search

toniesteves / covid19-chest-x-ray-detection

Detecção de COVID-19 a partir de imagens de raios X de tórax utilizando uma Deep Convolutional Neural Network otimizada.
https://bit.ly/3aeX77s
6 stars 4 forks source link
analysis-data computer-vision coronavirus coronavirus-analysis coronavirus-globaloutbreak coronavirus-info coronavirus-tracking covid-19 data-science data-visualization deep-learning disease-modeling disease-prediction disease-spread disease-surveillance infectious-disease infectious-disease-models infectious-diseases public-health public-health-care

readme

COVID-19 Detection from Chest X-Ray Images

Domain             : Computer Vision, Machine Learning
Sub-Domain         : Deep Learning, Image Recognition
Techniques         : Deep Convolutional Neural Network, ImageNet, Inception
Application        : Image Recognition, Image Classification, Medical Imaging

Description

1. Detecção de COVID-19  a partir de imagens de raios X de tórax utlizando uma Deep Convolutional Neural Network otimizada.

Code

GitHub Link          : COVID-19 Detection from Chest X-Ray Images 
Linkedin             : Antonio Esteves

Datasets

Dataset Name     : Chest X-Ray Images (Pneumonia)
Dataset Link     : Chest X-Ray Images (Pneumonia) Dataset (Kaggle)
                 : Chest X-Ray Images (Pneumonia) Dataset (Original Dataset - No Labeled)
Dataset Name     : COVID-19 image data collection
Dataset Link     : COVID-19 image data collection (Original Dataset)


Detalhes do Dataset
Nome do Dataset              : Imagens de raio X de toráx (COVID-19)
Número de Classes            : 2
Número/Tamanho das imagens   : Total      : 178 (98.8 Megabyte (MB))
                               Treino     : 76  (51.7 Megabyte (MB))
                               Validação  : 30  (9.1  Megabyte (MB))
                               Teste      : 72  (38.4 Megabyte (MB))

Parâmetros do Modelo
Machine Learning Library     : Keras
Base Model                   : Custom Deep Convolutional Neural Network
Otimizadores                 : Adam
Função de Perda              : categorical_crossentropy

Deep Convolutional Neural Network Otimizada: 
Parâmetros de Treino
Batch Size                   : 64
Número of Épocas             : 100
Tempo de Treino              : 40 Minutes

Saída (Prediction/ Recognition / Classification Metrics)
Teste
F1-Score                     : 84.79%
Accuracy                     : 83.33%
Loss                         : 0.07
Precision                    : 82%
Recall (COVID-19)            : 86.11% (Para as classes positivas)
Specificity                  : 80.56%
Sample Output:

See More Images

Confusion Matrix:

Confusion Matrix

ROC Curve:

ROC Curve

Tools / Libraries

Languages               : Python
Libraries               : Keras, TensorFlow, Inception, ImageNet

Dates

Duration                : March 2020 - Current
Current Version         : v1.0.0.0
Last Update             : 23.03.2020