{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,12]],"date-time":"2026-07-12T01:42:39Z","timestamp":1783820559625,"version":"3.55.0"},"reference-count":47,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T00:00:00Z","timestamp":1739750400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Public Health"],"abstract":"<jats:sec><jats:title>Objective<\/jats:title><jats:p>To develop and evaluate the effectiveness of multimodal machine learning approach for the differentiation of NTM from MTB.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>The clinical data and CT images of 175 patients were retrospectively obtained. We established clinical data-based model, radiomics-based model, and multimodal (clinical plus radiomics) model gradually using 5 machine learning algorithms (Logistic, XGBoost, AdaBoost, RandomForest, and LightGBM). Optimal algorithm in each model was selected after evaluating the differentiation performance both in training and validation sets. The model performance was further verified using external new MTB and NTM patient data. Performance was also compared with the existing approaches and model.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The clinical data-based model contained age, gender, and IL-6, and the RandomForest algorithm achieved the optimal learning model. Two key radiomics features of CT images were identified and then used to establish the radiomics model, finding that model from Logistic algorithm was the optimal. The multimodal model contained age, IL-6, and the 2 radiomics features, and the optimal model was from LightGBM algorithm. The optimal multimodal model had the highest AUC value, accuracy, sensitivity, and negative predictive value compared with the optimal clinical or radiomics models, and its\u2019 favorable performance was also verified in the external test dataset (accuracy\u202f=\u202f0.745, sensitivity\u202f=\u202f0.900). Additionally, the performance of multimodal model was better than that of the radiologist, NGS detection, and existing machine learning model, with an increased accuracy of 26, 4, and 6%, respectively.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>This is the first study to establish multimodal model to distinguish NTM from MTB and it performs well in differentiating them, which has the potential to aid clinical decision-making for experienced radiologists.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fpubh.2025.1470072","type":"journal-article","created":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T07:02:42Z","timestamp":1739775762000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Multimodal machine learning-based model for differentiating nontuberculous mycobacteria from mycobacterium tuberculosis"],"prefix":"10.3389","volume":"13","author":[{"given":"Hong-ling","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ri-zeng","family":"Zhi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hua-sheng","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mei","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Si-jie","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,2,17]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"1642","DOI":"10.1016\/S0140-6736(19)30308-3","article-title":"Tuberculosis","volume":"393","author":"Furin","year":"2019","journal-title":"Lancet"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"e20","DOI":"10.1016\/S2666-5247(22)00359-7","article-title":"WHO's global tuberculosis report 2022","volume":"4","author":"Bagcchi","year":"2023","journal-title":"Lancet Microbe"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"120","DOI":"10.1016\/j.ijid.2022.10.013","article-title":"Global trends of pulmonary infections with nontuberculous mycobacteria: a systematic review","volume":"125","author":"Dahl","year":"2022","journal-title":"Int J Infect Dis"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"303","DOI":"10.3389\/fimmu.2020.00303","article-title":"The rise of non-tuberculosis mycobacterial lung disease","volume":"11","author":"Ratnatunga","year":"2020","journal-title":"Front Immunol"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1186\/s12929-020-00667-6","article-title":"Of tuberculosis and non-tuberculous mycobacterial infections - a comparative analysis of epidemiology, diagnosis and treatment","volume":"27","author":"Gopalaswamy","year":"2020","journal-title":"J Biomed Sci"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1186\/s40249-021-00844-1","article-title":"Nontuberculous mycobacteria in China: incidence and antimicrobial resistance spectrum from a nationwide survey","volume":"10","author":"Liu","year":"2021","journal-title":"Infect Dis Poverty"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"4065","DOI":"10.1007\/s11033-020-05413-7","article-title":"Advances in diagnosis of tuberculosis: an update into molecular diagnosis of Mycobacterium tuberculosis","volume":"47","author":"Acharya","year":"2020","journal-title":"Mol Biol Rep"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"e0192351","DOI":"10.1371\/journal.pone.0192351","article-title":"Molecular epidemiology of tuberculosis in Tasmania and genomic characterisation of its first known multi-drug resistant case","volume":"13","author":"Gautam","year":"2018","journal-title":"PLoS One"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.7754\/Clin.Lab.2023.230823","article-title":"A case of tuberculosis misdiagnosed as sarcoidosis and then confirmed by NGS testing","volume":"70","author":"Yang","year":"2024","journal-title":"Clin Lab"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"2429","DOI":"10.1183\/13993003.congress-2015.pa2674","article-title":"Chest imaging comparison between non-tuberculous