# US 2021/0330183 A1 · Method for Acquiring Data with the Aid of Surgical Microscopy Systems > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0330183 A1 - Document ID: US20210330183A1 - Status: Published application - Kind: A1 - Application number: 17/241,844 - Filed: 2021-04-27 - Published: 2021-10-28 - Earliest claimed: 2020-04-28 - Assignees: Carl Zeiss Meditec AG - Applicants: Carl Zeiss Meditec AG - Inventors: Stefan Saur; Alexander Urich - Web page: https://patentgrep.com/patent/US20210330183A1 ## CPC - A61B 3/102: {for optical coherence tomography [OCT]} - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - A61B 3/13: Ophthalmic microscopes - G02B 21/0012: {Surgical microscopes (counterbalanced structures for surgical microscopes G02B7/001)} - G02B 21/0056: {based on optical coherence, e.g. phase-contrast arrangements, interference arrangements} - G02B 21/365: {Control or image processing arrangements for digital or video microscopes (G02B21/361, G02B21/362 take precedence)} - G06N 20/00: Machine learning - G16H 30/20: for handling medical images, e.g. DICOM, HL7 or PACS - G16H 30/40: for processing medical images, e.g. editing - G06N 3/02: Neural networks - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA - G02B: OPTICAL ELEMENTS, SYSTEMS OR APPARATUS - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A method for acquiring data with the aid of a surgical microscopy system comprises recording a data set having a multiplicity of images and deciding whether or not the new data set ought to be stored in a data memory of a database. The decision is taken by an existing classifier. A new classifier is determined on the basis of training data which comprise the data sets stored in the data memory of the database and/or comprise data obtained from the data sets stored in the data memory of the database. The new classifier is then used instead of the existing classifier when deciding whether a subsequently recorded new data set ought to be stored. ## Claims ### Claim 1 Method for acquiring data with the aid of a surgical microscopy system, comprising: recording a new data set comprising a multiplicity of images by means of the surgical microscopy system; analysing the new data set and deciding whether or not the new data set ought to be stored in a data memory of a database that is arranged remotely from a surgical microscope of the surgical microscopy system, specifically using an existing classifier on the basis of the analysis of the new data set; storing the new data set in the data memory of the database if it was decided that the at least one data set ought to be stored, and not storing the new data set in the data memory if it was decided that the new data set ought not to be stored; determining a new classifier on the basis of training data which comprise the data sets stored in the data memory of the database and/or comprise data obtained from the data sets stored in the data memory of the database; and using the new classifier instead of the existing classifier when deciding whether or not a subsequently recorded new data set ought to be stored in the data memory. ### Claim 2 (depends on 1) Method according to claim 1, furthermore comprising assigning at least one annotation to each of the data sets stored in the data memory. ### Claim 3 (depends on 2) Method according to claim 2, wherein the training data comprise the annotations. ### Claim 4 (depends on 2) Method according to claim 2, furthermore comprising assigning a significance to each of the data sets stored in the data memory, and wherein the training data comprise the significances. ### Claim 5 (depends on 4) Method according to claim 4, wherein determining the new classifier is effected after a plurality of new data sets have been stored in the data memory. ### Claim 6 (depends on 4) Method according to claim 4, wherein the new classifier is determined in such a way that the new data sets stored in the data memory have the effect that when significances are subsequently assigned to the data sets, the significance newly assigned to a respective data set is lower than the significance previously assigned to this data set. ### Claim 7 (depends on 1) Method according to claim 1, wherein deciding whether or not the new data set ought to be stored in the data memory using the existing classifier is furthermore based on further information specifying a type of a medical intervention, an age, a sex and/or a state of health of a patient on whom the medical intervention is performed, a quality of tissue imaged in the images, a characteristic of the user of the surgical microscopy system, a location at which the surgical microscopy system is situated, and/or a use history of the surgical microscopy system. ### Claim 8 (depends on 1) Method according to claim 1, wherein the multiplicity of images comprises a video having a plurality of frames, a volume image having a plurality of slice images, and a multiplicity of OCT scans. ### Claim 9 (depends on 1) Method according to claim 1, furthermore comprising training a machine learning instrument with the data stored in the database. ### Claim 10 System for acquiring data for machine learning, wherein the system comprises: at least one surgical microscopy system; a database having a data memory; and a module; wherein the database and the module are arranged remotely from the at least one surgical microscopy system; wherein the at least one surgical microscopy system is configured for: recording a new data set comprising a multiplicity of images; analysing the new data set; and deciding whether or not the new data set ought to be stored in the data memory, specifically using an existing classifier on the basis of the analysis of the new data set; transmitting the new data set to the database if it was decided that the new data set ought to be stored; not transferring the new data set to the database if it was decided that the new data set ought not to be stored; receiving a new classifier and replacing the existing classifier with the new classifier received; and using the new classifier when it is decided whether or not a subsequently recorded new data set ought to be stored in the data memory; wherein the database is configured for: receiving the new data set transmitted by the at least one surgical microscopy system; and storing the new data set received in the data memory, wherein the module is configured for: determining the new classifier on the basis of training data which comprise the data sets stored in the data memory of the database and/or comprise data obtained from the data sets stored in the data memory of the database; and transmitting the new classifier to the at least one surgical microscopy system. ### Claim 11 (depends on 10) System according to claim 10, wherein the module is furthermore configured for: assigning at least one annotation to each of the data sets stored in the data memory. ### Claim 12 (depends on 11) System according to claim 11, wherein the training data comprise the annotations. ### Claim 13 (depends on 11) System according to claim 11, wherein the module is furthermore configured for: assigning a significance to each of the data sets stored in the data memory, and wherein the training data comprise the significances. ### Claim 14 (depends on 13) System according to claim 13, wherein the module is configured to carry out the process of determining the new classifier after a plurality of new data sets have been stored in the data memory. ### Claim 15 (depends on 13) System according to claim 13, wherein the module is configured to determine the new classifier in such a way that the new data sets stored in the data memory have the effect that when significances are subsequently assigned to the data sets by means of the module, the significance newly assigned to a respective data set is lower than the significance previously assigned to this data set. ### Claim 16 (depends on 10) System according to claim 10, wherein the at least one surgical microscopy system is configured to carry out the process of deciding whether or not the new data set ought to be stored in the data memory using the existing classifier furthermore on the basis of further information specifying a type of a medical intervention, an age, a sex and/or a state of health of a patient on whom the medical intervention is performed, a quality of tissue imaged in the images, a characteristic of the user of the surgical microscopy system, a location at which the surgical microscopy system is situated, and/or a use history of the surgical microscopy system. ### Claim 17 (depends on 10) System according to claim 10, wherein the multiplicity of images comprises a video having a plurality of frames, a volume image having a plurality of slice images, and a multiplicity of OCT scans. ### Claim 18 (depends on 10) System according to claim 10, furthermore comprising an analysis module configured for training a machine learning instrument with the data stored in the database.