Assessing developmental disorders via eye tracking

1 . A computer-implemented method for using timestamped eye-tracking gaze coordinate data collected by tablet computing devices, the computer-implemented method comprising:

receiving, at a network-connected server in communication with multiple patient-side portable tablet computing devices, session data of multiple eye-tracking sessions, the session data of each session comprising timestamped eye-tracking gaze coordinate data of a corresponding patient collected by a corresponding one of the multiple patient-side portable tablet computing devices during presentation of a set of videos in the session;

simultaneously recalibrating, at the network-connected server, the timestamped eye-tracking gaze coordinate data of each session of the multiple eye-tracking sessions to a predefined resolution or coordinate system for providing processed session data for each session of the multiple eye-tracking sessions; and

simultaneously comparing, at the network-connected server, the processed session data of each session of the multiple eye-tracking sessions to pre-loaded model data such that, for each session of the multiple eye-tracking sessions, the network-connected server:

obtains a comparison result from a trained artificial intelligence model at the network-connected server that uses both the processed session data of the session and corresponding reference eye-tracking data of patients from a same age range as the corresponding patient and that outputs multiple numeric index scores indicative of a severity of a developmental condition, and

stores at a web portal of the network-connected server an assessment report for the corresponding patient that displays said multiple numeric index scores indicative of the severity of the developmental condition.

2 . The computer-implemented method of claim 1 , wherein the assessment report stored at the web portal of the network-connected server displays said multiple numeric index scores indicative of a social disability, a verbal ability, and a nonverbal ability contemporaneously with a graphic indicative of oculomotor function for the corresponding patient.

3 . The computer-implemented method of claim 1 , wherein the assessment report stored at the web portal of the network-connected server is accessible by an operator-side tablet computing device that is configured to wirelessly communicate with at least a plurality of said multiple patient-side portable tablet computing devices.

4 . The computer-implemented method of claim 3 , wherein the web portal that stores the assessment report for the corresponding patient includes a dashboard to selectively output a graphic visualization of the corresponding patient's point-of-gaze overlaid on corresponding visual stimulus stills from selected moments of the set of videos.

5 . The computer-implemented method of claim 1 , further comprising:

loading, at the network-connected server, the corresponding reference eye-tracking data for the multiple eye-tracking sessions in parallel with said simultaneously recalibrating the timestamped eye-tracking gaze coordinate data of each session of the multiple eye-tracking sessions.

6 . The computer-implemented method of claim 5 , wherein the network-connected server comprises a plurality of processing cores, and

wherein said simultaneously recalibrating the timestamped eye-tracking gaze coordinate data of each session of the multiple eye-tracking sessions comprises using a first plurality of processing cores to process the session data of the multiple eye-tracking sessions in parallel while also using a second, different plurality of processing cores to load the corresponding reference eye-tracking data for the multiple eye-tracking sessions, a number of the first plurality of processing cores being larger than a number of the second plurality of processing cores.

7 . The computer-implemented method of claim 6 , wherein said obtaining the comparison result from the trained artificial intelligence model for each session of the multiple sessions comprises:

using the plurality of processing cores including the first plurality of processing cores and the second plurality of processing cores.

8 . The computer-implemented method of claim 1 , wherein the timestamped eye-tracking gaze coordinate data is associated with a list of predetermined set of videos presented to the patient in the session, and wherein the session data comprises timestamp information indicative of when each video in the list of predetermined set of videos was presented in the session so that a subset of the timestamped eye-tracking gaze coordinate data is associated with each video in the list of predetermined videos is individually identified based on the timestamp information.

9 . The computer-implemented method of claim 1 , comprising:

for each session of the multiple eye-tracking sessions, assigning a respective container for the session, and, in the respective container at the network-connected server, using the session data of the session and the pre-loaded model data to obtain the comparison result from the trained artificial intelligence model and to generate the assessment report for the corresponding patient in the session.

10 . The computer-implemented method of claim 9 , further comprising linking the timestamped eye-tracking gaze coordinate data of each session with the list of predetermined set of videos in the session by, in the respective container at the network-connected server, breaking up the eye-tracking data into multiple portions based on corresponding timestamp information associated with each video in the list of predetermined set of videos, each portion of the timestamped eye-tracking gaze coordinate data being associated with one of a respective predetermined video or a corresponding calibration.

11 . The computer-implemented method of claim 10 , wherein said simultaneously recalibrating includes, in the respective container at the network-connected server, recalibrating said portions of the timestamped eye-tracking gaze coordinate data associated with respective predetermined visual stimulus based on at least one portion of the timestamped eye-tracking gaze coordinate data associated with the corresponding calibration.

12 . The computer-implemented method of 10 , further comprising:

in the respective container at the network-connected server, determining a calibration accuracy using at least one portion of the timestamped eye-tracking gaze coordinate data associated with the corresponding calibration and a plurality of predetermined locations where a plurality of calibration targets are presented in the corresponding calibration.

13 . The computer-implemented method of claim 9 , further comprising:

transmitting the assessment report to an operator-side computing device to display the assessment report at a user interface of the operator-side computing device.

