US 11,875,479 B2
Fusion of deep learning and handcrafted techniques in dermoscopy image analysis
Nabin K Mishra, Overland Park, KS (US); Reda Kasmi, Bejaia (DZ); William V. Stoecker, Rolla, MO (US); Jason R. Hagerty, Waynesville, MO (US); Pavani Jella, Campbell, CA (US); and Norsang Lama, Rolla, MO (US)
Filed by Nabin K Mishra, Overland Park, KS (US); Reda Kasmi, Bejaia (DZ); William V. Stoecker, Rolla, MO (US); Jason R. Hagerty, Waynesville, MO (US); Pavani Jella, Campbell, CA (US); and Norsang Lama, Rolla, MO (US)
Filed on Dec. 30, 2020, as Appl. No. 17/138,450.
Claims priority of provisional application 62/956,510, filed on Jan. 2, 2020.
Prior Publication US 2021/0209754 A1, Jul. 8, 2021
Int. Cl. G06T 7/00 (2017.01); G06T 5/00 (2006.01); G06T 7/90 (2017.01); G06T 7/11 (2017.01); G06V 10/56 (2022.01); G06V 10/46 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC G06T 5/002 (2013.01) [G06T 5/005 (2013.01); G06T 7/0012 (2013.01); G06T 7/11 (2017.01); G06T 7/90 (2017.01); G06V 10/462 (2022.01); G06V 10/56 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06T 2207/30088 (2013.01); G06V 2201/03 (2022.01)] 36 Claims
OG exemplary drawing
 
1. A system for identifying melanoma and other skin cancers in a dermoscopy image or other close-up skin image, the system comprising:
an image analyzer having at least one processor that instantiates components stored in a memory, the components comprising:
a segmenter configured to segment a lesion from the rest of the image,
a handcrafted feature component including:
a median color splitting model for separating the image into a plurality of color regions,
a vessel detection model for detecting elevated vascularity,
an atypical pigment network detection model for identifying a pigment network whose structure varies in size and shape,
a salient point detection model for detecting salient points based on an intensity plane of the image,
a color detection model for detecting at least one of a white area, a pink shade, a pink blush, and a semi-translucency,
an outside-lesion model, obtained by applying at least one of the handcrafted component model features to the area outside the segmented area, and
a manual classifier configured to provide a first analysis result,
a deep learning feature component including: a deep learning feature detector for automatically detecting melanoma and other skin cancer features, and an automatic classifier configured to provide a second analysis result, and
an overall classifier configured to provide an overall analysis result based on the first analysis result and the second analysis result.