in individuals and wit ith Park rkinson's dis isease usin ing - - PowerPoint PPT Presentation

▶
in individuals and wit ith park rkinson s dis isease usin
SMART_READER_LITE
LIVE PREVIEW

in individuals and wit ith Park rkinson's dis isease usin ing - - PowerPoint PPT Presentation

May 23, 2023 112 likes •283 views

Cla lassification of f spiral im images of f healthy in individuals and wit ith Park rkinson's dis isease usin ing convolutional neural networks Joo Paulo Folador jpfolador@gmail.com, doctoral student Prof. Adriano O. Andrade, PhD

slide-1
SLIDE 1

Cla lassification of f spiral im images of f healthy in individuals and wit ith Park rkinson's dis isease usin ing convolutional neural networks

João Paulo Folador

jpfolador@gmail.com, doctoral student

  • Prof. Adriano O. Andrade, PhD

adriano@ufu.br

Centre for Innovation and Technology Assessment in Health, Postgraduate Program in Electrical and Biomedical Engineering, Faculty of Electrical Engineering, Federal University of Uberlândia, Uberlândia, Brazil http://www.niats.feelt.ufu.br/

slide-2
SLIDE 2

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Parkinson’s disease

2

slide-3
SLIDE 3

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Motivation

  • Parkinson's disease (PD) is present in about 1% of the world's

population over 65 years, and still remains incurable.

  • PD is a disease that has a difficult diagnosis.
  • Know the various symptoms is the key to the correct diagnosis

and understanding of the disease.

  • Techniques involving Artificial Intelligence have been applied to

aid in the detection of symptoms, and techniques involving deep learning have achieved more expressive results than traditional techniques.

3

slide-4
SLIDE 4

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Data collection

  • The Research Ethics Committee of the Federal University of

Uberlândia approved the research under the number 07075413.6.0000.5152.

4

Group Total Sex (F/M) Age (years) Health 12 8/4 60,08 ± 6,13 PD 15 7/8 65,33 ± 9,17

slide-5
SLIDE 5

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Data collection

5

  • Original drawings done

by the participants

  • First, the participant

followed the model of Archimedes' spiral and then performed the drawing freely

  • The images were scanned

and preprocessed (Gimp software was used in this step)

slide-6
SLIDE 6

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Data collection

  • Each individual drew

about 3 (three) or 4 (four) spirals

  • The spirals were resized

to width and height of 256 x 256 pixels.

  • 51 images was collected

from each group, totalizing 102 images.

6

slide-7
SLIDE 7

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Convolutional Neural Network (C (CNN)

7

1: Input image of Convolutional Neural Network 2: Convolution layer that yields the feature maps 3: Pooling layer is used to dimensionality reduction 4: The Fully connected layer represents a vector with all features to classify the images (it looks like a multilayer perceptron network - MLP) 5: The last layer has one neuron (unit) to classify between two kinds of classes

slide-8
SLIDE 8

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Example of f a CNN

8

Reference: https://developer.nvidia.com/discover/convolutional-neural-network

slide-9
SLIDE 9

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Data augmentation

A technique to increase the data, there is the increase of the number of samples modifying the original sample and then apply it in CNN

  • Rotation (rotation_range=20°)
  • Vertical flipping (vertical_flip=true)
  • Shear (shear_range=0.2)
  • Horizontal shifting (width_shift_range=0.2)
  • Zoom (zoom_range=0.2)
  • Rescale (rescale=1./255)

9

  • riginal

rotation vertical flipping zoom

slide-10
SLIDE 10

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Development environment

  • API Keras, a library to explore the machine learning techniques
  • Program language Python 3.5
  • TensorFlow 1.2, a engine to work with machine learning
  • The CPU and GPU process all the calculus in parallel by the

library CUDA from NVidia

  • Intel i7 2.4 GHz + 8 GB RAM DDR 3 + Video board de 2GB Nvidia

GT 650

10

slide-11
SLIDE 11

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Results

11

75% training and 25% validation

slide-12
SLIDE 12

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Results

12

Test Steps per epoch Epochs Nº of convolution layers Average validation accuracy Specificity Time spent (s) 1 100 10 2 73.4 % 62.0 % 64 2 200 10 2 75.2 % 60.4 % 158 3 800 10 2 78.9 % 64.8 % 727 4 1000 10 2 78.0 % 59.8 % 1854

Accuracy is expected to measure how well the test predicts both categories Specificity the ability of the system to accurately predict the absence

  • f the condition for cases that do not actually have it.
slide-13
SLIDE 13

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Results

13

73.40% 75.20% 78.90% 78.00% 62.00% 60.40% 64.80% 59.80%

64 158 727 1854

200 400 600 800 1000 1200 1400 1600 1800 2000 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 1 2 3 4

Average validation accuracy Specificity Time spent (s)

slide-14
SLIDE 14

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Discuss and conclusion

  • The

classic configuration

  • f

CNN

  • btained

a satisfactory classification (average of 76.3%) in the identification of healthy individuals and Parkinson's disease spirals.

  • Larger data volume is required to perform other tests and get

better results

  • We need refine the network parameters, test other error

calculation functions other than the mean squared error, etc.

  • Test another architecture CNNs
  • A simple CNN network with few images brought a satisfactory

result illustrating the high performance of the Deep Learning techniques

14

slide-15
SLIDE 15

Núcleo de Inovação e Avaliação Tecnológica em Saúde

Acknowledgements

Thank you!

CAPES - Programa CAPES / DFATD-88887.159028 / 2017-00 FAPEMIG-APQ-00942-17

  • A. O. Andrade é Bolsista de Produtividade do CNPq, Brasil (304818/2018-6 e

305223 / 2014-3)

15

slide-16
SLIDE 16

Cla lassification of f spiral im images of f healthy in individuals and wit ith Park rkinson's dis isease usin ing convolutional neural networks

João Paulo Folador

jpfolador@gmail.com, doctoral student

  • Prof. Adriano O. Andrade, PhD

adriano@ufu.br

Centre for Innovation and Technology Assessment in Health, Postgraduate Program in Electrical and Biomedical Engineering, Faculty of Electrical Engineering, Federal University of Uberlândia, Uberlândia, Brazil http://www.niats.feelt.ufu.br/