Chest X-Rays have been used in most studies for multilabel disease classification and are the main resource for medical image analysis. Chest X-Ray images are obtained utilizing penetration imaging technology, which produces multiple levels of characteristics in an image as opposed to natural images. As a result, it is challenging to identify a disease's characteristics for further diagnosis. In practice, the proportion of healthy persons outnumbers those with various diseases, and the abilities of those diseases vary. Therefore, the labels are unbalanced. Two major challenges of diagnosis through chest X-ray images are extracting discriminative features from X-ray images and handling imbalanced data distribution. The preprocessing method is intended to overcome the issue of imbalanced data distribution.

This research uses modified DenseNet169, called, DenseNet-Proposed1, DenseNet-Proposed2, DenseNet-Proposed3, DenseNet-Proposed4, and DenseNet-Proposed5 for the classification of multilabel chest X-Ray. We evaluate our proposed models using different loss functions, namely Binary Cross entropy (BCE) and Weighted Cross Entropy (WCE) loss. In addition, we also compared our proposed model with other 6 standard deep learning models, i.e. DenseNet121, DenseNet169, DenseNet201, ResNet50, MobileNetV2, and Xception. Utilizing the National Institute of Health (NIH) ChestXray14 dataset, we evaluate the effectiveness of our algorithm against the other cutting-edge approaches for classifying pathologies.

The major comparison is to examine the results between classifiers using the Area Under the Curve (AUC), precision, recall, F1-score, and accuracy. Finally, we conclude that DenseNet-Proposed3 with WCE loss achieved the best performance when compared to other methods. We observe an extensive range outcome in our proposed model, where Hernia received the model's highest AUC of 0.979, in contrast, Infiltration received the lowest AUC of 0.678.

