Development of a Multi-Task Learning CNN Model for Pneumonia Detection and Pathogen Classification Based on Medical Images
How to cite (JITCE) :
Pneumonia is one of the leading causes of death from respiratory tract infections worldwide. Early detection and identification of the causative pathogen are crucial for determining appropriate treatment. This study aims to develop a Convolutional Neural Network (CNN) model based on Multi-Task Learning (MTL) to simultaneously detect pneumonia and classify the type of pathogen through chest X-ray images. The model architecture uses a shared convolutional layer for feature extraction, which then branches into two classification paths. The model was trained using a dataset of X-ray images labeled with disease status and pathogen type, with two loss functions optimized simultaneously. Based on the training process and model architecture design, the estimated accuracy achieved is approximately 92% for pneumonia detection and 89% for pathogen type classification. These results indicate that the CNN-MTL approach is effective and efficient in simultaneously addressing two clinical tasks. The proposed model has the potential to be applied as a clinical decision support system, particularly in healthcare facilities with limited resources.
[1] D. N. Rohmah, “Management Kasus Gagal Nafas Pada Penyakit Pneumonia,” J. Ber. Ilmu Keperawatan, vol. 13, no. 1, pp. 22–30, 2020.
[2] M. A. Astuti and N. Nurhaeni, “Pengaruh Durasi Menyusui terhadap Kejadian Pneumonia pada Balita,” Keperawatan Univ. Indones., no. 2013, pp. 99–106, 2018, [Online]. Available: https://journal.stikespemkabjombang.ac.id/index.php/jikep/article/download/240/382/1235
[3] I. K. Dewi, N. Setiyawati, and D. Estiwidani, “Factors Affecting Pneumonia among Children Under Five Years Old,” J. Kesehat. Ibu dan Anak, vol. 13, no. 2, pp. 88–96, 2019, [Online]. Available: http://e-journal.poltekkesjogja.ac.id/index.php/kia/
[4] S. Vandenhende, S. Georgoulis, W. Van Gansbeke, M. Proesmans, D. Dai, and L. Van Gool, “Multi-Task Learning for Dense Prediction Tasks: A Survey,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 7, pp. 3614–3633, 2022, doi: 10.1109/TPAMI.2021.3054719.
[5] R. I. Stantchev, X. Chen, and J. Hardwicke, “The version presented in WRAP is the author ’ s accepted manuscript and may differ from the Selecting Effective Blockchain Solutions,” pp. 0–34, 2020.
[6] J. Huang, K. Zhou, A. Xiong, and D. Li, “Smart Contract Vulnerability Detection Model Based on Multi‐Task Learning,” Sensors, vol. 22, no. 5, pp. 1–24, 2022, doi: 10.3390/s22051829.
[7] G. Crichton, S. Pyysalo, B. Chiu, and A. Korhonen, “A neural network multi-task learning approach to biomedical named entity recognition,” BMC Bioinformatics, vol. 18, no. 1, pp. 1–14, 2017, doi: 10.1186/s12859-017-1776-8.
[8] Z. Zhao, Z. Zhang, P. Tang, X. Wang, and L. Cui, “MT-GN: Multi-Task-Learning-Based Graph Residual Network for Tropical Cyclone Intensity Estimation,” Remote Sens., vol. 16, no. 2, pp. 1–22, 2024, doi: 10.3390/rs16020215.
[9] T. Zeng and S. Ji, “Deep convolutional neural networks for multi-instance multi-task learning,” Proc. - IEEE Int. Conf. Data Mining, ICDM, vol. 2016-Janua, no. October, pp. 579–588, 2016, doi: 10.1109/ICDM.2015.92.
[10] P. Linli, “A Cross-Language Information Retrieval Method Based on Multi-Task Learning,” vol. 4, no. July, pp. 853–862, 2024.
[11] A. H. Abdulnabi, G. Wang, J. Lu, and K. Jia, “Multi-Task CNN Model for Attribute Prediction,” IEEE Trans. Multimed., vol. 17, no. 11, pp. 1949–1959, 2015, doi: 10.1109/TMM.2015.2477680.
