US 12,133,730 B2
Systems and methods for evaluating mindfulness
David Yang, Doha (QA); Anis Troudi, Doha (QA); Dena Al Thani, Doha (QA); Amine Bermak, Doha (QA); and Mounir Hamdi, Doha (QA)
Assigned to QATAR FOUNDATION FOR EDUCATION, SCIENCE AND COMMUNITY DEVELOPMENT, Doha (QA)
Filed by Qatar Foundation for Education, Science and Community Development, Doha (QA)
Filed on Dec. 22, 2021, as Appl. No. 17/559,015.
Claims priority of provisional application 63/130,001, filed on Dec. 23, 2020.
Prior Publication US 2022/0192560 A1, Jun. 23, 2022
Int. Cl. G06N 3/00 (2023.01); A61B 5/00 (2006.01); A61B 5/16 (2006.01); G06N 3/02 (2006.01); G16H 20/70 (2018.01)
CPC A61B 5/165 (2013.01) [A61B 5/004 (2013.01); A61B 5/163 (2017.08); A61B 5/168 (2013.01); G06N 3/02 (2013.01); G16H 20/70 (2018.01)] 19 Claims
OG exemplary drawing
 
1. A system for detecting whether an individual's attention is maintained comprising:
a camera configured to capture video;
a memory; and
a processor in communication with the memory, the processor configured to:
capture, via the camera, video of the individual,
extract, via at least one first model, a plurality of features of the individual from the captured video, wherein the plurality of features include a plurality of positions of a left eye of the individual, a plurality of positions of a right eye of the individual, a plurality of roll angles of the individual's head, a plurality of yaw angles of the individual's head, a probability that the individual is smiling, a probability that the left eye is closed, and a probability that the right eye is closed,
determine, via at least one second model, whether the individual's attention is maintained based on the plurality of extracted features and a plurality of weighting factors, wherein to determine whether the individual's attention is maintained based on the plurality of extracted features and the plurality of weighting factors the processor is configured to:
calculate a left eye motion, wherein the left eye motion is a mean displacement of the plurality of positions of the left eye of the individual across a first set of frames of the captured video over a first predetermined time period;
calculate a right eye motion, wherein the right eye motion is a mean displacement of the plurality of positions of the right eye of the individual across a second set of frames of the captured video over a second predetermined time period;
calculate a roll angle motion, wherein the roll angle motion is a mean displacement of the plurality of roll angles of the individual's head across a third set of frames of the captured video over a third predetermined time period;
calculate a yaw angle motion, wherein the yaw angle motion is a mean displacement of the plurality of yaw angles of the individual's head across a fourth set of frames of the captured video over a fourth predetermined time period; and
determine whether the individual's attention is maintained based on the calculated left eye motion, right angle motion, roll angle motion, yaw angle motion, the probability that the individual is smiling, the probability that the left eye is closed, the probability that the right eye is closed, and the plurality of weighting factors, and
cause an alert to be generated in response to determining that the individual's attention is not maintained.