and tuberculosis mycobacteria in sputum acid fast bacilli smear-positive patients","volume":"19","author":"Chu","year":"2015","journal-title":"Eur Rev Med Pharmacol Sci"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"4148","DOI":"10.1002\/mp.12331","article-title":"Computerized detection of lung nodules through radiomics","volume":"44","author":"Ma","year":"2017","journal-title":"Med Phys"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"467","DOI":"10.1016\/j.jtho.2016.11.2226","article-title":"Radiomic-based pathological response prediction from primary tumors and lymph nodes in NSCLC","volume":"12","author":"Coroller","year":"2017","journal-title":"J Thorac Oncol"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"6287545","DOI":"10.1155\/2020\/6287545","article-title":"Machine learning-based differentiation of nontuberculous mycobacteria lung disease and pulmonary tuberculosis using CT images","volume":"2020","author":"Xing","year":"2020","journal-title":"Biomed Res Int"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1016\/j.ijid.2022.09.031","article-title":"T-SPOT with CT image analysis based on deep learning for early differential diagnosis of nontuberculous mycobacteria pulmonary disease and pulmonary tuberculosis","volume":"125","author":"Ying","year":"2022","journal-title":"Int J Infect Dis"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1016\/j.cell.2020.03.022","article-title":"How machine learning will transform biomedicine","volume":"181","author":"Goecks","year":"2020","journal-title":"Cell"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"696","DOI":"10.4103\/0971-5916.90999","article-title":"Mycobacterium indicus pranii as stand-alone or adjunct immunotherapeutic in treatment of experimental animal tuberculosis","volume":"134","author":"Faujdar","year":"2011","journal-title":"Indian J Med Res"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"2191","DOI":"10.2174\/1570159X22999240531160344","article-title":"Brain disorder detection and diagnosis using machine learning and deep learning - a bibliometric analysis","volume":"22","author":"Chaki","year":"2024","journal-title":"Curr Neuropharmacol"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"114","DOI":"10.1186\/s13045-023-01514-5","article-title":"Novel research and future prospects of artificial intelligence in cancer diagnosis and treatment","volume":"16","author":"Zhang","year":"2023","journal-title":"J Hematol Oncol"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"2370399","DOI":"10.1080\/22221751.2024.2370399","article-title":"Plasma immune profiling combined with machine learning contributes to diagnosis and prognosis of active pulmonary tuberculosis","volume":"13","author":"Yao","year":"2024","journal-title":"Emerg Microbes Infect"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1186\/s40779-023-00490-8","article-title":"From immunology to artificial intelligence: revolutionizing latent tuberculosis infection diagnosis with machine learning","volume":"10","author":"Li","year":"2023","journal-title":"Mil Med Res"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"111180","DOI":"10.1016\/j.ejrad.2023.111180","article-title":"Deep learning and radiomics of longitudinal CT scans for early prediction of tuberculosis treatment outcomes","volume":"169","author":"Nijiati","year":"2023","journal-title":"Eur J Radiol"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"15312","DOI":"10.1038\/s41598-024-66113-x","article-title":"Discovery of urinary biosignatures for tuberculosis and nontuberculous mycobacteria classification using metabolomics and machine learning","volume":"14","author":"Anh","year":"2024","journal-title":"Sci Rep"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"2401613","DOI":"10.1080\/07853890.2024.2401613","article-title":"A retrospective study differentiating nontuberculous mycobacterial pulmonary disease from pulmonary tuberculosis on computed tomography using radiomics and machine learning algorithms","volume":"56","author":"Zhou","year":"2024","journal-title":"Ann Med"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"1710","DOI":"10.12688\/f1000research.20096.1","article-title":"Recent advances in nontuberculous mycobacterial lung infections","volume":"8","author":"Horne","year":"2019","journal-title":"F1000Res"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"561","DOI":"10.1080\/14737159.2023.2223980","article-title":"Diagnosis of extrapulmonary tuberculosis by gene Xpert MTB\/RIF ultra assay","volume":"23","author":"Dahiya","year":"2023","journal-title":"Expert Rev Mol Diagn"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"45","DOI":"10.14715\/cmb\/2023.69.8.7","article-title":"Application value of next-generation sequencing of bronchial alveolar lavage fluid in emergency patients with infection","volume":"69","author":"Wu","year":"2023","journal-title":"Cell Mol biol (Noisy-le-grand)"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1213\/ANE.0000000000005247","article-title":"Logistic regression in medical research","volume":"132","author":"Schober","year":"2021","journal-title":"Anesth Analg"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1093\/bib\/bbs034","article-title":"Data mining in the life sciences with random Forest: a walk in the park or lost in the jungle?","volume":"14","author":"Touw","year":"2013","journal-title":"Brief Bioinform"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"821996","DOI":"10.3389\/fgene.2021.821996","article-title":"Identify DNA-binding proteins through the extreme gradient boosting algorithm","volume":"12","author":"Zhao","year":"2021","journal-title":"Front Genet"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics11091714","article-title":"Diagnosis