Engineering a Program to Digitally Analyze Genetic Tests for HPV

1
Engineering a Program to Digitally Analyze
Genetic Tests for HPV

• Stephen Trusheim and Daniel Mokhtari, 2007-2008

2
Human Papillomavirus

• Human Papillomavirus (HPV) is the most common
  cause of cervical cancer.
• Cervical cancer is the second most common type of
  cancer in women (Khan et al., 2005).
• HPV has multiple types, ranging from low-risk to
  high-risk for cervical cancer (Muñoz et al.,
  2003).

http//upload.wikimedia.org/wikipedia/en/a/ad/EM_o
f_pap_virus2C_basal_tissue_grafted_to_mouse.jpg
3
Diagnosing HPV

• Diagnosing HPV requires determining the type of
  HPV contained in a patient sample.
• Polyacrylamide gel electrophoresis is used for
  this purpose.

4
Gel Electrophoresis
5
Gel Electrophoresis
6
Gel Electrophoresis
7
Gel Electrophoresis
8
Gel Electrophoresis
9
Gel Electrophoresis
10
Diagnosing HPV

• Currently, a double-blind human review process is
  used to diagnose HPV types in completed
  gel-electrophoresis tests
• Human review is often qualitative, slow, and
  expensive
• Computers have the potential to automate
  interpretation for gel-electrophoresis tests

11
Goals

• To engineer a Java-based program to
  quantitatively interpret gel electrophoresis
  tests for HPV
• To diagnose more than 95 of genetic tests for
  HPV correctly
• To reduce the time, cost, and effort required in
  each gel test

12
Test Interpretation Program
Original Gel Image
Image-Analysis Algorithms
Test-Interpretation Algorithm
Test Result
HPV Type 16 High Risk
13
Test Interpretation Program
Original Gel Image
Image-Analysis Algorithms
Test-Interpretation Algorithm
Test Result
HPV Type 16 High Risk
14
(No Transcript)
15
Test Interpretation Program
Original Gel Image
Image-Analysis Algorithms
Test-Interpretation Algorithm
Test Result
HPV Type 16 High Risk
16
Image Analysis Algorithms

• Orientation
• Lane Detection
• Noise Reduction
• Band Detection

17
Image Analysis Algorithms

• Orientation
• Lane Detection
• Noise Reduction
• Band Detection

18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
Image Analysis Algorithms

• Orientation
• Lane Detection
• Noise Reduction
• Band Detection

26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
Image Analysis Algorithms

• Orientation
• Lane Detection
• Noise Reduction
• Band Detection

32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
Image Analysis Algorithms

• Orientation
• Lane Detection
• Noise Reduction
• Band Detection

37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
(No Transcript)
46
Test Interpretation Program
Original Gel Image
Image-Analysis Algorithms
Test-Interpretation Algorithm
47
(No Transcript)
48
Accepted Ranges for HPV Type 16
49
(No Transcript)
50
Average Correlation of HPV Type 16 78.9
51
(No Transcript)
52
(No Transcript)
53
This patient is diagnosed with HPV Type 16.
54
Test Interpretation Program
Original Gel Image
Image-Analysis Algorithms
Test-Interpretation Algorithm
Test Result
HPV Type 16 High Risk
55
Results
56
Future Studies

• Improve band detection
• Expand test interpretation capabilities for
  Chlamydia, Gonorrhea, Cystic Fibrosis, and
  Inherited Thrombophilia
• Improve program success rate with the goal of
  receiving FDA approval for HPV tests

57
Acknowledgements

• Mrs. Lois Fruen
• Dr. Ronald McGlennen MD
• Mr. Robert Schuldt
• Mr. Jeremy Raadt
• Ms. Bonita Bryant MT(ASCP), CLSpMB(NCA)
• Access Genetics
• Breck Team Research

58
Engineering a Program to Digitally Analyze
Genetic Tests for HPV

• Stephen Trusheim and Daniel Mokhtari, 2007-2008