Systems and methods for quantitative diagnosis of anemia

12268,498

17/291215

November 05, 2019

A smartphone-based hemoglobin (Hgb) assessment application quantitatively analyzes pallor in patient-sourced photos using image analysis algorithms to enable a noninvasive, accurate quantitative smartphone app for detecting anemia. A user takes a photo of his/her fingernail beds using the app and receives an accurate displayed Hgb level. Since fingernails do not contain melanocytes, the primary source of color of these anatomical features is blood Hgb. At the same time, quality control software minimizes the impact of common fingernail irregularities (e.g. leukonychia and camera flash reflection) on Hgb level measurement. Metadata recorded upon capturing the image is leveraged for determining a users' Hgb level thereby eliminating the need for external equipment. A personalized calibration of image data with measured Hgb levels improves the accuracy of the application.

EMORY UNIVERSITY (Atlanta, GA, US); CHILDREN'S HEALTHCARE OF ATLANTA, INC. (Atlanta, GA, US); GEORGIA TECH RESEARCH CORPORATION, INC. (Atlanta, GA, US); Sanguina, Inc (Peachtree Corners, GA, US)

EMORY UNIVERSITY (Atlanta, GA, US), CHILDREN'S HEALTHCARE OF ATLANTA, INC. (Atlanta, GA, US), GEORGIA TECH RESEARCH CORPORATION, INC. (Peachtree Corners, GA, US), Sanguina, Inc. (Peachtree Corners, GA, US)

1. A system for analyzing an image for estimating hemoglobin levels, the system comprising at least one processor configured to: access a camera of a mobile device; force on a flash functionality of the camera to normalize variable background lighting conditions; capture, via the camera, an image of one or more fingernail beds of a user with the flash on, wherein: the camera automatically uses at least one lighting condition setting; and the mobile device automatically associates metadata with the image, the metadata comprising information about the flash and the at least one lighting condition setting; receive the image from the camera; receive an indication of one or more regions of interest on the image based, at least in part, on a user input to the mobile device, the one or more regions of interest at least partially including the one or more fingernail beds; determine pixel intensity for each of the one or more regions of interest; average pixel intensity from color channels across each of the one or more regions of interest; transform the average pixel intensity from the color channels into a value that correlates with the user's approximate hemoglobin (Hgb) level using machine learning to correct for variations in the average pixel intensity; adjust the value based on the metadata associated with the image to compensate for the at least one lighting condition setting and the flash; determine the user's approximate Hbg level from the image based on the adjusted value; and output the user's approximate Hgb level to a display of the mobile device, wherein the Hgb level is used to monitor changes to the user's Hgb levels over time.

2. The system of claim 1 , wherein the at least one processor is further configured to: receive the indication of the one or more regions of interest; and automatically display a visual indication of the one or more regions of interest.

3. The system of claim 2 , wherein the user input comprises the user tapping the image to indicate the one or more regions of interest.

4. The system of claim 2 , wherein the visual indication of the one or more regions of interest comprise one or more boxes encompassing each of the one or more regions of interest.

5. The system of claim 1 , wherein the one or more regions of interest include an area of approximately 10 mm 2 .

6. The system of claim 1 , wherein the one or more regions of interest consist of one or more portions of the one or more fingernail beds.

7. The system of claim 1 , wherein the at least one processor is further configured to receive the indication of the one or more regions of interest automatically from a third-party computing system.

8. The system of claim 1 , wherein the at least one processor is further configured to receive the indication of the one or more regions of interest automatically from a remote server.

9. The system of claim 8 , wherein the remote server determines the one or more regions of interest via a machine learning algorithm.

10. The system of claim 1 , wherein the at least one processor determines the one or more regions of interest via a machine learning algorithm.

11. The system of claim 1 , wherein the system receives the image of the one or more fingernail beds of the user and quantitatively analyzes pallor of the one or more fingernail beds in the one or more regions of interest without the use of external hardware physically coupled to the system.

12. The system of claim 1 , wherein the determining pixel intensity for each of the one or more regions of interest comprises extracting color data from the image.

13. The system of claim 1 , wherein the at least one processor is further configured to exclude areas including leukonychia and/or camera flash reflection from the one or more regions of interest.

14. The system of claim 1 , wherein the user's approximate Hgb level is an approximate complete blood count (CBC) Hgb level.

15. The system of claim 1 , wherein the at least one processor is configured to save the user's approximate Hgb level in a text file.

16. The system of claim 1 , wherein the at least one processor is configured to exclude pixel intensity values outside of a particular range when averaging pixel intensity for each of the one or more regions of interest.

17. The system of claim 1 , wherein the at least one processor is further configured to confirm the image is acceptable by receiving a validation from the user.

18. The system of claim 1 , wherein the metadata includes color management metadata including a white balance of the image.

19. The system of claim 1 , wherein the approximate Hgb level is personally calibrated for the user based on at least one of: one or more previously determined Hgb levels or the user's medical history.

6104939 August 2000 Groner et al.
2003/0002722 January 2003 Jay et al.
2007/0255115 November 2007 Anglin
2009/0267893 October 2009 Kato
2011/0082711 April 2011 Poeze
2012/0265041 October 2012 Yamaguchi
2015/0044098 February 2015 Smart et al.
2016/0089062 March 2016 Sivathanu
2017/0311871 November 2017 Kikuchi
2018/0012365 January 2018 Chefd'hotel et al.
2018/0125610 May 2018 Carrier, Jr.

Indi, T. and Gunge, Y. “Early Stage Disease Diagnosis System Using Human Nail Image Processing.” International Journal of Information Technology and Computer Science. 8. 30-35. (Year: 2016). cited by examiner
Wang, E.J., Li, W., Hawkins, D., Gernsheimer, T., et al. HemaApp: Noninvasive Blood Screening of Hemoglobin Using Smartphone Cameras. Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp '16) (Year: 2016). cited by examiner
International Search Report and Written Opinion for International Application No. PCT/US2019/059742 dated Jan. 21, 2020. cited by applicant
Sanguina, “Over the counter Development” Oct. 10, 2019, 1 page. cited by applicant
International Preliminary Report on Patentability issued for Application No. PCT/US2019/059742, dated May 20, 2021. cited by applicant

edspgr.12268498