User: Guest  Login
Title:

MITI: SLAM Benchmark for Laparoscopic Surgery

Document type:
Forschungsdaten
Publication date:
20.05.2022
Responsible:
Hartwig, Regine
Authors:
Hartwig, Regine ; Ostler, Daniel ; Rosenthal, Jean-Claude ; Feussner, Hubertus ; Wilhelm, Dirk ; Wollherr, Dirk
Author affiliation:
TUM
Publisher:
TUM
Identifier:
doi:10.14459/2021mp1621941
End date of data production:
26.08.2021
Subject area:
DAT Datenverarbeitung, Informatik; MED Medizin
Resource type:
Experimente und Beobachtungen / experiments and observations; Abbildungen von Objekten / image of objects; Logfiles und Nutzungsdaten / log files and usage data
Data type:
Bilder / images; Texte / texts; Tabellen / tables
Description:
Our MITI Dataset provides all the necessary data by a complete recording of a handheld surgical intervention at Research Hospital Rechts der Isar of TUM. It contains multimodal sensor information from IMU, stereoscopic video, and infrared (IR) tracking as ground truth for evaluation. Furthermore, calibration for the stereoscope, accelerometer, magnetometer, the rigid transformations in the sensor setup, and time-offsets are available. We wisely chose a suitable intervention that contains very fe...     »
Method of data assessment:
During the intervention the surgical scenarios are diverse, containing movements of tissue and movement of the camera behind the tissue and also water reflections and movements. The intervention is called diagnostic laparoscopy and therefore almost no cutting and changing of tissue is conducted, which makes this intervention suitable testing SLAM applications for visceral surgery.
Links:
This dataset relates to the publication: https://ieeexplore.ieee.org/document/9812105
Key words:
Visual-Inertial; Localization; SLAM; SfM; Benchmark; Laparoscopy Surgery; Minimally Invasive; Visceral; Stereoscopic; Inertial Measurement Unit; Visual; Infrared Tracking
Technical remarks:
View and download (208 GB total, 89 files)
The data server also offers downloads with FTP
The data server also offers downloads with rsync (password m1621941):
rsync rsync://m1621941@dataserv.ub.tum.de/m1621941/
Language:
en
Rights:
by, http://creativecommons.org/licenses/by/4.0
 BibTeX