{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T18:28:52Z","timestamp":1793039332692,"version":"4.5.0"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T00:00:00Z","timestamp":1781049600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100008786","name":"Universidad Nacional Mayor de San Marcos","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100008786","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Digit. Health"],"abstract":"<jats:p>\n                    Pleural effusion cytology is critical for diagnosing benign and malignant conditions, yet manual interpretation remains time-consuming and prone to subjectivity. The increasing burden of malignant pleural effusion in resource-constrained settings highlights the need for automated diagnostic solutions. This study presents an ensemble deep learning framework combining ResNet50V2, DenseNet121, and InceptionV3 architectures with transfer learning for classifying pleural cytology images into negative for malignancy (NFM) and malignant (MAL) categories. Three scenarios were evaluated: no data augmentation, 50% augmentation, and 300% augmentation. A local dataset of 1,292 images from Hospital Nacional Cayetano Heredia and an external Kaggle dataset (693 images) were used for training, validation, and independent testing. Performance was measured using accuracy, precision, recall, and\n                    <jats:italic>F<\/jats:italic>\n                    1-score across individual models and ensemble voting strategies (hard and soft voting). The ResNet\u2009+\u2009DenseNet ensemble with soft voting and 300% data augmentation achieved the highest accuracy (96.2% on the local dataset; 89.6% on the external dataset), outperforming individual models across all scenarios. Increasing the dataset through data augmentation significantly improved generalization and robustness. The proposed ensemble-based approach supports cytological diagnosis, potentially reducing diagnostic uncertainty in pleural carcinoma detection. Our findings demonstrate that ensemble deep learning models, optimized with data augmentation, can provide accurate and reproducible diagnostic support for pleural cytology, offering practical potential for deployment in low-resource healthcare settings and contributing to improved cancer diagnosis accessibility.\n                  <\/jats:p>","DOI":"10.3389\/fdgth.2026.1849898","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T05:42:11Z","timestamp":1781070131000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Ensemble based in transfer learning for cytological classification in pleural fluid"],"prefix":"10.3389","volume":"8","author":[{"given":"Frida","family":"L\u00f3pez-C\u00f3rdova","sequence":"first","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hugo","family":"Vega-Huerta","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gisella Luisa Elena","family":"Maquen-Ni\u00f1o","sequence":"additional","affiliation":[{"name":"Universidad Nacional Pedro Ruiz Gallo","place":["Lambayeque, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaime","family":"C\u00e1ceres-Pizarro","sequence":"additional","affiliation":[{"name":"Hospital Nacional Cayetano Heredia","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ciro","family":"Rodriguez","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Calder\u00f3n","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan","family":"Gamarra-Moreno","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Percy","family":"De-la-Cruz-VdV","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luis","family":"Guerra-Grados","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Santiago","family":"Moquillaza-Henr\u00edquez","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Oscar","family":"Benito-Pacheco","sequence":"additional","affiliation":[{"name":"Universidad Nacional Mayor de San Marcos","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ivan","family":"Adrianz\u00e9n-Olano","sequence":"additional","affiliation":[{"name":"Universidad Nacional Toribio Rodr\u00edguez de Mendoza de Amazonas","place":["Bagua, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mario","family":"Chauca","sequence":"additional","affiliation":[{"name":"Universidad Ricardo Palma","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,10]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1055","DOI":"10.1016\/j.mcna.2011.08.005","article-title":"Pleural effusions","volume":"95","author":"Light","year":"2011","journal-title":"Med Clin North Am"},{"key":"B2","doi-asserted-by":"publisher","first-page":"870","DOI":"10.3390\/clinpract14030068","article-title":"Infective pleural effusions\u2014a comprehensive narrative review article","volume":"14","author":"Abdulelah","year":"2024","journal-title":"Clin