Method and system for anonymizing raw surgical procedure videos

1. A computer-implemented method for verifying that an anonymized surgical procedure video is free of personally-identifiable information, the method comprising:

receiving a surgical video corresponding to a surgical procedure;

removing personally-identifiable information (PII) from the surgical video to generate an anonymized surgical video;

selecting a set of verification video segments from the anonymized surgical video, which includes:

determining one or more time periods within the anonymized surgical video that are associated with higher probabilities to contain PII; and

randomly selecting the set of verification video segments from the determined one or more time periods within the anonymized surgical video;

determining whether each segment in the set of verification video segments is free of PII; and

if so, replacing the surgical video with the anonymized surgical video for storage,

if not, performing additional PII removal steps on the anonymized surgical video to generate an updated anonymized surgical procedure video.

2. The computer-implemented method of claim 1 , wherein determining the one or more time periods comprises:

segmenting the anonymized surgical video into a set of video portions including timing information based on a set of predefined surgical phases; and

identifying a subset of the set of predefined surgical phases statistically-known to contain PII, wherein

the set of verification video segments are randomly selected from a subset of the set of video portions corresponding to the identified subset of the set of predefined surgical phases.

3. The computer-implemented method of claim 2 , wherein performing additional PII removal steps on the anonymized surgical video includes:

identifying each video portion in the subset of video portions that contains a randomly-selected verification video segment which is again determined to contain PII; and

for each identified video portion determined to contain PII, performing PII removal steps to the identified video portion to de-identify remaining PII within the identified video portion.

4. The computer-implemented method of claim 3 , wherein identifying the subset of the set of predefined surgical phases statistically-known to contain PII comprises identifying one or more surgical phases among the set of predefined surgical phases which have higher probabilities to contain one or more out-of-body (OOB) events, wherein an OOB event corresponds to a time period when an endoscope is outside of a patient's body during the surgical procedure.

5. The computer-implemented method of claim 4 , wherein the time period when the endoscope is outside of a patient's body can include:

a time period when the endoscope is temporarily taken out of the patient's body during the surgical procedure;

a time period before the endoscope being inserted into the patient's body at a beginning of the surgical procedure; and

a time period after the endoscope being taken out of the patient's body at a end of the surgical procedure.

6. The method of claim 1 , wherein selecting the set of verification video segments from the anonymized surgical video comprises performing a fully-random sampling by randomly-selecting the set of verification video segments throughout the anonymized surgical video.

7. The method of claim 1 , wherein performing additional PII removal steps on the anonymized surgical video comprises performing a manual PII removal procedure on each segment within the set of verification video segments determined to still contain PII.

8. The method of claim 1 , wherein after performing additional PII removal steps on the anonymized surgical video, the method further comprises randomly-sampling another set of verification video segments in the updated anonymized surgical video for additional verification.

9. The computer-implemented method of claim 1 , wherein removing the PII from the surgical video comprises:

detecting one or more forms of personally-identifiable texts within each video image in the surgical video; and

when personally-identifiable text is detected within a video image, de-identifying the detected personally-identifiable text by blurring out or otherwise making the detected personally-identifiable text unreadable in the video image.

10. The computer-implemented method of claim 9 , wherein the one or more forms of personally-identifiable texts comprise:

text printed on a surgical tool captured in the one or more video images in the surgical video; and

text displayed inside an operation room (OR) and accidentally captured during the surgical procedure; and

a text box inserted into the surgical video to display surgical procedure-related information.

11. An apparatus for verifying that an anonymized surgical procedure video is free of personally-identifiable information, the apparatus comprising:

one or more processors; and

a memory coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the apparatus to:

receive a surgical video corresponding to a surgery;

remove personally-identifiable information (PII) from the surgical video to generate an anonymized surgical video;

select a set of verification video segments from the anonymized surgical video, which includes:

determining one or more time periods within the anonymized surgical video that are associated with higher probabilities to contain PII; and

randomly selecting the set of verification video segments from the determined one or more time periods within the anonymized surgical video;

determine whether each segment in the set of verification video segments is free of PII; and

if so, replace the surgical video with the anonymized surgical video for storage,

if not, perform additional PII removal steps on the anonymized surgical video to generate an updated anonymized surgical video.

12. The apparatus of claim 11 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to determine the one or more time periods by:

segmenting the anonymized surgical video into a set of video portions including timing information based on a set of predefined surgical phases; and

identifying a subset of the set of predefined surgical phases statistically-known to contain PII, wherein

the set of verification video segments are randomly selected from a subset of the set of video portions corresponding to the identified subset of the set of predefined surgical phases.

13. The apparatus of claim 12 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to perform additional PII removal operations on the anonymized surgical video by:

identifying each video portion in the subset of the set of predefined video portions that contains a randomly-selected verification video segment which is determined to contain PII; and

for each identified video portion determined to contain PII, performing PII removal steps to the identified video portion to de-identify remaining PII within the identified video portion.

14. The apparatus of claim 13 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to identify the subset of the set of predefined surgical phases statistically-known to contain PII by identifying one or more surgical phases among the set of predefined surgical phases which have higher probabilities to contain one or more out-of-body (OOB) events, wherein an OOB event corresponds to a time period when an endoscope is outside of a patient's body during the surgical procedure.

15. The apparatus of claim 11 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to select the set of verification video segments from the anonymized surgical video by randomly-selecting the set of verification video segments throughout the anonymized surgical video.

16. The apparatus of claim 11 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to remove the PII from the surgical video by:

detecting one or more forms of personally-identifiable texts within each video image in the surgical video; and

when personally-identifiable text is detected within a video image, de-identifying the detected personally-identifiable text by blurring out or otherwise making the detected personally-identifiable text unreadable in the video image.

17. A system for verifying that an anonymized surgical procedure video is free of personally-identifiable information, the system comprising:

one or more processors; and

a memory coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to:

receive an anonymized surgical video, wherein the anonymized surgical video was obtained by removing personally-identifiable information (PII) from a raw surgical video of a surgical procedure;

select a set of verification video segments from the anonymized surgical video, which includes:

determining one or more time periods within the anonymized surgical video that are associated with higher probabilities to contain PIT; and

randomly selecting the set of verification video segments from the determined one or more time periods within the anonymized surgical video;

determine whether each segment in the set of verification video segments is free of PII; and

if so, replace the raw surgical video with the anonymized surgical video for storage,

if not, perform additional PII removal steps on the anonymized surgical video to generate an updated anonymized surgical video.

18. The system of claim 17 , wherein the system is configured to determine the one or more time periods by:

segmenting the anonymized surgical video into a set of video portions including timing information based on a set of predefined surgical phases; and

identifying a subset of the set of predefined surgical phases statistically-known to contain PII, wherein

the set of verification video segments are randomly selected from a subset of the set of video portions corresponding to the identified subset of the set of predefined surgical phases.