14 . The computer-implemented method of claim 13 , wherein the assessment report output from the network-connected server contemporaneously displays:

information of the corresponding patient,

information of an entity performing the session for the corresponding patient,

information of a calibration accuracy in the session,

information of session data collection, and

said multiple numeric index scores for the corresponding patient.

15 . The computer-implemented method of claim 14 , wherein the assessment report indicates a likelihood that the corresponding patient has a developmental, cognitive, social, or mental disability or ability.

16 . The computer-implemented method of claim 15 , wherein the assessment report output from the network-connected server further displays: a graphic visualization of the corresponding patient's point-of-gaze overlaid on corresponding visual stimulus stills from selected moments of the set of videos.

17 . The computer-implemented method of claim 14 , wherein the assessment report indicates a likelihood that the corresponding patient has an Austin Spectrum Disorder (ASD) or a non-ASD.

18 . The computer-implemented method of claim 1 , wherein said receiving the session data of the multiple sessions comprises:

receiving, through the web portal, the session data of the multiple sessions from the multiple patient-side portable tablet computing devices at multiple remote sites.

19 . The computer-implemented method of claim 18 , further comprising:

in response to receiving session data of each session, adding a file pointer for the session data of the session in a processing queue to be processed.

20 . The computer-implemented method of claim 19 , further comprising:

storing the session data of the session using the file pointer for the session in a database; and

retrieving the session data of the session from the database using the file pointer for the session.

21 . The computer-implemented method of claim 1 , wherein the corresponding patient has an age in a range from 5 months to 7 years.

22 . A computer-implemented method for using timestamped eye-tracking gaze coordinate data collected by tablet computing devices, the computer-implemented method comprising:

receiving, at a network-connected server in communication with multiple patient-side portable tablet computing devices, session data of multiple eye-tracking sessions, the session data of each session comprising timestamped eye-tracking gaze coordinate data of a corresponding patient collected by a corresponding one of the multiple patient-side portable tablet computing devices during presentation of a set of videos in the session;

simultaneously recalibrating, at the network-connected server, the timestamped eye-tracking gaze coordinate data of each session of the multiple eye-tracking sessions to a predefined resolution or coordinate system for providing processed session data for each session of the multiple eye-tracking sessions; and

simultaneously comparing, at the network-connected server, the processed session data of each session of the multiple eye-tracking sessions to pre-loaded model data such that, for each session of the multiple eye-tracking sessions, the network-connected server:

obtains a comparison result from a trained artificial intelligence model at the network-connected server that uses both the processed session data of the session and corresponding reference eye-tracking data of patients from a same age range as the corresponding patient and that outputs multiple numeric index scores indicative of a severity of a developmental condition, wherein said simultaneously comparing the processed session data of each session of the multiple eye-tracking sessions to pre-loaded model data includes the network-connected server loading multiple session containers for parallel analysis so that one comparison result is obtained from the trained artificial intelligence model for each one of the multiple session containers, wherein each one of the multiple session containers stores the processed session data of one patient and the pre-loaded model data corresponding to said one patient, and

stores at a web portal of the network-connected server an assessment report for the corresponding patient that displays said multiple numeric index scores indicative of the severity of the developmental condition.

23 . The computer-implemented method of claim 22 , wherein the pre-loaded model data comprises historical eye-tracking data or results for patients having both the same age range and a same condition as said one patient.

24 . A computer-implemented method for using timestamped eye-tracking gaze coordinate data collected by tablet computing devices, the computer-implemented method comprising:

receiving, at a network-connected server in communication with multiple patient-side portable tablet computing devices, session data of multiple eye-tracking sessions, the session data of each session comprising timestamped eye-tracking gaze coordinate data of a corresponding patient collected by a corresponding one of the multiple patient-side portable tablet computing devices during presentation of a set of videos in the session, wherein said receiving the session data of the multiple sessions comprises: receiving, through a web portal of the network-connected server, the session data of the multiple sessions from the multiple patient-side portable tablet computing devices at multiple remote sites, and storing session data from one or more patient-side portable tablet computing devices associated with each remote site in a respective isolated repository that includes information of one or more operators or operator-side computing devices associated with the remote site, wherein the respective isolated repository is isolated from one or more other isolated repositories at the network-connected server such that the respective isolated repository is inaccessible by one or more other remote sites having access to the web portal;

simultaneously recalibrating, at the network-connected server, the timestamped eye-tracking gaze coordinate data of each session of the multiple eye-tracking sessions to a predefined resolution or coordinate system for providing processed session data for each session of the multiple eye-tracking sessions; and

simultaneously comparing, at the network-connected server, the processed session data of each session of the multiple eye-tracking sessions to pre-loaded model data such that, for each session of the multiple eye-tracking sessions, the network-connected server:

obtains a comparison result from a trained artificial intelligence model at the network-connected server that uses both the processed session data of the session and corresponding reference eye-tracking data of patients from a same age range as the corresponding patient and that outputs multiple numeric index scores indicative of a severity of a developmental condition, and

stores at the web portal of the network-connected server an assessment report for the corresponding patient that displays said multiple numeric index scores indicative of the severity of the developmental condition.

25 . The computer-implemented method of claim 24 , further comprising:

dynamically adjusting resources of the network-connected server based at least in part on a number of patient-side portable tablet computing devices that access the network-connected server.