Pract"},{"key":"B3","article-title":"Risk and mortality factors for pleural effusions requiring diagnostic thoracentesis","volume":"4","author":"Baranda","year":"2022","journal-title":"Open Respir Arch"},{"key":"B4","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1016\/j.cireng.2022.06.009","article-title":"Recommendations of the Spanish society of thoracic surgery for the management of malignant pleural effusion","volume":"100","author":"Recuero D\u00edaz","year":"2022","journal-title":"Cir Esp"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1086","DOI":"10.5858\/arpa.2019-0429-OA","article-title":"Malignancies in pleural, peritoneal, and pericardial effusions","volume":"144","author":"Dermawan","year":"2020","journal-title":"Arch Pathol Lab Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"100","DOI":"10.33821\/603","article-title":"Actualidad en el manejo cl\u00ednico del derrame pleural maligno","volume":"32","author":"Rivera Rivera","year":"2022","journal-title":"Oncolog\u00eda"},{"key":"B7","doi-asserted-by":"publisher","first-page":"15593","DOI":"10.1007\/s11042-019-07821-9","article-title":"Transfer learning with pre-trained deep convolutional neural networks for serous cell classification","volume":"79","author":"Baykal","year":"2020","journal-title":"Multimed Tools Appl"},{"key":"B8","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1016\/j.arbr.2014.03.012","article-title":"Etiolog\u00eda del derrame pleural: An\u00e1lisis de m\u00e1s de 3.000 toracocentesis consecutivas","volume":"50","author":"Porcel","year":"2014","journal-title":"Arch Bronconeumol"},{"key":"B9","doi-asserted-by":"publisher","first-page":"e253","DOI":"10.1016\/S1470-2045(19)30154-8","article-title":"Digital pathology and artificial intelligence","volume":"20","author":"Niazi","year":"2019","journal-title":"Lancet Oncol"},{"key":"B10","doi-asserted-by":"publisher","first-page":"3608","DOI":"10.3390\/cancers15143608","article-title":"Deep learning for medical image-based cancer diagnosis","volume":"15","author":"Jiang","year":"2023","journal-title":"Cancers"},{"key":"B11","doi-asserted-by":"publisher","first-page":"958310","DOI":"10.3389\/fonc.2023.958310","article-title":"Applications of machine and deep learning to thyroid cytology and histopathology: a review","volume":"13","author":"Slabaugh","year":"2023","journal-title":"Front Oncol"},{"key":"B12","doi-asserted-by":"publisher","first-page":"3266","DOI":"10.1158\/1078-0432.CCR-18-2495","article-title":"Deep learning predicts lung cancer treatment response from serial medical imaging","volume":"25","author":"Xu","year":"2019","journal-title":"Clin Cancer Res"},{"key":"B13","doi-asserted-by":"publisher","first-page":"3529","DOI":"10.3390\/cancers14143529","article-title":"Recent application of artificial intelligence in non-gynecological cancer cytopathology: a systematic review","volume":"14","author":"Thakur","year":"2022","journal-title":"Cancers"},{"key":"B14","doi-asserted-by":"publisher","first-page":"609","DOI":"10.1038\/s41379-021-00987-4","article-title":"Deep convolutional neural network-based classification of cancer cells on cytological pleural effusion images","volume":"35","author":"Xie","year":"2022","journal-title":"Mod Pathol"},{"key":"B15","doi-asserted-by":"crossref","first-page":"101","DOI":"10.3991\/ijoe.v20i10.49089","article-title":"Classification of Alzheimer\u2019s disease based on deep learning using medical images","volume":"20","author":"Vega-Huerta","year":"2024","journal-title":"Int J Online Biomed Eng"},{"key":"B16","doi-asserted-by":"publisher","first-page":"159","DOI":"10.4103\/joc.joc_111_22","article-title":"The international system for reporting serous fluid cytopathology: an institutional experience on its implication and assessment of risk of malignancy in effusion cytology","volume":"40","author":"Kala","year":"2023","journal-title":"J Cytol"},{"key":"B17","doi-asserted-by":"publisher","first-page":"2223","DOI":"10.3390\/diagnostics11122223","article-title":"Application of the international system for reporting serous fluid cytopathology with cytohistological correlation and risk of malignancy assessment","volume":"11","author":"Pergaris","year":"2021","journal-title":"Diagnostics"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1126","DOI":"10.3390\/diagnostics11061126","article-title":"Cytohistological correlation in pleural effusions based on the international system for reporting serous fluid cytopathology","volume":"11","author":"Pinto","year":"2021","journal-title":"Diagnostics"},{"key":"B19","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1186\/s13000-022-01241-4","article-title":"A retrospective analysis of serous effusions based on the newly proposed international system for reporting serous fluid cytopathology: a report of 3633 cases