Systems and methods for surgical data censorship

A non-transitory computer-readable medium comprising instructions configured to cause a computer system to perform a method, the method comprising:

• acquiring a plurality of features associated with a plurality of video frames;
• identifying a feature from the plurality of features for each of a classifier of a plurality of different classifiers, each classifier configured to classify a different feature than other classifiers of the plurality of different classifier;
• generating a classification of each feature of the plurality of features using a respective classifier of the plurality of different classifiers;
• providing the classification of each feature to a fusion model configured to output at least one classification prediction;
• classifying, by the fusion model, at least one video frame of the plurality of video frames as depicting a region inside or outside a patient's body based upon the at least one classification prediction;
• determining one or more intervals of consecutive classifications of the plurality of video frames depicting the region outside the patient's body;
• and
• in response to determining that a duration of the one or more intervals is greater than a threshold indicative of an inter-surgery period, excising the one or more intervals corresponding to the region outside the patient's body.

A computer system comprising:

• at least on processor;
• and
• at least one memory comprising instructions configured to cause the computer system to perform a method, the method comprising:
• acquiring a plurality of features associated with a plurality of video frames;
• identifying a feature from the plurality of features for each of a classifier of a plurality of different classifiers, each classifier configured to classify a different feature than other classifiers of the plurality of different classifier;
• generating a classification of each feature of the plurality of features using a respective classifier of the plurality of different classifiers;
• providing the classification of each feature to a fusion model configured to output at least one classification prediction;
• classifying, by the fusion model, at least one video frame of the plurality of video frames as depicting a region inside or outside a patient's body based upon the at least one classification prediction;
• determining one or more intervals of consecutive classifications of the plurality of video frames depicting the region outside the patient's body;
• and
• in response to determining that a duration of the one or more intervals is greater than a threshold indicative of an inter-surgery period, excising the one or more intervals corresponding to the region outside the patient's body.