Automatic identification of medically relevant video elements

US11410310B2 · kind B2 · utility

Assignee

• KARL STORZ SE & Co. KG · DE

Inventors

• Christoph Hiltl · Tuttlingen, DE
• Heinz-Werner Stiller · Jestetten, DE
• Nader Hassanzadeh · Tuttlingen, DE
• Mélissa Wiemuth · Tuttlingen, DE
• Frank Stock · Tuttlingen, DE
• Lars Mündermann · Tuttlingen, DE
• Bernd Münzer · Tuttlingen, DE

Key dates

• Filing date: May 28, 2020
• Grant date: Aug 9, 2022
• Expiry date: Nov 24, 2040

Classification

• Technology area (CPC G): Physics
• CPC primary: G06V2201/034
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Apparatus for an automatic identification of medically relevant video elements, the apparatus comprising a data input, configured to receive a data stream of image slices, wherein the data stream of image slices represents a temporal course of a view of image slices defined by a masking strip of video images from a video which has been recorded during a medical surgery on a patient an analysis apparatus configured to analyze the data stream of image slices via an analysis comprising at least one predefined analysis step for the presence of at least one sought-for feature and to output a result of the presence, and a processing device configured to output a start mark which indicates a correspondence between the presence and a position in the data stream of image slices if the result indicates the presence of the sought-for feature. Also, a corresponding method is disclosed.