in an oncological center","volume":"17","author":"Zhu","year":"2022","journal-title":"Diagn Pathol"},{"key":"B20","doi-asserted-by":"publisher","first-page":"821594","DOI":"10.3389\/fonc.2022.821594","article-title":"Deep learning-based classification of cancer cell in leptomeningeal metastasis on cytomorphologic features of cerebrospinal fluid","volume":"12","author":"Yu","year":"2022","journal-title":"Front Oncol"},{"key":"B21","doi-asserted-by":"crossref","DOI":"10.1109\/ICSPCC50002.2020.9259490","article-title":"A deep learning model for the automatic detection of malignancy in effusion cytology","author":"Aboobacker","year":"2020"},{"key":"B22","doi-asserted-by":"publisher","first-page":"103742","DOI":"10.1016\/j.media.2025.103742","article-title":"Interpretable multi-scale deep learning to detect malignancy in cell blocks and cytological smears of pleural effusion and identify aggressive endometrial cancer","volume":"106","author":"Wang","year":"2025","journal-title":"Med Image Anal"},{"key":"B23","doi-asserted-by":"publisher","first-page":"107625","DOI":"10.1016\/j.bspc.2025.107625","article-title":"Anomaly detection applied to the classification of cytology images","volume":"105","author":"Carlo Bruno","year":"2025","journal-title":"Biomed Signal Process Control"},{"key":"B24","doi-asserted-by":"crossref","first-page":"109710","DOI":"10.1016\/j.compbiomed.2025.109710","article-title":"A multimodal deep learning model for cervical pre-cancers and cancers prediction: development and internal validation study","volume":"191","author":"Madathil","year":"2025","journal-title":"Comput Biol Med"},{"key":"B25","doi-asserted-by":"publisher","first-page":"21929","DOI":"10.1007\/s11042-023-16351-4","article-title":"Hybrid models for classifying histological images: an association of deep features by transfer learning with ensemble classifier","volume":"83","author":"de Oliveira","year":"2024","journal-title":"Multimed Tools Appl"},{"key":"B26","doi-asserted-by":"publisher","first-page":"160","DOI":"10.1159\/000538465","article-title":"A novel validated real-world dataset for the diagnosis of multiclass serous effusion cytology according to the international system and ground-truth validation data","volume":"68","author":"Abd-Almoniem","year":"2024","journal-title":"Acta Cytol"},{"key":"B27","doi-asserted-by":"publisher","first-page":"89","DOI":"10.3390\/diagnostics14010089","article-title":"Combining state-of-the-art pre-trained deep learning models: a noble approach for skin cancer detection using max voting ensemble","volume":"14","author":"Hossain","year":"2024","journal-title":"Diagnostics"},{"key":"B28","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1016\/j.media.2017.01.008","article-title":"Deep ensemble learning of sparse regression models for brain disease diagnosis","volume":"37","author":"Suk","year":"2017","journal-title":"Med Image Anal"},{"key":"B29","doi-asserted-by":"crossref","first-page":"138","DOI":"10.3991\/ijoe.v21i07.54323","article-title":"Construction of a new data set of pleural fluid cytological images for research","volume":"21","author":"L\u00f3pez-C\u00f3rdova","year":"2025","journal-title":"Int J Online Biomed Eng"},{"key":"B30","doi-asserted-by":"crossref","first-page":"651","DOI":"10.5209\/RGID.54719","article-title":"Conceptual representation of digital medical images: integrating context and visual content | Representaci\u00f3n conceptual de im\u00e1genes m\u00e9dicas digitales: Integraci\u00f3n de contexto y contenido visual","volume":"26","author":"Roa Mart\u00ednez","year":"2016","journal-title":"Rev Gen Inf Doc"},{"key":"B31","doi-asserted-by":"publisher","first-page":"387","DOI":"10.18280\/isi.300210","article-title":"Classification model of skin cancer using convolutional neural network","volume":"30","author":"Vega-Huerta","year":"2025","journal-title":"Ing\u00e9nierie Syst Inf"},{"key":"B32","doi-asserted-by":"publisher","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":"B33","first-page":"4700","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"key":"B34","year":""},{"key":"B35","doi-asserted-by":"publisher","first-page":"e1249","DOI":"10.1002\/widm.1249","article-title":"Ensemble learning: a survey","volume":"8","author":"Sagi","year":"2018","journal-title":"WIRES Data Min Knowl Discov"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1849898\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T05:42:13Z","timestamp":1781070133000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1849898\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,10]]},"references-count":35,"alternative-id":["10.3389\/fdgth.2026.1849898"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1849898","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,10]]},"article-number":"1849898"}}