US 11,854,200 B2
Skin abnormality monitoring systems and methods
Kyoko Crawford, Chicago, IL (US); and Arthur Morrissette, Chicago, IL (US)
Assigned to SkinIO, Inc., Chicago, IL (US)
Filed by SkinIO, LLC, Chicago, IL (US)
Filed on Oct. 20, 2020, as Appl. No. 17/075,336.
Application 17/075,336 is a continuation in part of application No. 16/513,681, filed on Jul. 16, 2019, abandoned.
Application 16/513,681 is a continuation of application No. 15/726,367, filed on Oct. 5, 2017, granted, now 10,354,383, issued on Jul. 16, 2019.
Claims priority of provisional application 62/441,101, filed on Dec. 30, 2016.
Prior Publication US 2021/0104043 A1, Apr. 8, 2021
Int. Cl. G06K 9/00 (2022.01); G06T 7/00 (2017.01); G06T 7/60 (2017.01); G06T 11/60 (2006.01); G06T 5/20 (2006.01); G06T 7/90 (2017.01)
CPC G06T 7/0014 (2013.01) [G06T 7/0012 (2013.01); G06T 7/60 (2013.01); G06T 11/60 (2013.01); G06T 5/20 (2013.01); G06T 7/90 (2017.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30088 (2013.01); G06T 2207/30096 (2013.01); G06T 2210/22 (2013.01)] 4 Claims
OG exemplary drawing
 
1. A process for determining one or more outlier lesions, the process comprising the steps of:
receiving a plurality of image data from an image capture device, wherein each image data includes a skin portion, wherein each image data of the plurality of image data comprises a different skin image of the skin portion, the skin portion including a plurality of lesions;
extracting, via a machine learning system, a first set of information associated with each image data in the plurality of image data;
extracting, via the machine learning system, a second set of information associated with each image data in the plurality of image data;
determining a plurality of vector representations for each set of information associated with each image data in the plurality of image data;
determining an average vector based on the plurality of vector representations across the plurality of image data of the skin portion;
calculating a normalized distance metric based on the plurality of vector representations and the average vector; and
analyzing the plurality of lesions to identify one or more outlier lesions within the plurality of lesions based on the normalized distance metric.