{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,13]],"date-time":"2026-04-13T22:07:19Z","timestamp":1776118039863,"version":"3.50.1"},"reference-count":40,"publisher":"MDPI AG","issue":"14","license":[{"start":{"date-parts":[[2024,7,10]],"date-time":"2024-07-10T00:00:00Z","timestamp":1720569600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JCM"],"abstract":"<jats:p>Background: Pneumothorax detection is often challenging, particularly when radiographic features are subtle. This study introduces a deep learning model that integrates curriculum learning and ChatGPT to enhance the detection of pneumothorax in chest X-rays. Methods: The model training began with large, easily detectable pneumothoraces, gradually incorporating smaller, more complex cases to prevent performance plateauing. The training dataset comprised 6445 anonymized radiographs, validated across multiple sites, and further tested for generalizability in diverse clinical subgroups. Performance metrics were analyzed using descriptive statistics. Results: The model achieved a sensitivity of 0.97 and a specificity of 0.97, with an area under the curve (AUC) of 0.98, demonstrating a performance comparable to that of many FDA-approved devices. Conclusions: This study suggests that a structured approach to training deep learning models, through curriculum learning and enhanced data extraction via natural language processing, can facilitate and improve the training of AI models for pneumothorax detection.<\/jats:p>","DOI":"10.3390\/jcm13144042","type":"journal-article","created":{"date-parts":[[2024,7,10]],"date-time":"2024-07-10T15:22:05Z","timestamp":1720624925000},"page":"4042","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Utilizing ChatGPT for Curriculum Learning in Developing a Clinical Grade Pneumothorax Detection Model: A Multisite Validation Study"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1704-2755","authenticated-orcid":false,"given":"Joseph","family":"Chang","sequence":"first","affiliation":[{"name":"Department of Biomedical Engineering, College of Medicine and College of Engineering, National Taiwan University, No. 1, Sec. 1, Jen-Ai Road, Taipei 100, Taiwan"},{"name":"EverFortune.AI Co., Ltd., Taichung 403, Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kuan-Jung","family":"Lee","sequence":"additional","affiliation":[{"name":"EverFortune.AI Co., Ltd., Taichung 403, Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ti-Hao","family":"Wang","sequence":"additional","affiliation":[{"name":"EverFortune.AI Co., Ltd., Taichung 403, Taiwan"},{"name":"Department of Medicine, China Medical University, Taichung 404, Taiwan"},{"name":"Department of Radiation Oncology, China Medical University Hospital, Taichung 404, Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0023-5817","authenticated-orcid":false,"given":"Chung-Ming","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, College of Medicine and College of Engineering, National Taiwan University, No. 1, Sec. 1, Jen-Ai Road, Taipei 100, Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2024,7,10]]},"reference":[{"key":"ref_1","first-page":"1379","article-title":"Incidence of spontaneous pneumothorax in Olmsted County, Minnesota: 1950 to 1974","volume":"120","author":"Melton","year":"1979","journal-title":"Am. Rev. Respir. Dis."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"805","DOI":"10.1136\/thx.52.9.805","article-title":"Recurrence of primary spontaneous pneumothorax","volume":"52","author":"Sadikot","year":"1997","journal-title":"Thorax"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"590","DOI":"10.1378\/chest.119.2.590","article-title":"Management of spontaneous pneumothorax: An American College of Chest Physicians Delphi consensus statement","volume":"119","author":"Baumann","year":"2001","journal-title":"Chest"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1111\/j.1758-5910.2012.00161.x","article-title":"Thoracoscopic surgery for refractory cases of secondary spontaneous pneumothorax","volume":"6","author":"Odaka","year":"2013","journal-title":"Asian J. Endosc. Surg."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"050","DOI":"10.1055\/s-0034-1399782","article-title":"When is the optimal timing of the surgical treatment for secondary spontaneous pneumothorax?","volume":"65","author":"Jeon","year":"2015","journal-title":"Thorac. Cardiovasc. Surg."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1335","DOI":"10.1097\/TA.0b013e3181f6f525","article-title":"Occult Pneumothoraces Truly Occult or Simply Missed: Redux","volume":"69","author":"Brar","year":"2010","journal-title":"J. Trauma Acute Care Surg."