Systems and methods for providing digital health services
The present disclosure is directed to providing digital health services. In some embodiments, systems and methods for conducting virtual or remote sessions between patients and clinicians are disclosed. During the sessions, media content (e.g., images, video content, audio content, etc.) may be captured as the patient performs one or more tasks. The media content may be presented to the clinician and used to evaluate a condition of the patient or a state of the condition, adjust treatment parameters, provide therapy, or other operations to treat the patient. The analysis of the media content may be aided by one or more machine learning/artificial intelligence models that analyze various aspects of the media content, augment the media content, or other functionality to aid in the treatment of the patient.
1. A method of remotely programming an implantable medical device that provides therapy to a patient, comprising:
establishing a first communication between a patient controller (PC) device and the implantable medical device, wherein the implantable medical device provides therapy to the patient according to one or more programmable parameters, the PC device communicates signals to the implantable medical device to set or modify the one or more programmable parameters, and the PC device comprises a video camera;
establishing a video connection between the PC device and a clinician programmer (CP) device of a clinician for a remote programming session in a second communication that includes an audio/video (A/V) session;
communicating a value for a respective programmable parameter of the medical device from the CP device to the PC device during the remote programming session; and
modifying, by the PC device, the respective programming parameter of the medical device according to the communicated value from the CP device during the remote programming session;
wherein the method further comprises:
automatically analyzing, by one or more processors, video data of the patient from the A/V session to identify a plurality of landmark points on a face of the patient indicative of facial expression of the patient;
processing data related to the landmark points to identify a first metric representing a ratio of a first area on a first side of the facial expression and a second area on a second side of the facial expression;
processing data related to the landmark points to identify a plurality of second metrics with each second metric representing a ratio of distances across the facial expression of the patient, wherein the distances across the facial expression of the patient comprises a distance along an approximate vertical midline of a face of the patient and a horizontal distance along the approximate middle of a face of the patient;
providing the first and second metrics to a trained neural network to generate a patient metric indicative of patient condition; and
displaying the patient metric using the CP device during the remote programming session.
2. The method of claim 1 wherein the patient metric is related to a chronic pain state of the patient.
3. The method of claim 2 further comprising:
detecting asymmetry in the facial expression of the patient.
4. The method of claim 1 wherein the patient metric is related to a motor disorder state of the patient.
5. The method of claim 2 comprising detecting a blank affect of the patient with a motor disorder.
6. The method of claim 1 further comprising: processing a timeline of video data according to a sliding time window processing operation and repeating calculation of the first and second metrics for each time window of the sliding time window processing operation.
7. The method of claim 6 further comprising:
identifying frames of video where a probability of accurate classification exceeds a predetermined level.
8. The method of claim 7 further comprising:
modifying a video stream for frames of video to display the patient metric for the identified frames.
9. The method of claim 6 further comprising:
identifying frames of video where a probability of accurate classification fails to exceed a predetermined level.
10. The method of claim 9 further comprising:
omitting displaying the patient metric for the identified frames.
11. The method of claim 1 wherein the first and second areas are areas identified as disposed under a respective eye, on a respective side of a nose of the patient, and above a mouth of the patient.
12. The method of claim 1 further comprising:
receiving user input from a clinician to toggle between video augmented with one or more graphical user interface (GUI) components modified according to the calculated patient metric and a video stream without display of the one or more GUI components.
13. The method of claim 1 further comprising:
displaying, by the CP device, one or more graphical user interface (GUI) components defined by identified landmarks over a display of the patient in a video stream during the A/V session.