Robotically-assisted surgical procedure feedback techniques

1. A system comprising:

a robotic surgical device to automatically perform a portion of a surgical procedure on a patient and capture robotic surgical operation data associated with the portion of the surgical procedure;

a machine learning engine including a training engine, a prediction engine, and a feature determination engine, the training engine to generate a machine learning model trained based on machine learning training data, the machine learning training data including a set of robotic surgical procedure weights generated based on a subset of the robotic surgical operation data associated with a positive surgical procedure outcome, the set of robotic surgical procedure weights associated with surgical outcome metrics, the feature determination engine to generate the set of robotic surgical procedure weights based on a correlation of each of a plurality of pre-operative deformities with each of a plurality of successful intra-operative interventions taken by the robotic surgical device, each of the plurality of successful intra-operative interventions based on a patient surgical outcome received from the patient; and

a processor and memory including instructions, which when executed by the processor, cause the processor to:

record an action taken by the robotic surgical device during the portion of the surgical procedure;

intra-operatively input the action taken by the robotic surgical device to the prediction engine;

generate, in response to the input of the action taken at the prediction engine, a robotic surgical procedure prediction output at the prediction engine based on an application of the set of robotic surgical procedure weights within the generated machine learning model to the action taken by the robotic surgical device, the robotic surgical procedure prediction output including at least one result or next action taken after the action;

determine a robotic surgical procedure recommendation intra-operatively, based on the robotic surgical procedure prediction output, to the portion of the surgical procedure performed by the robotic surgical device; and

output the robotic surgical procedure recommendation by intra-operatively providing the robotic surgical procedure recommendation to a surgeon operating the robotic surgical device.

2. The system of claim 1 , wherein the instructions further cause the processor to submit a plan to machine learning engine to receive feedback preoperatively.

3. The system of claim 1 , wherein the robotic surgical procedure recommendation is a surgeon specific recommendation.

4. The system of claim 3 , wherein the surgeon specific recommendation is based on surgeon preferences.

5. The system of claim 1 , wherein the instructions further cause the processor to simulate the portion of the surgical procedure to determine a plurality of recommended changes, and select the robotic surgical procedure recommendation from the plurality of recommended changes based on outcome likelihoods of the plurality of recommended changes.

6. The system of claim 1 , wherein the robotic surgical device is further to perform a validation action, and wherein the instructions further cause the processor to update the robotic surgical procedure recommendation based on the validation action.

7. The system of claim 1 , wherein to output the robotic surgical procedure recommendation, the instructions cause the processor to output the robotic surgical procedure recommendation without surgeon input as an alert.

8. The system of claim 1 , wherein to output the robotic surgical procedure recommendation, the instructions cause the processor to output the robotic surgical procedure recommendation in response to receiving a request for the robotic surgical procedure recommendation from the surgeon via a user interface.

9. The system of claim 1 , wherein the machine learning training data includes patient-specific information about a past procedure performed on the patient.

10. The system of claim 1 , wherein the machine learning training data includes demographic-specific information corresponding to the patient, the demographic-specific information including at least one of patient size, surgical procedure type, or patient age.

11. The system of claim 1 , wherein the instructions further cause the processor to output a post-operative recommendation for the patient, including a physical therapy recommendation, based on demographic-specific information.

12. The system of claim 1 , wherein the instructions further cause the processor to:

store anonymized data related to the action taken by the robotic surgical device on a first server;

receive a code entered by the patient;

pull the anonymized data onto a second server; and

tie patient identifying information to the anonymized data on the second server.

13. The system of claim 1 , wherein the instructions further cause the processor to determine a plurality of recommendations, each having a corresponding action and confidence score, and select the robotic surgical procedure recommendation from the plurality of recommendations based on the confidence score of the robotic surgical procedure recommendation.

14. The system of claim 1 , wherein the robotic surgical procedure recommendation includes recommending the next action, and wherein the robotic surgical device is further to perform the next action after receiving confirmation from the surgeon.

15. A method comprising:

generating a set of robotic surgical procedure weights at a feature determination engine based on a correlation of each of a plurality of pre-operative deformities with each of a plurality of successful intra-operative interventions taken by a robotic surgical device, each of the plurality of successful intra-operative interventions based on a patient surgical outcome received from a patient;

generating a machine learning model trained based on machine learning training data, the machine learning training data including the set of robotic surgical procedure weights generated based on a subset of robotic surgical operation data captured by a robotic surgical device, the subset of robotic surgical operation data associated with a positive surgical procedure outcome, the set of robotic surgical procedure weights associated with surgical outcome metrics;

performing, with the robotic surgical device, a portion of a surgical procedure on a patient;

recording, using a processor, an action taken by the robotic surgical device during the portion of the surgical procedure;

intra-operatively input the action taken by the robotic surgical device to a prediction engine;

generate, in response to the input of the action taken at the prediction engine, a robotic surgical procedure prediction output at a prediction engine within a machine learning engine based on an application of the set of robotic surgical procedure weights within the generated machine learning model to the action taken by the robotic surgical device, the robotic surgical procedure prediction output including at least one result or next action taken after the action;

determining a robotic surgical procedure recommendation, based on the robotic surgical procedure prediction output, to the portion of the surgical procedure performed by the robotic surgical device; and

outputting the robotic surgical procedure recommendation by intra-operatively displaying the robotic surgical procedure recommendation to a surgeon operating the robotic surgical device on a display device.

16. The method of claim 15 , further comprising:

storing anonymized data related to the action taken by the robotic surgical device on a first server;

receiving a code entered by the patient;

pulling the anonymized data onto a second server; and

tying patient identifying information to the anonymized data on the second server.