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1231","DOI":"10.1097\/01.CCM.0000164542.86954.B4","article-title":"Ultrasound Diagnosis of Occult Pneumothorax","volume":"33","author":"Lichtenstein","year":"2005","journal-title":"Crit. Care Med."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"378","DOI":"10.1016\/j.annemergmed.2010.08.012","article-title":"A Method to Detect Occult Pneumothorax with Chest Radiography","volume":"57","author":"Matsumoto","year":"2011","journal-title":"Ann. Emerg. Med."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"720","DOI":"10.1378\/chest.115.3.720","article-title":"Miss Rate of Lung Cancer on the Chest Radiograph in Clinical Practice","volume":"115","author":"Quekel","year":"1999","journal-title":"Chest"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13017-016-0061-x","article-title":"Diagnostic Accuracy of Oblique Chest Radiograph for Occult Pneumothorax: Comparison with Ultrasonography","volume":"11","author":"Matsumoto","year":"2016","journal-title":"World J. Emerg. Surg."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"e200190","DOI":"10.1148\/ryai.2021200190","article-title":"Deep Learning Systems for Pneumothorax Detection on Chest Radiographs: A Multicenter External Validation Study","volume":"3","author":"Thian","year":"2021","journal-title":"Radiol. Artif. Intell."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1350","DOI":"10.1016\/j.acra.2021.09.013","article-title":"Detection of Pneumothorax with Deep Learning Models: Learning From Radiologist Labels vs Natural Language Processing Model Generated Labels","volume":"29","author":"Hallinan","year":"2021","journal-title":"Acad. Radiol."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Rajpurkar, P., Irvin, J., Ball, R.L., Zhu, K., Yang, B., Mehta, H., Duan, T., Ding, D., Bagul, A., and Langlotz, C. (2018). Deep Learning for Chest Radiograph Diagnosis: A Retrospective Comparison of the CheXNeXt Algorithm to Practicing Radiologists. PLoS Med., 15.","DOI":"10.1371\/journal.pmed.1002686"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1186\/s40537-019-0197-0","article-title":"A Survey on Image Data Augmentation for Deep Learning","volume":"6","author":"Shorten","year":"2019","journal-title":"J. Big Data"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"657","DOI":"10.14245\/ns.1938396.198","article-title":"Deep Learning in Medical Imaging","volume":"16","author":"Kim","year":"2019","journal-title":"Neurospine"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"570","DOI":"10.3348\/kjr.2017.18.4.570","article-title":"Deep Learning in Medical Imaging: General Overview","volume":"18","author":"Lee","year":"2017","journal-title":"Korean J. Radiol."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Tsang, R. (2023). Practical Applications of ChatGPT in Undergraduate Medical Education. J. Med. Educ. Curric. Dev., 10.","DOI":"10.1177\/23821205231178449"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e48568","DOI":"10.2196\/48568","article-title":"Utility of ChatGPT in Clinical Practice","volume":"25","author":"Liu","year":"2023","journal-title":"J. Med. Internet Res."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Dave, T., Athaluri, S.A., and Singh, S. (2023). ChatGPT in Medicine: An Overview of Its Applications, Advantages, Limitations, Future Prospects, and Ethical Considerations. Front. Artif. Intell., 6.","DOI":"10.3389\/frai.2023.1169595"},{"key":"ref_20","first-page":"e37589","article-title":"Evaluation of ChatGPT\u2019s Capabilities in Medical Report Generation","volume":"15","author":"Zhou","year":"2023","journal-title":"Cureus"},{"key":"ref_21","first-page":"e38317","article-title":"ChatGPT in Dentistry: A Comprehensive Review","volume":"15","author":"Alhaidry","year":"2023","journal-title":"Cureus"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"A44","DOI":"10.1136\/thoraxjnl-2013-204457.89","article-title":"Multi-centre prospective comparison of the BTS and ACCP guidelines to determine size in primary spontaneous pneumothorax","volume":"68","author":"Nikolic","year":"2013","journal-title":"Thorax"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1830","DOI":"10.1016\/j.rmed.2008.07.026","article-title":"Comparison of Size Classification of Primary Spontaneous Pneumothorax by Three International Guidelines: A Case for International Consensus?","volume":"102","author":"Kelly","year":"2008","journal-title":"Respir. Med."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1230","DOI":"10.1097\/CCM.0b013e31816a0380","article-title":"Real-time Detection of Pneumothorax Using Electrical Impedance Tomography","volume":"36","author":"Costa","year":"2008","journal-title":"Crit. Care Med."