Method and device for analysing an image

Patent metadata

Field	Value
Publication number	US-10499845-B2
Application number	US-201515507107-A
Country	US
Kind code	B2
Filing date	Aug 25, 2015
Priority date	Aug 25, 2014
Publication date	Dec 10, 2019
Grant date	Dec 10, 2019

Abstract

A method for analyzing an image of a lesion on the skin of a subject including (a) identifying the lesion in the image by differentiating the lesion from the skin; (b) segmenting the image; and (c) selecting a feature of the image and comparing the selected feature to a library of predetermined parameters of the feature. The feature of the lesion belongs to any one selected from the group: color, border, asymmetry and texture of the image.

First claim

The invention claimed is:

1. A method for analysing analyzing an image of a lesion on the skin of a subject, the method comprising:
(a) identifying the lesion in the image by differentiating the lesion from the skin;
(b) segmenting the image; and
(c) selecting a feature of the image and comparing the selected feature to a library of pre-determined parameters of the feature, wherein the feature of the lesion belongs to any one selected from the group: colour, border, asymmetry and texture of the image,
(d) quantifying the color variation and border irregularity of the image of the lesion, wherein the irregularity of the border is determined by:
(a) providing lines along the border;
(b) determining the angles between two adjacent lines; and
(c) determining the average and variance of the angles,
wherein the number of lines chosen is any number 8, 12, 16, 20, 24 or 28.

2. The method according to claim 1 , wherein the image is processed prior to identifying the lesion in the image.

3. The method according to claim 2 , wherein processing comprises down-sampling the image.

4. The method according to claim 3 , wherein segmenting the image further comprises a first segmenting and a second segmenting, the first segmenting is a coarse segmentation to determine an uncertain region on the downsampled image and the second segmenting refines the uncertain region to obtain segment boundary details.

5. The method according to claim 4 , wherein the uncertain region is +/−2 pixels around the coarse segmentation region boundary.

6. The method according to claim 4 , wherein the second segmenting is carried out using a MST-based algorithm.

7. The method according to claim 1 , wherein the group is further divided into sub-groups and the feature is selected by comparing the feature to other features belonging to other sub-groups and other features within the same sub-group.

8. The method according to claim 1 , wherein the lesion in the image is identified by comparing a colour of the skin to a library of pre-determined colours.

9. The method according to claim 1 , wherein segmenting the lesion further comprising removing segments of the lesion that are connected to a skin boundary.

10. The method according to claim 1 , wherein segmenting the image is a result of two segmentations,
(a) a minimal intra-class-variance thresholding algorithm to locate smoothly-changing borders; and
(b) a minimal-spanning-tree based algorithm to locate abruptly-changing borders.

11. The method according to claim 1 , wherein segmenting is carried out by a region-based method.

12. The method according to claim 1 , wherein the color variation is quantified by:
(a) dividing image into N-partitions, each partitions further divided into M-subparts;
(b) calculating an average pixel value for each subpart and assigning a vector to the subpart; and
(c) determining a maximum distance between the vectors,
wherein a value of N is any value 4, 8, 12 or 16; and a value of M is any value 2, 4 or 8.

13. The method according to claim 1 , wherein the lesion is present in a tissue having a dermal-epidermal junction and an epidermal layer.

14. The method according to claim 1 , wherein the lesion is an acral lentiginous melanoma.

15. The method according to claim 1 , wherein the method further comprising acquiring the image on a computing device and the analysis is carried out on the same computing device.

16. A device for analyzing an image of an object and evaluating the risk or likelihood of a disease or condition, the system comprising:
(a) an image capturing device for capturing the image of an object; and
(b) a processor for executing a set of instructions stored in the device for analyzing the image, the set of instructions includes a library of algorithms stored in the device to carry out a method comprising:
identifying the lesion in the image by differentiating the lesion from the skin;
segmenting the image; and
selecting a feature of the image and comparing the selected feature to a library of pre-determined parameters of the feature, wherein the feature of the lesion belongs to any one selected from the group: color, border, asymmetry and texture of the image,
quantifying the color variation and border irregularity of the image of the lesion, wherein the irregularity of the border is determined by:
(a) providing lines along the border;
(b) determining the angles between two adjacent lines; and
(c) determining the average and variance of the angles,
wherein the number of lines chosen is any number 8, 12, 16, 20, 24 or 28.

17. The device according to claim 16 , wherein the object is a lesion on a patient's body.

18. The device according to claim 16 , wherein the disease is melanoma.

19. The device according to claim 16 , wherein the device further comprising a graphical user interface for indicating to a user the results of the analysis.

20. A non-transitory computer-readable medium including executable instructions to carry out a method for analyzing an image of a lesion on the skin of a subject, the method comprising:
(a) identifying the lesion in the image by differentiating the lesion from the skin;
(b) segmenting the image; and
(c) selecting a feature of the image and comparing the selected feature to a library of pre-determined parameters of the feature, wherein the feature of the lesion belongs to any one selected from the group: colour, border, asymmetry and texture of the image,
(d) quantifying the color variation and border irregularity of the image of the lesion, wherein the irregularity of the border is determined by:
(a) providing lines along the border;
(b) determining the angles between two adjacent lines; and
(c) determining the average and variance of the angles,
wherein the number of lines chosen is any number 8, 12, 16, 20, 24 or 28.

Assignees

• Univ Singapore Technology & Design
• Nat Skin Centre Singapore Pte Ltd

Inventors

• Do Thanh-Toan
• Zhou Yiren
• Pomponiu Victor
• Cheung Ngai-Man
• Koh Dawn Chin Ing
• Hoon Tan Suat

Classifications

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G06T2207/20016

  Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title

• G06T2207/10024

  Color image · CPC title

• A61B5/7264

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title