Skin Cancer Surface Shape Based Classification Steven McDonagh - - PowerPoint PPT Presentation

▶

Oct 11, 2022 563 likes •825 views

Skin Cancer Surface Shape Based Classification Steven McDonagh March 10, 2008 Steven McDonagh Skin Cancer Surface Shape Based Classification Motivation A non-trivial classification problem Clinicians time is valuable Good track

SLIDE 1

Skin Cancer Surface Shape Based Classification

Steven McDonagh March 10, 2008

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 2

Motivation

◮ A non-trivial classification

problem

◮ Clinician’s time is valuable ◮ Good track record if caught

early enough

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 3

Hypothesis

A classification system using a combination of standard and depth based image features is more successful at the task of classifying skin lesion images than a system which uses standard image features alone.

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 4

The system

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 5

The system

◮ Data capture and pre-processing ◮ Feature calculation and selection ◮ Training and classification

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 6

Data capture

Figure: Stereo-scopic geometry Figure: Stereo camera rig

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 7

Data capture

Figure: Lesion colour data Figure: Lesion reconstruction from range data

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 8

Pre-processing

Figure: Sample 1 Figure: Sample 2

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 9

Pre-processing

Figure: Sample 1 Figure: Sample 2

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 10

Pre-processing

◮ Image Segmentation: Separate lesion from surrounding skin

◮ Automated thresholding techniques ◮ Do it manually!

Figure: Hand segmented image

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 11

The system

◮ Data capture and pre-processing ◮ Feature calculation and selection ◮ Training and classification

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 12

Feature calculation

◮ What is a feature? ◮ Standard 2D image based features

◮ Asymmetry ◮ Border irregularity ◮ Colour variegation ◮ Diameter

Figure: Sample features

feature vector X = [−2.42, 412, 0.63, 50.75]

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 13

3D features

◮ ∆spotheight (avg spot z

depth − avg skin z depth)

◮ Rz = σ[z]

spot

σ[z]

skin

local texture roughness ratio

◮ Peak and pit density

#peaks + #pits spot area

Figure: ∆spotheight

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 14

Investigating a feature

∆spotheight

Figure: 1D Scatter plot for i1 feature

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 15

Investigating a feature

∆spotheight

Figure: Scatter plot for i1 feature

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 16

Feature selection

◮ n features? ⇒ 2n subset combinations! ◮ How do we select the best feature subset? ◮ Feature subset selection

◮ Sieve out the irrelevant / redundant features ◮ Goal: Small subset of features that give high predictive

accuracy

◮ For any given image we can now compute a feature vector

X = [f1, f2, ..., fm] m ≤ n

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 17

Figure: Evolution of subset classification accuracy

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 18

The system

◮ Data capture and pre-processing ◮ Feature calculation and selection ◮ Training and classification

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 19

Training and classification

◮ Goal: Given a previously unseen lesion image X ⇒ compute

the most likely class k that it belongs to

◮ k ∈ {Basel cell carcinoma, Squamous cell carcinoma,

Seborrheic keratosis, Melanocytic naevus, Actinic keratosis }

◮ We need to find P(class = k|X) for each class k ◮ Classify new image data X as:

1. Class k which provides highest conditional probability
2. Assign sample X to the class j which minimises the quantity

ΣkLkjP(class = k|X)

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 20

Experimental results

Diagnosis AK BCC ML SCC SK AK 11 100% BCC 57 1 4 3 87.6% True ML 4 48 2 7 78.6% SCC 4 1 19 1 76% SK 4 8 5 55 76.3% Overall accuracy 83.7% Misclassification cost 196

Table: Best feature set found by accuracy based criterion

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 21

Experimental results

Diagnosis AK BCC ML SCC SK AK 11 100% BCC 47 3 14 1 72.3% True ML 4 48 4 5 78.6% SCC 1 23 1 92% SK 6 15 5 46 63.8% Overall accuracy 81.3% Misclassification cost 183

Table: Best feature set found by cost based criterion

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 22

Experimental results

Diagnosis AK BCC ML SCC SK AK 11 100% BCC 55 3 2 5 84.6% True ML 7 49 5 80.3% SCC 9 2 11 3 44% SK 7 8 1 56 77.7% Overall accuracy 77.3% Misclassification cost 306

Table: Best feature set found when constrained to colour feature pool

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 23

Discussion

◮ Novel data capture techniques may provide us with useful

information

◮ Feature calculation and selection methods are just as

important as data quality

◮ Success measure is vital!

Steven McDonagh Skin Cancer Surface Shape Based Classification

SLIDE 24

Thanks for listening!

◮ Questions?

Steven McDonagh Skin Cancer Surface Shape Based Classification