Movement feedback for orthopedic patient
Systems and methods may be used for presenting motion feedback for an orthopedic patient. In an example, images may be captured of a patient in motion attempting to perform a task, for example after completion of an orthopedic surgery on the patient. The images may be analyzed to generate a movement metric of the patient corresponding to the task. The movement metric may be compared to a baseline metric (e.g., an average metric or a previous patient metric) for the task. An indication of the comparison may be presented, for example including a qualitative result of the comparison.
1. A method comprising:
capturing pre-operative video of a patient;
identifying, using processing circuitry, a pre-operative gait of the patient based on walking movement performed by the patient in the pre-operative video;
determining a gait type using a machine learning model, the machine learning model trained using gait patterns from a plurality of patients having a plurality of co-morbidities and a plurality of gait types;
generating, using the processing circuitry, an orthopedic intervention plan for the patient based on the gait type; and
outputting information indicative of the orthopedic intervention plan for display, wherein outputting the information indicative of the orthopedic intervention plan for display includes outputting a set of instant loading data on a knee of the patient at corresponding stages of the identified pre-operative gait.
2. The method of claim 1 , wherein the gait type includes a walking speed.
3. The method of claim 1 , wherein the gait type includes a walking stiffness.
4. The method of claim 1 , wherein the gait type includes a pain value.
5. The method of claim 1 , wherein the orthopedic intervention plan includes a kinematic alignment of a knee of the patient, the kinematic alignment including between 1-4, inclusive, degrees of varus.
6. The method of claim 1 , wherein the orthopedic intervention plan includes a surgical procedure.
7. The method of claim 1 , wherein the gait type includes a classification generated by a classifier, the classification including a knee loading pattern.
8. The method of claim 1 , wherein the pre-operative gait is based on a healthy knee of the patient and wherein the orthopedic intervention plan includes one or more interventions to return the patient to the pre-operative gait.
9. The method of claim 1 , wherein the gait patterns from the plurality of patients are labeled with surgical outcomes.
10. The method of claim 1 , wherein outputting the set of instant loading data on the knee includes outputting an identification of pre-operative wear to the knee using a kinematic alignment.
11. At least one non-transitory machine-readable medium including instructions, which when executed by processing circuitry, cause the processing circuitry to perform operations to:
capture pre-operative video of a patient;
identify a pre-operative gait of the patient based on walking movement performed by the patient in the pre-operative video;
determine a gait type using a machine learning model, the machine learning model trained using gait patterns from a plurality of patients having a plurality of co-morbidities and a plurality of gait types;
generate an orthopedic intervention plan for the patient based on the gait type; and
output information indicative of the orthopedic intervention plan for display, wherein to output the information indicative of the orthopedic intervention plan for display includes to output a set of instant loading data on a knee of the patient at corresponding stages of the identified pre-operative gait.
12. The at least one non-transitory machine-readable medium of claim 11 , wherein the gait type includes a walking speed.
13. The at least one non-transitory machine-readable medium of claim 11 , wherein the gait type includes a walking stiffness.
14. The at least one non-transitory machine-readable medium of claim 11 , wherein the gait type includes a pain value.
15. The at least one non-transitory machine-readable medium of claim 11 , wherein the orthopedic intervention plan includes a kinematic alignment of a knee of the patient, the kinematic alignment including between 1-4, inclusive, degrees of varus.
16. The at least one non-transitory machine-readable medium of claim 11 , wherein the orthopedic intervention plan includes a surgical procedure.
17. The at least one non-transitory machine-readable medium of claim 11 , wherein the gait type includes a classification generated by a classifier.
18. The at least one non-transitory machine-readable medium of claim 11 , wherein the pre-operative gait is based on a healthy knee of the patient and wherein the orthopedic intervention plan includes one or more interventions to return the patient to the pre-operative gait.