of diabetes mellitus using gradient boosting machine (light GBM)","volume":"11","author":"Rufo","year":"2021","journal-title":"Diagnostics (Basel)"},{"key":"ref31","article-title":"Light GBM: a highly efficient gradient boosting decision tree","author":"Ke","year":"2017","journal-title":"Neural Inform Proces Syst"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"e201960241","DOI":"10.1002\/jbio.201960241","article-title":"Adaptive boosting (Ada boost)-based multiwavelength spatial frequency domain imaging and characterization for ex vivo human colorectal tissue assessment","volume":"13","author":"Li","year":"2020","journal-title":"J Biophotonics"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"4293","DOI":"10.1007\/s00259-021-05432-x","article-title":"Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework","volume":"48","author":"Wang","year":"2021","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"1540","DOI":"10.3201\/eid2908.230378","article-title":"Spatial epidemiologic analysis and risk factors for nontuberculous mycobacteria infections, Missouri, USA, 2008-2019","volume":"29","author":"Mejia-Chew","year":"2023","journal-title":"Emerg Infect Dis"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"881","DOI":"10.1164\/rccm.201111-2016OC","article-title":"Prevalence of nontuberculous mycobacterial lung disease in U.S. Medicare beneficiaries","volume":"185","author":"Adjemian","year":"2012","journal-title":"Am J Respir Crit Care Med"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"356","DOI":"10.1016\/j.ejim.2014.03.008","article-title":"Management of nontuberculous mycobacterial infection in the elderly","volume":"25","author":"Mirsaeidi","year":"2014","journal-title":"Eur J Intern Med"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"977","DOI":"10.1164\/rccm.201003-0503OC","article-title":"Pulmonary nontuberculous mycobacterial disease prevalence and clinical features: an emerging public health disease","volume":"182","author":"Winthrop","year":"2010","journal-title":"Am J Respir Crit Care Med"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1016\/j.pulmoe.2022.01.011","article-title":"Nontuberculous mycobacteria in Portugal: trends from the last decade","volume":"30","author":"Santos","year":"2022","journal-title":"Pulmonology"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"e12926","DOI":"10.1111\/acel.12926","article-title":"Nontuberculous mycobacterium M. avium infection predisposes aged mice to cardiac abnormalities and inflammation","volume":"18","author":"Headley","year":"2019","journal-title":"Aging Cell"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"275","DOI":"10.4046\/trd.2012.72.3.275","article-title":"Impaired expression of MAPK is associated with the downregulation of TNF-alpha, IL-6, and IL-10 in Mycobacterium abscessus lung disease","volume":"72","author":"Sim","year":"2012","journal-title":"Tuberc Respir Dis (Seoul)"},{"key":"ref41","doi-asserted-by":"publisher","first-page":"105175","DOI":"10.1016\/j.micinf.2023.105175","article-title":"Long non-coding RNA-XLOC_002383 enhances the inhibitory effects of THP-1 macrophages on Mycobacterium avium and functions as a competing endogenous RNA by sponging mi R-146a-5p to target TRAF6","volume":"25","author":"Hu","year":"2023","journal-title":"Microbes Infect"},{"key":"ref42","doi-asserted-by":"publisher","first-page":"328","DOI":"10.1148\/radiol.2020191145","article-title":"The image biomarker standardization initiative: standardized quantitative Radiomics for high-throughput image-based phenotyping","volume":"295","author":"Zwanenburg","year":"2020","journal-title":"Radiology"},{"key":"ref43","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1007\/s12149-016-1062-2","article-title":"Differentiating the grades of thymic epithelial tumor malignancy using textural features of intratumoral heterogeneity via (18) F-FDG PET\/CT","volume":"30","author":"Lee","year":"2016","journal-title":"Ann Nucl Med"},{"key":"ref44","doi-asserted-by":"publisher","first-page":"2273","DOI":"10.2169\/internalmedicine.53.2558","article-title":"Nontuberculous mycobacterium diseases and chronic thromboembolic pulmonary hypertension","volume":"53","author":"Kuroda","year":"2014","journal-title":"Intern Med"},{"key":"ref45","doi-asserted-by":"publisher","first-page":"235","DOI":"10.4103\/ijmy.ijmy_208_22","article-title":"Hematological profiles of patients with tuberculosis and nontuberculous mycobacteria infections in Bamako, Mali","volume":"12","author":"Sanogo","year":"2023","journal-title":"Int J Mycobacteriol"},{"key":"ref46","doi-asserted-by":"publisher","first-page":"116","DOI":"10.3389\/fmicb.2020.00116","article-title":"Differential responses by human macrophages to infection with Mycobacterium tuberculosis and non-tuberculous mycobacteria","volume":"11","author":"Feng","year":"2020","journal-title":"Front Microbiol"},{"key":"ref47","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1186\/s13244-023-01395-9","article-title":"A deep learning model using chest X-ray for identifying TB and NTM-LD patients: a cross-sectional study","volume":"14","author":"Liu","year":"2023","journal-title":"Insights Imaging"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2025.1470072\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T07:02:45Z","timestamp":1739775765000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2025.1470072\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,17]]},"references-count":47,"alternative-id":["10.3389\/fpubh.2025.1470072"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2025.1470072","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,17]]},"article-number":"1470072"}}