{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T18:04:11Z","timestamp":1790359451060,"version":"4.1.0"},"reference-count":44,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T00:00:00Z","timestamp":1790294400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01 HL145411"],"award-info":[{"award-number":["R01 HL145411"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01 CA276301"],"award-info":[{"award-number":["R01 CA276301"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"name":"NIH SIG","award":["S10OD026983 and S10OD030269"],"award-info":[{"award-number":["S10OD026983 and S10OD030269"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    Tuberculosis (TB), caused by\n                    <jats:italic>Mycobacterium tuberculosis<\/jats:italic>\n                    (\n                    <jats:italic>M.tb<\/jats:italic>\n                    ), remains a major global health challenge, with approximately 10.8 million new cases and 1.25 million deaths reported in 2023. Human responses to\n                    <jats:italic>M.tb<\/jats:italic>\n                    are heterogeneous with clinical outcomes including bacterial clearance, asymptomatic latent\n                    <jats:italic>M.tb<\/jats:italic>\n                    infection, and severe fatal pulmonary TB. Here, we aim to address knowledge gaps in the organization of\n                    <jats:italic>M.tb<\/jats:italic>\n                    granulomas by identifying cell-based spatial features indicating asymptomatic lung infection. To address this gap, which cannot be directly studied in humans, we used lung sections from\n                    <jats:italic>M.tb-<\/jats:italic>\n                    infected Diversity Outbred mice with acute pulmonary TB, asymptomatic\n                    <jats:italic>M.tb<\/jats:italic>\n                    infection, or chronic pulmonary TB that were stained for T cells, B cells, macrophages, and bronchiolar epithelial cells by multiplexed immunofluorescence. We first developed a new, accurate model to automatically segment lung granulomas, detect\/quantify the cell types within granulomas, and extract the location of immune cells in granulomas for each disease state. Analysis of model-derived results shows that lung granulomas from asymptomatic mice have a characteristic spatial profile consisting of higher CD4+ and CD8a+\u2009T cell densities, closer B cell proximity to bronchiolar epithelium, and increased T cell-macrophage proximity in asymptomatic\n                    <jats:italic>M.tb<\/jats:italic>\n                    infection. Next, we propose a second novel approach to utilize a large language model (LLM) to independently decode complex cellular patterns within granulomas, and distinguish key immunological signatures: balanced immune expression in asymptomatic\n                    <jats:italic>M.tb<\/jats:italic>\n                    mice, dysfunctional responses with low cellularity in acute TB, and highest immune-cell abundance in chronic TB. Overall, the results from this study show that lung granulomas in asymptomatic infection are characterized by increased T cell density, increased numbers of peribronchiolar B cells, and T cells closer to macrophages as compared to acute and chronic TB. These methods and results help establish an automated pipeline to extract and analyze data from multiplexed fluorescence images and provide a foundation to better understand how granuloma architecture varies by disease state.