new biomedical applications of digital image processing … · 2013. 2. 28. · • carlos lopes...
Post on 26-Sep-2020
0 Views
Preview:
TRANSCRIPT
José Fonseca André Mora jmf@uninova.pt atm@uninova.pt
Biomedical Applications of
Digital Image Processing
Biomedical Applications of Digital Image Processing
• Retinography applications
• Biology and Bacteria's applications
• Conclusions
JORTEC 2013 - José Fonseca & André Mora 2
OUTLINE
RETINAL DRUSEN DETECTION
JORTEC 2013 - José Fonseca & André Mora 3
• Drusen are retina abnormalities caused by accumulation of extracellular materials beneath retina surface
• ARMD indicator and can produce loss of vision
• Visible in retina Fundus photography as yellowish spots around the macula
• Objective: To detect and quantify affected area
JORTEC 2013 - José Fonseca & André Mora 4
DRUSEN DETECTION
Transversal view
Surface view
• Assumption: retina images allow 3D shape estimation
– In a 2D grey scale image the intensity value can represent the
3rd coordinate (depth)
• Proposed algorithm:
– Step 1 – Illumination correction
– Step 2 – detection of retina spots that can be related to Drusen
– Step 3 – modelling of each retina spot using a 3D function
JORTEC 2013 - José Fonseca & André Mora 5
CONSIDERATIONS
2D view Estimated
3D view
JORTEC 2013 - José Fonseca & André Mora 6
AD3RI METHODOLOGY
ROI Definition Correct illumination Locate Drusen spots
(GPL algorithm) Green Channel Norm. Contrast
7
GRADIENT PATH LABELLING (GPL)
11
11
10
10
10
10
2
Image gradient (Sobel operator)
Labels propagation
Step1
Step2 Step3
Apply label
compatibilities
3
12 13
14 15 16 17 13 18
19 20 21 18 22
23 24 25 25 25 22 26
27 28 29 30 31 32 33
34 35 36 37 37 38 39 40
1
1
1
5
5
4
4
6
6
6 6
8 7
7
7
7
7 7
7
7
9
1 1 1 1 1 1 7 7
1 1 1 1 1 7 7 7
1 1 1 1 7 7 7 7
1 1 1 7 7 7 7 7
1 1 7 7 7 7 7 7
1 7 7 7 7 7 7 7
1 7 7 7 7 7 7 7
1 7 7 7 7 7 7 7
Ori
gin
al im
ag
e
JORTEC 2013 - José Fonseca & André Mora
JORTEC 2013 - José Fonseca & André Mora 8
AD3RI METHODOLOGY
ROI Definition Correct illumination Locate Drusen spots
(GPL algorithm)
Model Drusen spots Threshold model
Green Channel Norm. Contrast
JORTEC 2013 - José Fonseca & André Mora 9
EVALUATION TESTS – PIXEL LEVEL
AD3RI
MD3RI
Pixel level Comparison
(TP, FP, TN, FN)
Pixel level analysis
Sensitivity / Specificity
Kappa coefficient
MD3RI AD3RI
SS = 0.69
SP = 0.96
K = 0.58
IMAGE ARTIFACTS DETECTION
JORTEC 2013 - José Fonseca & André Mora 10
• Images’ quality can be affected by: – Bad collaboration from the patient
– Dust over the lens
• The effects are: – Non-uniform illumination
– Flares (artifacts)
– Central Artifact
• MSc João Soares MIEBM
JORTEC 2013 - José Fonseca & André Mora 11
FUNDUS IMAGE ACQUISITION PROBLEMS
Fingerprint
JORTEC 2013 - José Fonseca & André Mora 12
DETECTION PROCEDURE
Template Matching
Saturation Channel
Image Template
Flare validation
Image and Template
correlation
ARTERIOLAR TO VENULAR RATIO (AVR)
JORTEC 2013 - José Fonseca & André Mora 13
• The retina arteriolar to venular diameter ratio is suggested to reflect generalized vascular alterations and to predict the risk of cardiovascular diseases.