[12] A. Pal, A. Chaturvedi, U. Garain, A. Chandra, and R. Chatterjee, “Severity grading of psoriatic plaques using deep CNN based multi-task learning,” Proc. - Int. Conf. Pattern Recognit., vol. 0, pp. 1478–1483, 2016, doi: 10.1109/ICPR.2016.7899846.
[13] W. Zhao et al., “A multi-task learning approach for image captioning,” IJCAI Int. Jt. Conf. Artif. Intell., vol. 2018-July, pp. 1205–1211, 2018, doi: 10.24963/ijcai.2018/168.
[14] Z. Xie, J. Chen, Y. Feng, K. Zhang, and Z. Zhou, “End to end multi-task learning with attention for multi-objective fault diagnosis under small sample,” J. Manuf. Syst., vol. 62, no. April, pp. 301–316, 2022, doi: 10.1016/j.jmsy.2021.12.003.
[15] H. Dong, S. Liu, Z. Fu, S. Han, and D. Zhang, “Semantic Structure Extraction for Spreadsheet Tables with a Multi-task Learning Architecture,” Work. Doc. Intell. (DI 2019) NeurIPS 2019, no. NeurIPS, pp. 2–5, 2019, [Online]. Available: https://www.microsoft.com/en-us/research/publication/semantic-structure-extraction-for-spreadsheet-tables-with-a-multi-task-learning-architecture/
[16] B. Li et al., “Improving LSTM hydrological modeling with spatiotemporal deep learning and multi-task learning: A case study of three mountainous areas on the Tibetan Plateau,” J. Hydrol., vol. 620, 2023, doi: 10.1016/j.jhydrol.2023.129401.
[17] P. Oza and V. M. Patel, “Deep CNN-based Multi-task Learning for Open-Set Recognition,” 2019, [Online]. Available: http://arxiv.org/abs/1903.03161
[18] S. Chen, J. Zhao, Q. Jin, and S. Wang, “Multimodal multi-task learning for dimensional and continuous emotion recognition,” AVEC 2017 - Proc. 7th Annu. Work. Audio/Visual Emot. Challenge, co-located with MM 2017, no. October, pp. 19–26, 2017, doi: 10.1145/3133944.3133949.
[19] M. Djandji, F. Baly, W. Antoun, and H. Hajj, “Multi-Task Learning using AraBert for Offensive Language Detection,” Proc. 4th Work. Open-Source Arab. Corpora Process. Tools, with a Shar. Task Offensive Lang. Detect., no. May, pp. 97–101, 2020, [Online]. Available: https://www.aclweb.org/anthology/2020.osact-1.16
[20] A. I. Alharbi and M. Lee, “Multi-task Learning Using a Combination of Contextualised and Static Word Embeddings for Arabic Sarcasm Detection and Sentiment Analysis,” WANLP 2021 - 6th Arab. Nat. Lang. Process. Work. Proc. Work., pp. 318–322, 2021.
[21] O. S. Mossad, M. Elnainay, and M. Torki, “Deep convolutional neural network with multi-task learning scheme for modulations recognition,” 2019 15th Int. Wirel. Commun. Mob. Comput. Conf. IWCMC 2019, no. April, pp. 1644–1649, 2019, doi: 10.1109/IWCMC.2019.8766665.
[22] T. Tohti, M. Abdurxit, and A. Hamdulla, “Medical QA Oriented Multi-Task Learning Model for Question Intent Classification and Named Entity Recognition,” Inf., vol. 13, no. 12, 2022, doi: 10.3390/info13120581.
[23] Y. Zhang and Q. Yang, “An overview of multi-task learning,” Natl. Sci. Rev., vol. 5, no. 1, pp. 30–43, 2018, doi: 10.1093/nsr/nwx105.
[24] P. Zhai, Y. Tao, H. Chen, T. Cai, and J. Li, “Multi-Task Learning for Lung Nodule Classification on Chest CT,” IEEE Access, vol. 8, pp. 180317–180327, 2020, doi: 10.1109/ACCESS.2020.3027812.
