Dataset Information

Heidelberg colorectal data set for surgical data science in the sensor operating room.

ABSTRACT: Image-based tracking of medical instruments is an integral part of surgical data science applications. Previous research has addressed the tasks of detecting, segmenting and tracking medical instruments based on laparoscopic video data. However, the proposed methods still tend to fail when applied to challenging images and do not generalize well to data they have not been trained on. This paper introduces the Heidelberg Colorectal (HeiCo) data set - the first publicly available data set enabling comprehensive benchmarking of medical instrument detection and segmentation algorithms with a specific emphasis on method robustness and generalization capabilities. Our data set comprises 30 laparoscopic videos and corresponding sensor data from medical devices in the operating room for three different types of laparoscopic surgery. Annotations include surgical phase labels for all video frames as well as information on instrument presence and corresponding instance-wise segmentation masks for surgical instruments (if any) in more than 10,000 individual frames. The data has successfully been used to organize international competitions within the Endoscopic Vision Challenges 2017 and 2019.

SUBMITTER: Maier-Hein L

PROVIDER: S-EPMC8042116 | biostudies-literature | 2021 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Heidelberg colorectal data set for surgical data science in the sensor operating room.

Maier-Hein Lena L Wagner Martin M Ross Tobias T Reinke Annika A Bodenstedt Sebastian S Full Peter M PM Hempe Hellena H Mindroc-Filimon Diana D Scholz Patrick P Tran Thuy Nuong TN Bruno Pierangela P Kisilenko Anna A Müller Benjamin B Davitashvili Tornike T Capek Manuela M Tizabi Minu D MD Eisenmann Matthias M Adler Tim J TJ Gröhl Janek J Schellenberg Melanie M Seidlitz Silvia S Lai T Y Emmy TYE Pekdemir Bünyamin B Roethlingshoefer Veith V Both Fabian F Bittel Sebastian S Mengler Marc M Mündermann Lars L Apitz Martin M Kopp-Schneider Annette A Speidel Stefanie S Nickel Felix F Probst Pascal P Kenngott Hannes G HG Müller-Stich Beat P BP

Scientific data 20210412 1

Image-based tracking of medical instruments is an integral part of surgical data science applications. Previous research has addressed the tasks of detecting, segmenting and tracking medical instruments based on laparoscopic video data. However, the proposed methods still tend to fail when applied to challenging images and do not generalize well to data they have not been trained on. This paper introduces the Heidelberg Colorectal (HeiCo) data set - the first publicly available data set enabling ...[more]

PMID: 33846356

Similar Datasets

Project description:IntroductionSurgical-related injuries are frequent, in fact the reported percentage of musculoskeletal disorders in surgeons is between 47% and 87%. These conditions are caused by long periods of standing, incorrect postures, repeated movements, little rest between operations, the lack of integrated operator rooms, the correct number and arrangement of monitors and the use of non-ergonomic instruments. This survey aims to assess the Italian overview both highlighting how prevalent surgical-related injury is in our surgeons and whether there is an operating room ergonomics education program in Italian surgical specialty schools.MethodsAn anonymous questionnaire was designed through SurveyMonkey© web application. This survey was composed of 3 different sections concerning the general characteristics of the participants, their surgical background and any training performed, and any injuries or ailments related to the surgical activity. The survey was carried out in the period 1th of December 2022 and the 6th of February 2023.ResultsAt the close of our survey, 300 responses were collected. Among the participants, the two most represented specialties were Gynecology and Obstetrics (42.3%) and General Surgery (39.7%) and surgeons were mainly employed in the Northern regions of Italy (54.8%). Analyzing the participants' background, 61.7% of the respondents had laparoscopic training during their training and only 53.1% had a pelvic trainer during their residency. In accordance with 98.7% of the respondents, during surgery we have the feeling of being in an uncomfortable position that causes discomfort or muscle pain, and regarding the frequency of these discomforts, the majority of our study population experiences these problems monthly (46.2%), while in 29.6% it is experienced weekly, 12.1% annually and finally 12.1% daily. The surgical approach that is most correlated with these disorders is laparoscopy (62.7%) while the one that causes the least discomfort is robotic surgery (1.4%). These discomforts cause 43.9% of our population to take a break or do short exercises to reduce pain during surgery, and the body areas most affected are the back (61.6%), neck (40.6%) and shoulders (37.8%).ConclusionDespite this, our survey allows us to highlight some now-known gaps present in the surgical training program of our schools and the lack of protection toward our surgeons during their long career.

Project description:Purpose Resident involvement would likely lead to prolonged operative time of a surgical case performed at academic medical centers. However, little is known about factors beneath this phenomenon. The purpose of this study was to investigate whether factors from case (procedure type, surgical case complexity, and surgical approach), teacher (attending surgeon experience and gender), and learner (resident postgraduate training year and gender) would influence operative time of surgical cases involved teaching a resident (SCT). Methods A single-institution retrospective analysis of 3 common general surgery procedures, including cholecystectomies, colectomies, and inguinal hernia, with involvement of general surgery residents between 2016 and 2020 was conducted. Surgical operative time was defined as the "cut-to-close” time from incision to completion of wound closure. Analysis of variance for continuous variables and multivariable linear regression were applied. Results A total of 4,417 eligible SCT were included. The average operative time was 114.8 ± 78.7 min. SCT with male resident involvement showed a significantly longer operative time than those with female residents (117 vs. 112, p = 0.01). Comparable operative time was observed between male and female attending surgeon cases (115.5 vs. 110.8, p = 0.15). SCT operating time decreased with increased resident training level, except for SCT with involvement of Year2 residents. SCT with Year5 residents demonstrated the lowest time to case completion (110.5 min); SCT with major complications took least time to complete (105.7 min). Univariate and multivariate analysis revealed resident training year level, resident gender, and case complexity as factors associated with significant differences in operative time. Attending surgeon experience, surgeon gender, surgical approach, and procedure type did not impact SCT operative time. Conclusion Our study findings suggest resident training level, resident gender, and case complexity are factors significantly associated with SCT operative time of cholecystectomies, colectomies, and inguinal hernia. Attending surgeons are recommended to factor them into pre-operative planning.

Dataset Information

Heidelberg colorectal data set for surgical data science in the sensor operating room.

Publications

Heidelberg colorectal data set for surgical data science in the sensor operating room.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets