Healthc Inform Res.  2021 Jan;27(1):82-91. 10.4258/hir.2021.27.1.82.

Codeless Deep Learning of COVID-19 Chest X-Ray Image Dataset with KNIME Analytics Platform

• An JY
• Seo H
• Kim YG
• Lee KE
• Kim S
• Kong HJ

• KMID: 2512628
• DOI: http://doi.org/10.4258/hir.2021.27.1.82

Abstract

Objectives
This paper proposes a method for computer-assisted diagnosis of coronavirus disease 2019 (COVID-19) through chest X-ray imaging using a deep learning model without writing a single line of code using the Konstanz Information Miner (KNIME) analytics platform.
Methods
We obtained 155 samples of posteroanterior chest X-ray images from COVID-19 open dataset repositories to develop a classification model using a simple convolutional neural network (CNN). All of the images contained diagnostic information for COVID-19 and other diseases. The model would classify whether a patient was infected with COVID-19 or not. Eighty percent of the images were used for model training, and the rest were used for testing. The graphic user interface-based programming in the KNIME enabled class label annotation, data preprocessing, CNN model training and testing, performance evaluation, and so on.
Results
1,000 epochs training were performed to test the simple CNN model. The lower and upper bounds of positive predictive value (precision), sensitivity (recall), specificity, and f-measure are 92.3% and 94.4%. Both bounds of the model’s accuracies were equal to 93.5% and 96.6% of the area under the receiver operating characteristic curve for the test set.
Conclusions
In this study, a researcher who does not have basic knowledge of python programming successfully performed deep learning analysis of chest x-ray image dataset using the KNIME independently. The KNIME will reduce the time spent and lower the threshold for deep learning research applied to healthcare.

Keyword

COVID-19, Mass Chest X-Ray, Diagnosis, Computer Assisted, Deep Learning, KNIME

Figure

• 

  Figure 1 Installation procedures of Anaconda.

• 

  Figure 2 Installation procedures of extensions. (A) KNIME extensions should be installed to execute deep learning model. (B) Select “KNIME Deep Learning – Keras Integration.” (C) Select “KNIME Image Processing,” “KNIME Image Processing – Deep Learning Extension,” and “KNIME Image Processing – Python Extensions.” (D) Click “Finish.” (E) Agree to “License agreements.” (F) Restart.

• 

  Figure 3 Conda and Deep Learning Conda environment setting. (A) Select “Python3.” (B) Create “New Conda environment.” (C) Create Python environment in which correct version is matched automatically. (D) Select “Use Special Deep Learning Configuration as Defined Below.” (E) Create “New Deep Learning Conda environment.” (F) Create deep learning environment in which correct version is matched automatically.

• 

  Figure 4 Example of overall preprocessing flow. (A) All nodes for preprocessing. (B) Import filenames of a dataset. (C) Sort by filenames. (D) Classify files as “positive” or “negative.” (E) Perform stratified sampling and Open “Load and Preprocess Images” node by right-clicking the node→component→open. (F) Import components. (G) Add image column before filename column. (H) Normalize images. (I) Resize images by 150 × 150. (J) Filter out other columns except image and class columns.

• 

  Figure 5 Example of overall deep learning flow. (A) All nodes for deep learning flow. (B) Import table of preprocessed dataset. (C) Divide the dataset into two, one for training and the other for testing. (D) Create a convolutional neural network. (E) Apply the trained model. (F) Rename the column of ROC Curve. (G) Scores the model. (H) Show area under curve. (I) Train the model.

• 

  Figure 6 Architecture of the simple convolutional neural network algorithm used in the study.

• 

  Figure 7 Performance of a simple convolutional neural network (CNN) model in detecting COVID-19: “Scorer” node shows (A) accuracy statistics (true positive, false positive, true negative, false negative, sensitivity, specificity, F-measure, accuracy and Cohen’s kappa; (B) confusion matrix (accuracy, Cohen’s kappa). (C) “ROC Curve” node shows area under the curve.

Reference

References

1. Grand View Research. Artificial intelligence in healthcare market size, share & trends analysis report by component (hardware, software, services), by application, by region, competitive insights, and segment forecasts, 2019–2025 [Internet]. San Francisco (CA): Grand View Research;2019. [cited at 2021 Jan 27]. Available from: https://www.grand-viewresearch.com/industry-analysis/artificial-intelligence-ai-healthcare-market?utm_source=prnewswire&utm_medium=referral&utm_campaign=hc_16-dec-19&utm_term=ai-healthcare-market&utm_content=rd/toc.

2. Guo Y, Hao Z, Zhao S, Gong J, Yang F. Artificial intelligence in health care: bibliometric analysis. J Med Internet Res. 2020; 22(7):e18228.

3. Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol. 2017; 2(4):230–43.

4. Fontani D. Keras UI: a GUI to manage image classification [Internet] [place unknown]. Code Project. 2019. [cited at 2021 Jan 27]. Available from: https://www.codeproject.com/Articles/4053651/Keras-UI-A-GUI-to-Manage-Image-Classification.

5. Blume PA. The LabVIEW style book. Boston (MA): Prentice-Hall;2011.

6. Frank E, Hall M, Holmes G, Kirkby R, Pfahringer B, Witten IH, et al. Weka: a machine learning workbench for data mining. Maimon O, Rokach L, editors. Data mining and knowledge discovery handbook. Boston (MA): Springer;2009. p. 1269–77.

7. KNIME Analytics Platform [Internet]. Zurich, Switzerland: KNIME AG;c2020. [cited at 2021 Jan 27]. Available from: https://www.knime.com/knime-analytics-platform.

8. Lin Q, Zhao S, Gao D, Lou Y, Yang S, Musa SS, et al. A conceptual model for the coronavirus disease 2019 (COVID-19) outbreak in Wuhan, China with individual reaction and governmental action. Int J Infect Dis. 2020; 93:211–6.

9. RSNA pneumonia detection challenge [Internet]. San Francisco (CA): Kaggle.com;2018. [cited at 2021 Jan 27]. Available from: https://www.kaggle.com/c/rsna-pneumonia-detection-challenge.

10. Cohen JP, Morrison P, Dao L, Roth K, Duong TQ, Ghassemi M. Covid-19 image data collection: prospective predictions are the future [Internet]. Ithaca (NY): arXiv.org;2020. [cited at 2020 Jan 28]. Available from: https://arxiv.org/abs/2006.11988.

11. Shaheen F, Verma B, Asafuddoula M. Impact of automatic feature extraction in deep learning architecture. In : Proceedings of 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA); 2016 Nov 30–Dec 2; Gold Coast, Australia. p. 1–8.

12. Vinayakumar R, Soman KP, Poornachandran P. Applying convolutional neural network for network intrusion detection. In : Proceedings of 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI); 2017 Sep 13–16; Udupi, India. p. 1222–8.

13. Swapna G, Soman Kp, Vinayakumar R. Automated detection of diabetes using CNN and CNN-LSTM network and heart rate signals. Procedia computer science. 2018; 132:1253–62.

14. Gao T. Chest X-ray image analysis and classification for COVID-19 pneumonia detection using deep CNN. medRxiv. 2020. [Epub]. https://doi.org/10.1101/2020.08.20.20178913.

15. Ismael AM, Sengur A. Deep learning approaches for COVID-19 detection based on chest X-ray images. Expert Syst Appl. 2021; 164:114054.

16. Maghdid HS, Asaad AT, Ghafoor KZ, Sadiq AS, Khan MK. Diagnosing COVID-19 pneumonia from X-ray and CT images using deep learning and transfer learning algorithms [Internet]. Ithaca (NY): arXiv.org;2020. [cited at 2020 Jan 28]. Available from: https://arxiv.org/abs/2004.00038.

17. Phillips NA, Rajpurkar P, Sabini M, Krishnan R, Zhou S, Pareek A, et al. Chexphoto: 10,000+ smartphone photos and synthetic photographic transformations of chest x-rays for benchmarking deep learning robustness [Internet]. Ithaca (NY): arXiv.org;2020. [cited at 2020 Jan 28]. Available from: https://arxiv.org/abs/2007.06199.