US 11,915,428 B2
Method for determining severity of skin disease based on percentage of body surface area covered by lesions
Yanqing Chen, San Diego, CA (US); Charles Tang, Berkeley, CA (US); and Ernesto J. Munoz-Elias, Rancho Santa Fe, CA (US)
Assigned to Janssen Biotech, Inc., Horsham, PA (US)
Filed by Janssen Biotech, Inc., Horsham, PA (US)
Filed on Nov. 10, 2022, as Appl. No. 18/054,372.
Application 18/054,372 is a continuation of application No. 17/115,036, filed on Dec. 8, 2020, granted, now 11,538,167.
Claims priority of provisional application 62/945,642, filed on Dec. 9, 2019.
Prior Publication US 2023/0060162 A1, Mar. 2, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/11 (2017.01)
CPC G06T 7/11 (2017.01) [G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30088 (2013.01); G06T 2207/30096 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A system for determining severity of skin disease, comprising:
a non-transitory computer-readable medium with instructions stored thereon, which when executed by a processor perform steps comprising:
segmenting at least one test image into a set of image regions with a Felzenszwalb segmentation algorithm;
inputting the set of image regions to a trained convolutional neural network;
with the trained neural network, classifying each region of the set of image regions as (a) lesion regions, (b) non-lesion body regions, or (c) non-body regions;
calculating a percentage of body surface area (BSA) covered by lesions by using the combined areas of the lesion regions and the combined areas of the lesion regions and non-lesion body regions.