[1] T. Agrawal and P. Choudhary, “Segmentation and classification on chest radiography: a systematic survey,” Vis. Comput., vol. 39, no. 3, pp. 875–913, 2022, doi: 10.1007/s00371-021-02352-7.
[2] S. Candemir and S. Antani, “A review on lung boundary detection in chest X-rays,” Int. J. Comput. Assist. Radiol. Surg., vol. 14, no. 4, pp. 563–576, 2019, doi: 10.1007/s11548-019-01917-1.
[3] H. Sharma, J. S. Jain, P. Bansal, and S. Gupta, “Feature extraction and classification of chest X-ray images using CNN to detect pneumonia,” Proc. Conflu. 2020 - 10th Int. Conf. Cloud Comput. Data Sci. Eng., pp. 227–231, 2020, doi: 10.1109/Confluence47617.2020.9057809.
[4] A. Giełczyk, A. Marciniak, M. Tarczewska, and Z. Lutowski, “Pre-processing methods in chest X-ray image classification,” PLoS One, vol. 17, no. 4 April, pp. 1–11, 2022, doi: 10.1371/journal.pone.0265949.
[5] G. Litjens et al., “A survey on deep learning in medical image analysis,” Med. Image Anal., vol. 42, no. 1995, pp. 60–88, 2017, doi: 10.1016/j.media.2017.07.005.
[6] S. Palani, A. Kulkarni, A. Kochara, and K. M, “Detection of Thoracic Diseases using Deep Learning,” ITM Web Conf., vol. 32, p. 03024, 2020.
[7] T. Ching et al., Opportunities and obstacles for deep learning in biology and medicine, vol. 15, no. 141. 2018.
[8] X. Wang, Y. Peng, L. Lu, Z. Lu, M. Bagheri, and R. M. Summers, “ChestX-ray8: Hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases,” Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, vol. 2017-Janua, pp. 3462–3471, 2017, doi: 10.1109/CVPR.2017.369.
[9] J. Irvin et al., “CheXpert: A large chest radiograph dataset with uncertainty labels and expert comparison,” 33rd AAAI Conf. Artif. Intell. AAAI 2019, 31st Innov. Appl. Artif. Intell. Conf. IAAI 2019 9th AAAI Symp. Educ. Adv. Artif. Intell. EAAI 2019, pp. 590–597, 2019, doi: 10.1609/aaai.v33i01.3301590.
[10] A. E. W. Johnson et al., “MIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs,” vol. 14, pp. 1–7, 2019, [Online]. Available: http://arxiv.org/abs/1901.07042.
[11] A. Bustos, A. Pertusa, J. M. Salinas, and M. de la Iglesia-Vayá, “PadChest: A large chest x-ray image dataset with multilabel annotated reports,” Med. Image Anal., vol. 66, pp. 1–35, 2020, doi: 10.1016/j.media.2020.101797.
[12] F. Ahmad, A. Farooq, and M. U. Ghani, “Deep Ensemble Model for Classification of Novel Coronavirus in Chest X-Ray Images,” vol. 2021, 2021.
[13] I. Allaouzi and M. Ben Ahmed, “A Novel Approach for Multilabel Chest X-Ray Classification of Common Thorax Diseases,” IEEE Access, vol. 7, pp. 64279–64288, 2019, doi: 10.1109/ACCESS.2019.2916849.
[14] L. Seyyed-Kalantari, G. Liu, M. McDermott, I. Y. Chen, and M. Ghassemi, “CheXclusion: Fairness gaps in deep chest X-ray classifiers,” Pac. Symp. Biocomput., vol. 26, pp. 232–243, 2021, doi: 10.1142/9789811232701_0022.
[15] I. M. Baltruschat, H. Nickisch, M. Grass, T. Knopp, and A. Saalbach, “Comparison of Deep Learning Approaches for Multilabel Chest X-Ray Classification,” Sci. Rep., vol. 9, no. 1, pp. 1–11, 2019, doi: 10.1038/s41598-019-42294-8.
[16] J. Xu, H. Li, and X. Li, “MS-ANet: deep Learning for Automated Multilabel Thoracic Disease Detection and Classification,” PeerJ Comput. Sci., vol. 7, pp. 1–12, 2021, doi: 10.7717/PEERJ-CS.541.
[17] M. S. Majdi, K. N. Salman, M. F. Morris, N. C. Merchant, and J. J. Rodriguez, “DEEP LEARNING CLASSIFICATION OF CHEST X-RAY IMAGES,” 2020 IEEE Southwest Symp. Image Anal. Interpret., pp. 116–119, 2020.
[18] G. A. Shadeed, M. A. Tawfeeq, and S. M. Mahmoud, “Deep learning model for thorax diseases detection,” Telkomnika (Telecommunication Comput. Electron. Control., vol. 18, no. 1, pp. 441–449, 2020, doi: 10.12928/TELKOMNIKA.v18i1.12997.
[19] Z. Ge, D. Mahapatra, S. Sedai, R. Garnavi, and R. Chakravorty, “Chest X-rays Classification: A Multilabel and Fine-Grained Problem,” pp. 1–9, 2018, [Online]. Available: http://arxiv.org/abs/1807.07247.
[20] A. S. Pillai, “Multilabel Chest X-Ray Classification via Deep Learning,” J. Intell. Learn. Syst. Appl., vol. 14, no. 04, pp. 43–56, 2022, doi: 10.4236/jilsa.2022.144004.
[21] D. Bhusal and D. S. P. Panday, “Multilabel Classification of Thoracic Diseases using Dense Convolutional Network on Chest Radiographs,” pp. 1–11, 2022, [Online]. Available: http://arxiv.org/abs/2202.03583.
[22] Z. Ge, D. Mahapatra, X. Chang, Z. Chen, L. Chi, and H. Lu, “Improving multilabel chest X-ray disease diagnosis by exploiting disease and health labels dependencies,” Multimed. Tools Appl., vol. 79, no. 21–22, pp. 14889–14902, 2020, doi: 10.1007/s11042-019-08260-2.
[23] X. Xu, Q. Guo, J. Guo, and Z. Yi, “DeepCXray: Automatically diagnosing diseases on chest x-rays using deep neural networks,” IEEE Access, vol. 6, pp. 66972–66982, 2018, doi: 10.1109/ACCESS.2018.2875406.
[24] H. Wang and Y. Xia, “ChestNet: A Deep Neural Network for Classification of Thoracic Diseases on Chest Radiography,” pp. 1–8, 2018, [Online]. Available: http://arxiv.org/abs/1807.03058.
[25] S. Xu, H. Wu, and R. Bie, “CXNet-m1: Anomaly Detection on Chest X-Rays with Image-Based Deep Learning,” IEEE Access, vol. 7, pp. 4466–4477, 2019, doi: 10.1109/ACCESS.2018.2885997.
[26] P. Rajpurkar et al., “CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning,” pp. 3–9, 2017, [Online]. Available: http://arxiv.org/abs/1711.05225.
[27] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, vol. 2017-Janua, pp. 2261–2269, 2017, doi: 10.1109/CVPR.2017.243.
[28] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 770–778, 2016, doi: 10.1109/CVPR.2016.90.
[29] A. G. Howard et al., “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,” 2017, [Online]. Available: http://arxiv.org/abs/1704.04861.
[30] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, vol. 2017-Janua, pp. 1800–1807, 2017, doi: 10.1109/CVPR.2017.195.
[31] M. L. Zhang and Z. H. Zhou, “A review on multilabel learning algorithms,” IEEE Trans. Knowl. Data Eng., vol. 26, no. 8, pp. 1819–1837, 2014, doi: 10.1109/TKDE.2013.39.
[32] R. Venkatesan and M. J. Er, “Multilabel classification method based on extreme learning machines,” 2014 13th Int. Conf. Control Autom. Robot. Vision, ICARCV 2014, pp. 619–624, 2014, doi: 10.1109/ICARCV.2014.7064375.