[25] Y. Xu, X. Li, H. Yuan, Y. Yang, and L. Zhang, “Multi-Task Learning With Multi-Query Transformer for Dense Prediction,” IEEE Trans. Circuits Syst. Video Technol., vol. 34, no. 2, pp. 1228–1240, 2024, doi: 10.1109/TCSVT.2023.3292995.
[26] Y. Fang, Z. Ma, Z. Zhang, X. Y. Zhang, and X. Bai, “Dynamic multi-task learning with convolutional neural network,” IJCAI Int. Jt. Conf. Artif. Intell., vol. 0, no. August, pp. 1668–1674, 2017, doi: 10.24963/ijcai.2017/231.
[27] S. Zhao, T. Liu, S. Zhao, and F. Wang, “A neural multi-task learning framework to jointly model medical named entity recognition and normalization,” 33rd AAAI Conf. Artif. Intell. AAAI 2019, 31st Innov. Appl. Artif. Intell. Conf. IAAI 2019 9th AAAI Symp. Educ. Adv. Artif. Intell. EAAI 2019, no. 2016, pp. 817–824, 2019, doi: 10.1609/aaai.v33i01.3301817.
[28] I. Misra, “Cross-stitch Networks for Multi-task Learning”.
[29] X. Ouyang et al., “A 3D-CNN and LSTM based multi-task learning architecture for action recognition,” IEEE Access, vol. 7, pp. 40757–40770, 2019, doi: 10.1109/ACCESS.2019.2906654.
[30] X. Fan et al., “Joint Optical Performance Monitoring and Modulation Format/Bit-Rate Identification by CNN-Based Multi-Task Learning,” 2018. doi: 10.1109/JPHOT.2018.2869972.
[31] A. Vaidya, F. Mai, and Y. Ning, “Empirical analysis of multi-task learning for reducing identity bias in toxic comment detection,” Proc. 14th Int. AAAI Conf. Web Soc. Media, ICWSM 2020, no. Icwsm, pp. 683–693, 2020, doi: 10.1609/icwsm.v14i1.7334.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Please find the rights and licenses in the Journal of Information Technology and Computer Engineering (JITCE).
1. License
The non-commercial use of the article will be governed by the Creative Commons Attribution license as currently displayed on Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
2. Author(s)’ Warranties
The author(s) warrants that the article is original, written by stated author(s), has not been published before, contains no unlawful statements, does not infringe the rights of others, is subject to copyright that is vested exclusively in the author and free of any third party rights, and that any necessary permissions to quote from other sources have been obtained by the author(s).
3. User Rights
JITCE adopts the spirit of open access and open science, which disseminates articles published as free as possible under the Creative Commons license. JITCE permits users to copy, distribute, display, and perform the work for non-commercial purposes only. Users will also need to attribute authors and JITCE on distributing works in the journal.
4. Rights of Authors
Authors retain the following rights:
- Copyright, and other proprietary rights relating to the article, such as patent rights,
- the right to use the substance of the article in future own works, including lectures and books,
- the right to reproduce the article for own purposes,
- the right to self-archive the article.
- the right to enter into separate, additional contractual arrangements for the non-exclusive distribution of the article's published version (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal (Journal of Information Technology and Computer Engineering).
5. Co-Authorship
If the article was jointly prepared by other authors; upon submitting the article, the author is agreed on this form and warrants that he/she has been authorized by all co-authors on their behalf, and agrees to inform his/her co-authors. JITCE will be freed on any disputes that will occur regarding this issue.
7. Royalties
By submitting the articles, the authors agreed that no fees are payable from JITCE.
8. Miscellaneous
JITCE will publish the article (or have it published) in the journal if the article’s editorial process is successfully completed and JITCE or its sublicensee has become obligated to have the article published. JITCE may adjust the article to a style of punctuation, spelling, capitalization, referencing and usage that it deems appropriate. The author acknowledges that the article may be published so that it will be publicly accessible and such access will be free of charge for the readers.