices.

BACKGROUND: Since the increase of the number of technological aids in the operating room (OR), equipment-related incidents have come to be a common kind of adverse events. This underlines the importance of adequate equipment management to improve the safety in the OR. OBJECTIVE: A system was developed to monitor the safety status (periodic maintenance and registered malfunctions) of OR devices and to facilitate the notification of malfunctions. The objective was to assess whether the system is suitable for use in an busy OR setting and to analyse its effect on the notification of malfunctions. METHODS: The system checks automatically the safety status of OR devices through constant communication with the technical facility management system, informs the OR staff real-time and facilitates notification of malfunctions. The system was tested for a pilot period of six months in four ORs of a Dutch teaching hospital and 17 users were interviewed on the usability of the system. RESULTS: The users provided positive feedback on the usability. For 86.6% of total time, the localisation of OR devices was accurate. 62 malfunctions of OR devices were reported, an increase of 12 notifications compared to the previous year. CONCLUSIONS: The safety status system was suitable for an OR complex, both from a usability and technical point of view, and an increase of reported malfunctions was observed. The system eases monitoring the safety status of equipment and is a promising tool to improve the safety related to OR devices.