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0358334","type":"journal-article","created":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T17:52:05Z","timestamp":1790358725000},"page":"e0358334","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Deep learning-based spatial immunoprofiling of multiplex immunofluorescence images distinguishes tuberculosis disease states in diversity outbred mice"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7262-3034","authenticated-orcid":true,"given":"Usama","family":"Sajjad","sequence":"first","affiliation":[{"name":"The Ohio State University Wexner Medical Center","place":["Columbus, Ohio, United States of America"],"department":["Department of Pathology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicholas A.","family":"Crossland","sequence":"additional","affiliation":[{"name":"Boston University","place":["Boston, Massachusetts, United States of America"],"department":["National Emerging Infectious Diseases Laboratories (NEIDL)"]},{"name":"Boston University Chobanian & Avedisian School of Medicine","place":["Boston, Massachusetts, United States of America"],"department":["Department of Pathology and Laboratory Medicine"]},{"name":"Boston University Chobanian & Avedisian School of Medicine","place":["Boston, Massachusetts, United States of America"],"department":["Department of Virology, Immunology, & Microbiology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Metin N.","family":"Gurcan","sequence":"additional","affiliation":[{"name":"Wake Forest University School of Medicine","place":["Winston-Salem, North Carolina, United States of America"],"department":["Center for Artificial Intelligence Research"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gillian","family":"Beamer","sequence":"additional","affiliation":[{"name":"Texas Biomedical Research Institute","place":["San Antonio, Texas, United States of America"],"department":["Population Health"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Khalid Khan","family":"Niazi","sequence":"additional","affiliation":[{"name":"The Ohio State University Wexner Medical Center","place":["Columbus, Ohio, United States of America"],"department":["Department of Pathology"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,25]]},"reference":[{"key":"pone.0358334.ref001","doi-asserted-by":"crossref","DOI":"10.2471\/B09195","volume-title":"Global report on infection prevention and control 2024","author":"World Health Organization","year":"2024"},{"key":"pone.0358334.ref002","doi-asserted-by":"crossref","first-page":"103094","DOI":"10.1016\/j.ebiom.2020.103094","article-title":"Automatic discovery of clinically interpretable imaging biomarkers for Mycobacterium tuberculosis supersusceptibility using deep learning","volume":"62","author":"TE Tavolara","year":"2020","journal-title":"EBioMedicine"},{"issue":"7","key":"pone.0358334.ref003","doi-asserted-by":"crossref","DOI":"10.1128\/iai.00263-23","article-title":"B cells in perivascular and peribronchiolar granuloma-associated lymphoid tissue and B-cell signatures identify asymptomatic Mycobacterium tuberculosis lung infection in Diversity Outbred mice","volume":"92","author":"D Koyuncu","year":"2024","journal-title":"Infect Immun"},{"key":"pone.0358334.ref004","article-title":"Incipient and subclinical tuberculosis: defining early disease states in the context of host immune response","author":"JM Achkar","year":"2011","journal-title":"J Infect Dis"},{"issue":"6","key":"pone.0358334.ref005","doi-asserted-by":"crossref","first-page":"1563","DOI":"10.1183\/13993003.01245-2015","article-title":"Management of latent Mycobacterium tuberculosis infection: WHO guidelines for low tuberculosis burden countries","volume":"46","author":"H Getahun","year":"2015","journal-title":"Eur Respir J"},{"key":"pone.0358334.ref006","doi-asserted-by":"crossref","first-page":"139127","DOI":"10.1155\/2012\/139127","article-title":"The tuberculous granuloma: an unsuccessful host defence mechanism providing a safety shelter for the bacteria?","volume":"2012","author":"M Silva Miranda","year":"2012","journal-title":"Clin Dev Immunol"},{"key":"pone.0358334.ref007","doi-asserted-by":"crossref","first-page":"804838","DOI":"10.3389\/fneur.2022.804838","article-title":"Tuberculous granuloma: emerging insights from proteomics and metabolomics.","volume":"13","author":"AR Sholeye","year":"2022","journal-title":"Front Neurol."},{"issue":"9","key":"pone.0358334.ref008","doi-asserted-by":"crossref","first-page":"1100","DOI":"10.1038\/ng2087","article-title":"On the subspecific origin of the laboratory mouse","volume":"39","author":"H Yang","year":"2007","journal-title":"Nat gene."},{"issue":"7","key":"pone.0358334.ref009","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1038\/ng.847","article-title":"Subspecific origin and haplotype diversity in the laboratory mouse","volume":"43","author":"H Yang","year":"2011","journal-title":"Nat Genet"},{"issue":"2","key":"pone.0358334.ref010","doi-asserted-by":"crossref","DOI":"10.1128\/mSphere.00097-20","article-title":"The diversity outbred mouse population is an improved animal model of vaccination against tuberculosis that reflects heterogeneity of protection","volume":"5","author":"SL Kurtz","year":"2020","journal-title":"mSphere"},{"issue":"9","key":"pone.0358334.ref011","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1007\/s00335-012-9414-2","article-title":"The diversity outbred mouse population","volume":"23","author":"GA Churchill","year":"2012","journal-title":"Mamm Genome"},{"key":"pone.0358334.ref012","unstructured":"Rehman Akbar A, Sajjad U, Su Z, Li W, Xing F, Ruiz J, et al. CellEcoNet: decoding the Cellular language of pathology with deep learning for invasive lung adenocarcinoma recurrence prediction. arXiv 2025 Aug:arXiv-2508."},{"key":"pone.0358334.ref013","doi-asserted-by":"crossref","unstructured":"Sajjad U, Akbar AR, Su Z, Knight D, Frankel WL, Gurcan MN, et al. Morphology-aware prognostic model for five-year survival prediction in colorectal cancer from H&E whole slide images. arXiv 2025:arXiv:2510.14800.","DOI":"10.3390\/cancers18071150"},{"key":"pone.0358334.ref014","doi-asserted-by":"crossref","first-page":"103388","DOI":"10.1016\/j.ebiom.2021.103388","article-title":"Deep learning predicts gene expression as an intermediate data modality to identify susceptibility patterns in Mycobacterium tuberculosis infected Diversity Outbred mice","volume":"67","author":"TE Tavolara","year":"2021","journal-title":"EBioMedicine"},{"key":"pone.0358334.ref015","doi-asserted-by":"crossref","first-page":"103143","DOI":"10.1016\/j.media.2024.103143","article-title":"Cellvit: vision transformers for precise cell segmentation and classification","volume":"94","author":"F H\u00f6rst","year":"2024","journal-title":"Med Image Anal"},{"key":"pone.0358334.ref016","doi-asserted-by":"crossref","unstructured":"Sajjad U, Khan Niazi K, Beamer G, Gurcan MN. Clinically interpretable imaging biomarker discovery in BCG-vaccinated mycobacterium tuberculosis-infected diversity outbred mice using deep learning. In: Medical Imaging 2025: Digital and Computational Pathology; 2025. pp. 23. https:\/\/doi.org\/10.1117\/12.3047288","DOI":"10.1117\/12.3047288"},{"key":"pone.0358334.ref017","unstructured":"Su Z, Akbar AR, Sajjad U, Parwani AV, Niazi MK. Streamline pathology foundation model by cross-magnification distillation. arXiv. 2025:arXiv:2509.23097."},{"key":"pone.0358334.ref018","doi-asserted-by":"crossref","unstructured":"Akbar AR, Sajjad U, Su Z, Li W, Xing F, Ruiz J, et al. Learning the language of histopathology images reveals prognostic subgroups in invasive lung adenocarcinoma patients. Res Sq. 2026:rs-3.","DOI":"10.21203\/rs.3.rs-8089525\/v1"},{"key":"pone.0358334.ref019","doi-asserted-by":"crossref","unstructured":"Tavolara TE, Niazi MKK, Beamer G, Gurcan MN. Identifying lung imaging biomarkers of BCG vaccination after infection with Mycobacterium tuberculosis. In: Medical Imaging 2021: Digital Pathology; 2021. p. 7. https:\/\/doi.org\/10.1117\/12.2581945","DOI":"10.1117\/12.2581945"},{"key":"pone.0358334.ref020","doi-asserted-by":"crossref","first-page":"103094","DOI":"10.1016\/j.ebiom.2020.103094","article-title":"Automatic discovery of clinically interpretable imaging biomarkers for Mycobacterium tuberculosis supersusceptibility using deep learning","volume":"62","author":"TE Tavolara","year":"2020","journal-title":"EBioMedicine"},{"issue":"528","key":"pone.0358334.ref021","doi-asserted-by":"crossref","DOI":"10.1126\/scitranslmed.aay0233","article-title":"Immune correlates of tuberculosis disease and risk translate across species","volume":"12","author":"M Ahmed","year":"2020","journal-title":"Sci Transl Med"},{"issue":"9","key":"pone.0358334.ref022","first-page":"1141","article-title":"Lung necrosis and neutrophils reflect common pathways of susceptibility to Mycobacterium tuberculosis in genetically diverse, immune-competent mice","volume":"8","author":"MKK Niazi","year":"2015","journal-title":"Dis Model Mech"},{"issue":"12","key":"pone.0358334.ref023","doi-asserted-by":"crossref","first-page":"661","DOI":"10.3390\/microorganisms7120661","article-title":"Automatic detection of granuloma necrosis in pulmonary tuberculosis using a two-phase algorithm: 2D-TB","volume":"7","author":"P Kus","year":"2019","journal-title":"Microorganisms"},{"issue":"6","key":"pone.0358334.ref024","doi-asserted-by":"crossref","DOI":"10.1371\/journal.ppat.1011915","article-title":"Systems genetics uncover new loci containing functional gene candidates in Mycobacterium tuberculosis-infected Diversity Outbred mice","volume":"20","author":"DM Gatti","year":"2024","journal-title":"PLoS Pathog"},{"issue":"8","key":"pone.0358334.ref025","doi-asserted-by":"crossref","DOI":"10.1371\/journal.ppat.1009773","article-title":"CXCL1: a new diagnostic biomarker for human tuberculosis discovered using Diversity Outbred mice","volume":"17","author":"D Koyuncu","year":"2021","journal-title":"PLoS Pathog"},{"key":"pone.0358334.ref026","first-page":"3","article-title":"Unet++: A nested u-net architecture for medical image segmentation.","volume-title":"2018, proceedings","author":"Z Zhou","year":"2018"},{"key":"pone.0358334.ref027","doi-asserted-by":"crossref","unstructured":"Stringer C, Wang T, Michaelos M, Pachitariu M. Cellpose: a generalist algorithm for cellular segmentation. Nat Method. 2021;18(1):100\u20136.","DOI":"10.1038\/s41592-020-01018-x"},{"key":"pone.0358334.ref028","doi-asserted-by":"crossref","unstructured":"Dundar MM, Badve S, Raykar VC, Jain RK, Sertel O, Gurca MN. A multiple instance learning approach toward optimal classification of pathology slides. In: 2010 20th International Conference on Pattern Recognition; 2010. pp. 2732\u20135. https:\/\/doi.org\/10.1109\/icpr.2010.669","DOI":"10.1109\/ICPR.2010.669"},{"issue":"7","key":"pone.0358334.ref029","doi-asserted-by":"crossref","first-page":"1977","DOI":"10.1109\/TBME.2011.2110648","article-title":"Computerized classification of intraductal breast lesions using histopathological images","volume":"58","author":"MM Dundar","year":"2011","journal-title":"IEEE Trans Biomed Eng"},{"issue":"13","key":"pone.0358334.ref030","doi-asserted-by":"crossref","first-page":"3428","DOI":"10.3390\/cancers15133428","article-title":"NRK-ABMIL: subtle metastatic deposits detection for predicting lymph node metastasis in breast cancer whole-slide images","volume":"15","author":"U Sajjad","year":"2023","journal-title":"Cancers (Basel)"},{"key":"pone.0358334.ref031","unstructured":"Ilse M, Tomczak J, Welling M. Attention-based deep multiple instance learning. Int Conf Mach Learn. 2018:2127\u201336."},{"key":"pone.0358334.ref032","unstructured":"Shen M, Li Y, Chen L, Yang Q. From mind to machine: the rise of manus ai as a fully autonomous digital agent. arXiv 2025;arXiv:2505.02024."},{"key":"pone.0358334.ref033","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1186\/1471-2342-11-11","article-title":"Quantitative metric profiles capture three-dimensional temporospatial architecture to discriminate cellular functional states","volume":"11","author":"L McKeen-Polizzotti","year":"2011","journal-title":"BMC Med Imaging"},{"issue":"8033","key":"pone.0358334.ref034","doi-asserted-by":"crossref","first-page":"466","DOI":"10.1038\/s41586-024-07618-3","article-title":"A multimodal generative AI copilot for human pathology","volume":"634","author":"MY Lu","year":"2024","journal-title":"Nature"},{"issue":"8","key":"pone.0358334.ref035","doi-asserted-by":"crossref","first-page":"101284","DOI":"10.1016\/j.patter.2025.101284","article-title":"HistoChat: instruction-tuning multimodal vision language assistant for colorectal histopathology on limited data","volume":"6","author":"U Afzaal","year":"2025","journal-title":"Patterns (N Y)"},{"key":"pone.0358334.ref036","doi-asserted-by":"crossref","unstructured":"Afzaal U, Su Z, Sajjad U, Stack T, Rezapour M, Lu H, et al. Instruction tuning for colorectal histopathology: a multimodal vision-language assistant on human-evaluated data. In: Medical Imaging 2025: Digital and Computational Pathology; 2025. Vol. 13413, pp. 127\u201334.","DOI":"10.1117\/12.3047352"},{"issue":"3","key":"pone.0358334.ref037","doi-asserted-by":"crossref","first-page":"104998","DOI":"10.1016\/j.labinv.2025.104998","article-title":"711 an AI virtual agent for prognostic interpretation: translating black-box predictions into clinical insights","volume":"106","author":"U Sajjad","year":"2026","journal-title":"Lab Invest"},{"key":"pone.0358334.ref038","doi-asserted-by":"crossref","unstructured":"Amirizaniani M, Martin E, Sivachenko M, Mashhadi A, Shah C. Can LLMs reason like humans? Assessing theory of mind reasoning in LLMs for open-ended questions. In: Proceedings of the 33rd ACM International Conference on Information and Knowledge Management, 2024. pp. 34\u201344. https:\/\/doi.org\/10.1145\/3627673.3679832","DOI":"10.1145\/3627673.3679832"},{"key":"pone.0358334.ref039","doi-asserted-by":"crossref","unstructured":"Li D, Cao S, Griggs T, Liu S, Mo X, Tang E, et al. LLMs can easily learn to reason from demonstrations structure, not content, is what matters! arXiv 2025:arXiv:2502.07374.","DOI":"10.18653\/v1\/2025.findings-emnlp.866"},{"issue":"2","key":"pone.0358334.ref040","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1002\/path.1906","article-title":"New insights into the function of granulomas in human tuberculosis","volume":"208","author":"T Ulrichs","year":"2006","journal-title":"J Pathol"},{"issue":"5","key":"pone.0358334.ref041","doi-asserted-by":"crossref","first-page":"531","DOI":"10.1038\/nm.4073","article-title":"Inflammatory signaling in human tuberculosis granulomas is spatially organized","volume":"22","author":"MJ Marakalala","year":"2016","journal-title":"Nat Med"},{"issue":"6","key":"pone.0358334.ref042","doi-asserted-by":"crossref","DOI":"10.1084\/jem.20221392","article-title":"Spatial mapping reveals granuloma diversity and histopathological superstructure in human tuberculosis","volume":"220","author":"AJ Sawyer","year":"2023","journal-title":"J Exp Med"},{"issue":"3","key":"pone.0358334.ref043","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1016\/j.jrid.2018.08.001","article-title":"Spatial and temporal evolution of lung granulomas in a cynomolgus macaque model of Mycobacterium tuberculosis infection","volume":"5","author":"RW Gregg","year":"2018","journal-title":"Radiol Infect Dis"},{"issue":"10","key":"pone.0358334.ref044","doi-asserted-by":"crossref","first-page":"4631","DOI":"10.1128\/IAI.00592-09","article-title":"Quantitative comparison of active and latent tuberculosis in the cynomolgus macaque model","volume":"77","author":"PL Lin","year":"2009","journal-title":"Infect Immun"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358334","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T17:52:17Z","timestamp":1790358737000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358334"}},"subtitle":[],"editor":[{"given":"Selvakumar","family":"Subbian","sequence":"first","affiliation":[{"name":"Rutgers Biomedical and Health Sciences","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,25]]},"references-count":44,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,25]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0358334","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,25]]}}}