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"245","DOI":"10.1007\/s42979-020-00251-7","article-title":"Deep Curriculum Learning Optimization","volume":"1","author":"Ghebrechristos","year":"2020","journal-title":"SN Comput. Sci."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Wang, Y., Yue, Y., Lu, R., Liu, T.-D., Zhong, Z., Song, S., and Huang, G. (2022). EfficientTrain: Exploring Generalized Curriculum Learning for Training Visual Backbones. arXiv.","DOI":"10.1109\/ICCV51070.2023.00538"},{"key":"ref_27","unstructured":"Simionescu, C. (2022). Deep Learning Training Procedure Augmentations. arXiv."},{"key":"ref_28","first-page":"395","article-title":"Ensembled EfficientNetB3 Architecture for Multi-class Classification of Tumours in MRI Images","volume":"17","author":"Dudeja","year":"2023","journal-title":"Intell. Decis. Technol."},{"key":"ref_29","unstructured":"Sheela, R.K., Nagaraju, Y., and Sahu, D.A. (2022, January 27\u201329). Histopathological Image Classification of Breast Cancer Using EfficientNet. Proceedings of the 2022 3rd International Conference for Emerging Technology (INCET), Belgaum, India."},{"key":"ref_30","unstructured":"Tan, M., and Le, Q. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1485","DOI":"10.18280\/ts.380524","article-title":"EfficientNet-B0 Based Monocular Dense-Depth Map Estimation","volume":"38","author":"Tadepalli","year":"2021","journal-title":"Trait. Du Signal"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"16874","DOI":"10.1038\/s41598-022-21426-7","article-title":"Systematic Analysis of the Test Design and Performance of AI\/ML-Based Medical Devices Approved for Triage\/Detection\/Diagnosis in the USA and Japan","volume":"12","author":"Yuba","year":"2022","journal-title":"Sci. Rep."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"100612","DOI":"10.1016\/j.websem.2020.100612","article-title":"Less Is More: Data-Efficient Complex Question Answering Over Knowledge Bases","volume":"65","author":"Hua","year":"2020","journal-title":"J. Web Semant."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Guo, S., Huang, W., Zhang, H., Zhuang, C., Dong, D., Scott, M.R., and Huang, D. (2018, January 8\u201314). CurriculumNet: Weakly Supervised Learning from Large-Scale Web Images. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01249-6_9"},{"key":"ref_35","unstructured":"Abbe, E., Cornacchia, E., and Lotfi, A. (2024). Provable Advantage of Curriculum Learning on Parity Targets with Mixed Inputs. Adv. Neural Inf. Process. Syst., 36."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Wang, Y., Gan, W., Yang, J., Wu, W., and Yan, J. (November, January 27). Dynamic Curriculum Learning for Imbalanced Data Classification. Proceedings of the IEEE\/CVF International Conference on Computer Vision 2019, Seoul, Republic of Korea. Available online: https:\/\/arxiv.org\/abs\/1901.06783.","DOI":"10.1109\/ICCV.2019.00512"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Gong, Y., Liu, C., Yuan, J., Yang, F., Cai, X., Wan, G., Chen, J., Niu, R., and Wang, H. (2021, January 1\u20135). Density-Based Dynamic Curriculum Learning for Intent Detection. Proceedings of the 30th ACM International Conference on Information & Knowledge Management, Gold Coast, Australia.","DOI":"10.1145\/3459637.3482082"},{"key":"ref_38","unstructured":"Lin, H., Zhang, H., Ma, Y., He, T., Zhang, Z., Zha, S., and Li, M. (2019). Dynamic Mini-Batch SGD for Elastic Distributed Training: Learning in the Limbo of Resources. arXiv."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Doshi, R., Amin, K., Khosla, P., Bajaj, S., Chheang, S., and Forman, H.P. (2023). Utilizing Large Language Models to Simplify Radiology Reports: A Comparative Analysis of ChatGPT-3.5, ChatGPT-4.0, Google Bard, and Microsoft Bing. medRxiv.","DOI":"10.1101\/2023.06.04.23290786"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1657","DOI":"10.1093\/jamia\/ocad133","article-title":"Inferring Cancer Disease Response from Radiology Reports Using Large Language Models with Data Augmentation and Prompting","volume":"30","author":"Tan","year":"2023","journal-title":"J. Am. Med. Inform. Assoc."}],"container-title":["Journal of Clinical Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2077-0383\/13\/14\/4042\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T15:13:00Z","timestamp":1760109180000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2077-0383\/13\/14\/4042"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7,10]]},"references-count":40,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2024,7]]}},"alternative-id":["jcm13144042"],"URL":"https:\/\/doi.org\/10.3390\/jcm13144042","relation":{},"ISSN":["2077-0383"],"issn-type":[{"value":"2077-0383","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,7,10]]}}}