• Procedure: – Optic disk detection
– Vessels detection
– Arteriole / Venule Classification
– AVR calculation
– MSc Student Tânia Tomaz MIEBM
JORTEC 2013 - José Fonseca & André Mora 14
ARTERIOLAR TO VENULAR RATIO (AVR)
JORTEC 2013 - José Fonseca & André Mora 15
OPTIC DISK DETECTION
Morphologic Close
Edge Detection
Threshold Hough
Transform
JORTEC 2013 - José Fonseca & André Mora 16
VESSELS DETECTION AND CLASSIFICATION
Circular projection
Vessels detection
Vessels classification
DETECT ABNORMAL IMAGES
JORTEC 2013 - José Fonseca & André Mora 17
– Objective: On screening and using Retinal Auto-fluorescence images, detect patients that have retinal abnormalities and need to visit their ophthalmologist
– Procedure: • Detect deviations from a normal retina
auto-fluorescence pattern
– MSc Students: • Sara Praça MIEBM
• Carlos Lopes MIEEC
JORTEC 2013 - José Fonseca & André Mora 18
DETECTION OF ABNORMAL IMAGES
Norm
al
Optomap
(ultra-wide field) HRA
Ab
no
rma
l
BACTERIA SEGMENTATION
JORTEC 2013 - José Fonseca & André Mora 19
• The need of reliable and fast methods of single cell evaluation for statistically significant studies
• Extraction of individual cells information from bacterial populations with large number of elements
• Avoid manual analysis which is fastidious, time consuming and subject to observer variances
JORTEC 2013 - José Fonseca & André Mora 20
WHY BACTERIA SEGMENTATION AND TRACKING?
JORTEC 2013 - José Fonseca & André Mora 21
THE PROBLEM
DIC - Diferential Interference Contrast Microscopy
image Confocal Fluorescence Microscopy image
JORTEC 2013 - José Fonseca & André Mora 22
IMAGE PROCESSING STEPS
DIC image
Segmented
image
Cleaned
image
GPL
Discard classifier
Merge classifier
Merged
image
Fluorescence
image
Cells evaluation
Image alignment
Aligned
image 1
Aligned
image 2
Aligned
image 3
Aligned
image 4
Aligned
image n
Temporal filtering
JORTEC 2013 - José Fonseca & André Mora 23
BACTERIA (OVER)SEGMENTATON
Gradient Path Labelling (Sobel operator) Original image
JORTEC 2013 - José Fonseca & André Mora 24
OVERSEGMENTED IMAGE
JORTEC 2013 - José Fonseca & André Mora 25
“CLEANED” IMAGE BY CART DECISION TREE
JORTEC 2013 - José Fonseca & André Mora 26
“MERGED IMAGE” BY CART DECISION TREE
JORTEC 2013 - José Fonseca & André Mora 27
DIC AND FLUORESCENCE IMAGES ALIGNMENT
JORTEC 2013 - José Fonseca & André Mora 28
DIC AND FLUORESCENCE IMAGES ALIGNED
JORTEC 2013 - José Fonseca & André Mora 29
INTER-FRAME CORRECTION
frame n-1 frame n frame n+1
frame n after inter-frame correction
JORTEC 2013 - José Fonseca & André Mora 30
INDIVIDUAL CELLS FLUORESCENCE EVALUATION
• This work was developed in close collaboration with the Laboratory of Biosystem Dynamics, Department of Signal Processing, Tampere University of Technology (TUT)
• Special thanks to Professor André Ribeiro (TUT)
• Local contributors: – MSc Leonardo Martins (finished) - MIEBM
– MSc Cátia Queimadelas (finished) - MIEBM
– MSc João Rodrigues (on going) - MIEBM
– MSc Pedro Ferreira (just started) – MIEEC
JORTEC 2013 - José Fonseca & André Mora 31
CREDITS
• The proposed method shown to be able to extract the necessary information from individual cells with acceptable error
• The method can be used to support studies if gene expression dynamics using tsr-venus proteins to assess the kinetics of expression of the gene of interest
• The evaluation of more time-series and possible improvements on image quality will certainly lead to error reduction and improved cell evaluation
JORTEC 2013 - José Fonseca & André Mora 32
CONCLUSIONS
THANK YOU FOR YOUR ATTENTION
JORTEC 2013 - José Fonseca & André Mora 33
top related