Detection, identification, and verification of patients involved in diagnostic sessions
Inventors
Atkinson, Ian • Kramer, Ryan • Glik, Eliezer • STOERTZ, Aaron Haywood
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12437860-B2
Publication Date
2025-10-07
Expiration Date
Abstract
Introduced here are approaches to ensuring that digital images generated during diagnostic sessions are properly associated with the appropriate patients. By implementing these approaches, a diagnostic platform can minimize the time needed to manually input information prior to a diagnostic session and improve the accuracy of this information.
Core Innovation
The disclosure describes computer-implemented approaches for detecting, identifying, and verifying patients in diagnostic sessions by automatically associating digital images with the correct patient. A retinal camera generates digital images of an eye, and a notification instructs a patient to place an eye near the retinal camera to enable image acquisition.
A first digital image captured during a diagnostic session is compared to a database of digital images associated with past diagnostic sessions. The approach determines that the first digital image matches an entry in the database, identifies a profile associated with the matching entry, and populates a data structure with information obtained from the profile prior to encoding a second digital image.
The retinal camera’s second digital image is generated based on light reflected by a retina of the eye and through a lens of the retinal camera. The disclosure also includes additional mechanisms such as using a secondary/pupil camera to determine left/right eye and label image data structures, alternative identification using identity documents via OCR and verification services, and identification using machine-readable identifiers such as QR code, Bluetooth Low Energy, NFC, RFID, and audible codes.
To improve reliability, the disclosure includes closed-loop analysis in which additional retinal images are captured under different conditions when diagnostic model confidence is below a threshold, and it provides supporting system/network and a processing system overview.
Claims Coverage
The document includes one independent claim (clm-00001) directed to a non-transitory medium for retinal-camera-based patient identification/verification in diagnostic sessions. The claim is then refined by dependent claims that add image alignment using a secondary camera, constrain where the processor is located, and narrow the comparison to a biometric algorithm based on iris color and/or retinal blood-vessel pattern.
Retinal-camera patient identification via database matching and profile-driven data structure
A non-transitory medium with instructions for generating a notification instructing a patient to place an eye near a retinal camera, acquiring a first digital image from the retinal camera, comparing the first digital image to a database of digital images associated with past diagnostic sessions, determining that the first digital image matches an entry in the database, identifying a profile associated with the matching entry, and populating a data structure with information obtained from the profile before encoding a second digital image generated by the retinal camera based on light reflected by a retina of the eye and through a lens of the retinal camera.
Secondary-camera-assisted eye alignment for first digital image
The medium further instructs that a secondary camera near the retinal camera lens is used to generate the first digital image.
Processor housed in the retinal camera
The medium specifies that a processor is housed in the retinal camera.
Processor housed in a computing device connected to the retinal camera
The medium specifies that the processor is housed in a computing device communicatively connected to the retinal camera.
Biometric comparing using iris color and/or retinal blood-vessel pattern
The medium narrows the comparing operation by executing an algorithm that determines a matching entry based on the color of the iris and/or the pattern of blood vessels in the retina.
Overall, the claim set covers retinal-camera image acquisition in diagnostic sessions, database comparison to find a matching entry, identification of an associated patient profile, and population of a data structure prior to encoding an additional retinal-camera image. Dependent claims further refine patient-image capture using a secondary camera for alignment, constrain processor placement, and narrow matching to biometric comparison based on iris color and/or retinal blood-vessel pattern.
Stated Advantages
Reducing manual data entry time and errors.
Improving security by limiting sensitive data stored on the imaging device.
Documented Applications
Patient detection, identification, and verification in diagnostic sessions by associating digital retinal images with the correct patient.
Closed-loop diagnostic model confidence handling by capturing additional retinal images under different conditions when confidence is below a threshold.
Patient identification using identity documents via OCR and verification services.
Patient identification using machine-readable identifiers such as QR code, Bluetooth Low Energy, NFC, RFID, and audible codes.
Interested in licensing this patent?