visionpro® viditm deep learning-based image analysis software  (Cognex Inc)

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: The VisionPro Deep Learning GUI loaded with the X-ray images from the COVIDx dataset . On the left of the GUI, there are options to select various parameters for training the model, such as model type, model size, epoch count, minimum epochs and patience, train and validation split, class weights, threshold, heatmap and the different data augmentation options of flip, rotation, contrast, zoom, brightness, sharpen, blur, distortion and noise. In the middle, the selected image is shown. On the right, thumbnails of all the images in the training and test set are shown. On the top, there is the tool selection option. In the figure, the green tool has been selected for classification. Clicking on the ‘brain’ shaped icon in the green tool, starts the training of the model

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Biomarker Discovery, Selection

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: The segmented lungs after training the Red tool in VisionPro Deep Learning. Anything outside the segmented lungs is not considered to be part of the Region of Interest (ROI) and is not used for classification. This makes sure that VisionPro Deep Learning trains only on the lungs and not on the artefacts around it

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques:

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: Left: confusion matrix on the 300 test images COGNEX VisionPro Deep Learning with entire ROI selected. Confusion matrix on the 300 test images. COGNEX VisionPro Deep Learning with entire ROI selected. Right: interpretation of the confusion matrix. VisionPro Deep learning GUI does not display numbers of correctly classified or misclassified images on the confusion matrix, but if any point on the confusion matrix is clicked, it displays not only the number of images in that category, but also all of the images belonging to that category, with the prediction percentage and whether the prediction it made is correct or not

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques:

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: A snippet of the report generated on the 300 test images by VisionPro Deep Learning with the entire image selected as the ROI. The report contains the confusion matrix with the evaluation metrics: sensitivity (recall), positive predictive value (precision) and F score for each class. The test images are also shown with the correct labels, the predicted labels and the confidence percentage of each class. In this image, 5 images are classified correctly, and 1 image misclassified (marked in red)

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Generated

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: A snippet of the report generated on the 300 test images by VisionPro Deep Learning with the segmented lungs as the ROI. In this image, all four images are classified correctly

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Generated

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: Four COVID-19 X-ray images from the test set of the COVIDx dataset along with the predicted heatmaps generated by VisionPro Deep Learning. Heatmaps can be a great indicator for radiologists to identify whether the predictions made by the Deep learning algorithm is based on actual infection or some artefacts

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Generated, Infection

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: (a) Sensitivity for each infection type, (b) sensitivity calculated with 95% confidence Interval

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Infection

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: (a) Positive predictive value (PPV) for each infection type, (b) positive predictive value (PPV) calculated with 95% confidence Interval

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Infection

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: (a) F score for each infection type, (b) F score calculated with 95% confidence interval

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: Infection

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: Sensitivity, positive predictive value and F score with 95% confidence interval on the previous COVIDx dataset

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques:

Journal: Sn Computer Science

Article Title: Identification of Images of COVID-19 from Chest X-rays Using Deep Learning: Comparing COGNEX VisionPro Deep Learning 1.0™ Software with Open Source Convolutional Neural Networks

doi: 10.1007/s42979-021-00496-w

Figure Lengend Snippet: VisionPro Deep Learning tested on a previous version of the COVIDx dataset. This dataset has many more images for the test set, in the Normal and Non-COVID-19 classes, but only 91 images in the COVID-19 class. We see the confidence interval improve significantly in classes with a higher number of test images

Article Snippet: COGNEX VisionPro Deep Learning software is provided by COGNEX Corporation.

Techniques: