{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,12]],"date-time":"2026-02-12T16:39:40Z","timestamp":1770914380003,"version":"3.50.1"},"reference-count":31,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T00:00:00Z","timestamp":1770681600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Medical Sciences"],"abstract":"<jats:p>Background: Pulmonary fibrosis (PF) and post-infectious fibrotic lung disease are characterized by marked heterogeneity in radiologic patterns, physiologic impairment, and clinical presentation. Conventional analytic approaches often fail to capture non-linear and multimodal relationships between structural imaging findings and functional limitation. Integrating imaging-derived representations with clinical and functional data using artificial intelligence (AI) may provide a more comprehensive characterization of disease heterogeneity. Objectives: The objective of this study was to develop and evaluate a multimodal AI framework combining imaging-derived embeddings and structured clinical data to identify atypical clinical\u2013radiologic profiles in patients with pulmonary fibrosis using unsupervised anomaly detection. Methods: A retrospective cohort of 41 patients with radiologically confirmed pulmonary fibrosis or post-infectious fibrotic lung disease was analyzed. Deep imaging embeddings were extracted from baseline thoracic CT examinations using a pretrained convolutional neural network and integrated with standardized clinical and functional variables. A multimodal variational autoencoder (VAE) was trained in an unsupervised manner to learn the distribution of typical patient profiles. Patient-specific anomaly scores were derived from reconstruction error plus latent regularization (\u03b2\u00b7KL divergence). Associations between anomaly scores, disease severity, and clinical markers were assessed using Spearman rank correlation. Results: Anomaly scores were right-skewed (median 26.91, IQR 22.87\u201332.11; range 19.75\u201346.18). Patients above the 85th percentile (anomaly score \u2265 33.85) comprised 7\/41 (17.1%) of the cohort and occurred across all clinician-assigned severity categories (mild 3, moderate 1, severe 3). Anomaly scores overlapped substantially across severity groups, with similar medians (mild 26.47, moderate 28.55, severe 28.23). Correlations with conventional severity markers were weak and non-significant, including DLCO (% predicted; \u03c1 = \u22120.25, p = 0.115) and FEV1 (% predicted; \u03c1 = \u22120.22, p = 0.165), a pattern consistent with anomaly scores reflecting multimodal deviation rather than severity alone, while acknowledging the exploratory nature of the analysis. Highly anomalous patients frequently exhibited discordant clinical\u2013radiologic profiles, including preserved functional capacity despite marked imaging-derived deviation or disproportionate physiological impairment relative to imaging patterns. Conclusions: This proof-of-concept study demonstrates that multimodal VAE-based anomaly detection integrating imaging-derived embeddings with clinical data can quantify clinical\u2013radiologic heterogeneity in pulmonary fibrosis beyond conventional severity stratification. Unsupervised anomaly detection provides a complementary framework for identifying atypical multimodal profiles and supporting individualized phenotyping and hypothesis generation in fibrotic lung disease. Given the modest cohort size, these findings should be interpreted as illustrative and hypothesis-generating rather than generalizable.<\/jats:p>","DOI":"10.3390\/medsci14010076","type":"journal-article","created":{"date-parts":[[2026,2,11]],"date-time":"2026-02-11T09:16:08Z","timestamp":1770801368000},"page":"76","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Multimodal Autoencoder\u2013Based Anomaly Detection Reveals Clinical\u2013Radiologic Heterogeneity in Pulmonary Fibrosis"],"prefix":"10.3390","volume":"14","author":[{"given":"Constantin","family":"Ghimu\u0219","sequence":"first","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2791-3400","authenticated-orcid":false,"given":"C\u0103lin Gheorghe","family":"Buzea","sequence":"additional","affiliation":[{"name":"National Institute of Research and Development for Technical Physics, IFT Ia\u0219i, 700050 Iasi, Romania"},{"name":"Clinical Emergency Hospital \u201cProf. Dr. Nicolae Oblu\u201d Ia\u0219i, 700309 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0024-0237","authenticated-orcid":false,"given":"Alin Hora\u021biu","family":"Nedelcu","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"},{"name":"Radiology Clinic, Recovery Hospital, 700661 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-3003-5420","authenticated-orcid":false,"given":"Vlad Florin","family":"Oiegar","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8147-3632","authenticated-orcid":false,"given":"Ancu\u021ba","family":"Lupu","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6325-7370","authenticated-orcid":false,"given":"R\u0103zvan Tudor","family":"Tepordei","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Simona Alice Partene","family":"Vicoleanu","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ana Maria","family":"Dumitrescu","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Manuela","family":"Ursaru","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"},{"name":"Radiology Clinic, \u201cSf Spiridon\u201d County Clinical Emergency Hospital, 700111 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gabriel","family":"Statescu","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8244-0522","authenticated-orcid":false,"given":"Emil","family":"Anton","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2640-8795","authenticated-orcid":false,"given":"Vasile Valeriu","family":"Lupu","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1789-4277","authenticated-orcid":false,"given":"Paraschiva","family":"Postolache","sequence":"additional","affiliation":[{"name":"Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,2,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1811","DOI":"10.1056\/NEJMra1705751","article-title":"Idiopathic Pulmonary Fibrosis","volume":"378","author":"Lederer","year":"2018","journal-title":"N. Engl. J. Med."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1800692","DOI":"10.1183\/13993003.00692-2018","article-title":"IPF Consensus Working Group. What\u2019s in a name? That which we call IPF, by any other name would act the same","volume":"51","author":"Wells","year":"2018","journal-title":"Eur. Respir. J."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"180076","DOI":"10.1183\/16000617.0076-2018","article-title":"Presentation, diagnosis and clinical course of the spectrum of progressive-fibrosing interstitial lung diseases","volume":"27","author":"Cottin","year":"2018","journal-title":"Eur. Respir. Rev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"807","DOI":"10.1016\/S2213-2600(20)30225-3","article-title":"Pulmonary fibrosis and COVID-19: The potential role for antifibrotic therapy","volume":"8","author":"George","year":"2020","journal-title":"Lancet Respir. Med."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1242","DOI":"10.1136\/thoraxjnl-2021-217031","article-title":"Pulmonary fibrosis 4 months after COVID-19 is associated with severity of illness and blood leucocyte telomere length","volume":"76","author":"McGroder","year":"2021","journal-title":"Thorax"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1136\/thoraxjnl-2019-213865","article-title":"Serial CT analysis in idiopathic pulmonary fibrosis: Comparison of visual features that determine patient outcome","volume":"75","author":"Jacob","year":"2020","journal-title":"Thorax"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"138","DOI":"10.1016\/S2213-2600(17)30433-2","article-title":"Diagnostic criteria for idiopathic pulmonary fibrosis: A Fleischner Society White Paper","volume":"6","author":"Lynch","year":"2018","journal-title":"Lancet Respir. Med."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"837","DOI":"10.1016\/S2213-2600(18)30286-8","article-title":"Deep learning for classifying fibrotic lung disease on high-resolution computed tomography: A case-cohort study","volume":"6","author":"Walsh","year":"2018","journal-title":"Lancet Respir. Med."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1318","DOI":"10.1007\/s00330-017-5053-z","article-title":"Serial automated quantitative CT analysis in idiopathic pulmonary fibrosis: Functional correlations and comparison with changes in visual CT scores","volume":"28","author":"Jacob","year":"2018","journal-title":"Eur. Radiol."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1359","DOI":"10.1016\/j.chest.2018.08.1083","article-title":"Incidence and prevalence of idiopathic pulmonary fibrosis","volume":"154","author":"Raghu","year":"2018","journal-title":"Chest"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1016\/j.ccm.2011.12.005","article-title":"Idiopathic pulmonary fibrosis: Phenotypes and comorbidities","volume":"33","author":"Fell","year":"2012","journal-title":"Clin. Chest Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"152","DOI":"10.1016\/j.rmed.2017.06.013","article-title":"Correlation of pulmonary function and usual interstitial pneumonia computed tomography patterns in idiopathic pulmonary fibrosis","volume":"129","author":"Arcadu","year":"2017","journal-title":"Respir. Med."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1164\/rccm.201006-0894CI","article-title":"Clinical course and prediction of survival in idiopathic pulmonary fibrosis","volume":"183","author":"Ley","year":"2011","journal-title":"Am. J. Respir. Crit. Care Med."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"160035","DOI":"10.1038\/sdata.2016.35","article-title":"MIMIC-III, a freely accessible critical care database","volume":"3","author":"Johnson","year":"2016","journal-title":"Sci. Data"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1148\/radiol.16142770","article-title":"Natural Language Processing in Radiology: A Systematic Review","volume":"279","author":"Pons","year":"2016","journal-title":"Radiology"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"24","DOI":"10.1038\/s41591-018-0316-z","article-title":"A guide to deep learning in healthcare","volume":"25","author":"Esteva","year":"2019","journal-title":"Nat. Med."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1038\/s41591-018-0300-7","article-title":"High-performance medicine: The convergence of human and artificial intelligence","volume":"25","author":"Topol","year":"2019","journal-title":"Nat. Med."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"883","DOI":"10.1164\/rccm.202112-2684OC","article-title":"Deep learning\u2013based outcome prediction in progressive fibrotic lung disease using high-resolution computed tomography","volume":"206","author":"Walsh","year":"2022","journal-title":"Am. J. Respir. Crit. Care Med."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Sakurada, M., and Yairi, T. (2014). Anomaly detection using autoencoders with nonlinear dimensionality reduction. MLSDA\u201914: Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis, Association for Computing Machinery.","DOI":"10.1145\/2689746.2689747"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"26094","DOI":"10.1038\/srep26094","article-title":"Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records","volume":"6","author":"Miotto","year":"2016","journal-title":"Sci. Rep."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Choi, E., Bahadori, M.T., Searles, E., Coffey, C., and Sun, J. (2016). Multi-layer representation learning for medical concepts. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD\u201916), ACM.","DOI":"10.1145\/2939672.2939823"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"1347","DOI":"10.1056\/NEJMra1814259","article-title":"Machine Learning in Medicine","volume":"380","author":"Rajkomar","year":"2019","journal-title":"N. Engl. J. Med."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"57","DOI":"10.1186\/s12931-019-1022-1","article-title":"The natural history of progressive fibrosing interstitial lung diseases","volume":"20","author":"Kolb","year":"2019","journal-title":"Respir. Res."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Johnston, J., Dorrian, D., Linden, D., Stanel, S.C., Rivera-Ortega, P., and Chaudhuri, N. (2023). Pulmonary Sequelae of COVID-19: Focus on Interstitial Lung Disease. Cells, 12.","DOI":"10.3390\/cells12182238"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"e72500","DOI":"10.7554\/eLife.72500","article-title":"Investigating phenotypes of pulmonary COVID-19 recovery: A longitudinal observational prospective multicenter trial","volume":"11","author":"Sonnweber","year":"2022","journal-title":"Elife"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"m3001","DOI":"10.1136\/bmj.m3001","article-title":"Long term respiratory complications of COVID-19","volume":"370","author":"Fraser","year":"2020","journal-title":"BMJ"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"465","DOI":"10.1164\/rccm.202311-2185OC","article-title":"Deep Learning-based Segmentation of Computed Tomography Scans Predicts Disease Progression and Mortality in Idiopathic Pulmonary Fibrosis","volume":"210","author":"Thillai","year":"2024","journal-title":"Am. J. Respir. Crit. Care Med."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1148\/radiol.2021204164","article-title":"Content-based Image Retrieval by Using Deep Learning for Interstitial Lung Disease Diagnosis with Chest CT","volume":"302","author":"Choe","year":"2022","journal-title":"Radiology"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"e18","DOI":"10.1164\/rccm.202202-0399ST","article-title":"Idiopathic Pulmonary Fibrosis (an Update) and Progressive Pulmonary Fibrosis in Adults: An Official ATS\/ERS\/JRS\/ALAT Clinical Practice Guideline","volume":"205","author":"Raghu","year":"2022","journal-title":"Am. J. Respir. Crit. Care Med."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"729","DOI":"10.1016\/S2213-2600(22)00223-5","article-title":"2022 update on clinical practice guidelines for idiopathic pulmonary fibrosis and progressive pulmonary fibrosis","volume":"10","author":"Tzilas","year":"2022","journal-title":"Lancet Respir. Med."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"2103187","DOI":"10.1183\/13993003.03187-2021","article-title":"Progressive pulmonary fibrosis: An expert group consensus statement","volume":"61","author":"Rajan","year":"2023","journal-title":"Eur. Respir. J."}],"container-title":["Medical Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2076-3271\/14\/1\/76\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,12]],"date-time":"2026-02-12T15:57:02Z","timestamp":1770911822000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2076-3271\/14\/1\/76"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,10]]},"references-count":31,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["medsci14010076"],"URL":"https:\/\/doi.org\/10.3390\/medsci14010076","relation":{},"ISSN":["2076-3271"],"issn-type":[{"value":"2076-3271","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,10]]}}}