{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,24]],"date-time":"2025-10-24T05:32:42Z","timestamp":1761283962916,"version":"build-2065373602"},"reference-count":105,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,10,24]],"date-time":"2025-10-24T00:00:00Z","timestamp":1761264000000},"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. Radiol."],"abstract":"\n The increasing number of computed tomography (CT) scan examinations and the time-intensive nature of manual analysis necessitate efficient automated methods to assist radiologists in managing their increasing workload. While deep learning approaches primarily classify abnormalities from three-dimensional (3D) CT images, radiologists also incorporate clinical indications and patient demographics, such as age and sex, for diagnosis. This study aims to enhance multilabel abnormality classification and automated report generation by integrating imaging and non-imaging data. We propose a multimodal deep learning model that combines 3D chest CT scans, clinical information reports, patient age, and sex to improve diagnostic accuracy. Our method extracts visual features from 3D volumes using a visual encoder, textual features from clinical indications via a pretrained language model, and demographic features through a lightweight feedforward neural network. These extracted features are projected into a shared representation space, concatenated, and processed by a projection head to predict abnormalities. For the multilabel classification task, incorporating clinical indications and patient demographics into an existing visual encoder, called CT-Net, improves the F1 score to 51.58, representing a\n \n \n +<\/mml:mo>\n \u0394<\/mml:mi>\n 6.13<\/mml:mn>\n %<\/mml:mi>\n <\/mml:math>\n <\/jats:inline-formula>\n increase over CT-Net alone. For the automated report generation task, we extend two existing methods, CT2Rep and CT-AGRG, by integrating clinical indications and demographic data. This integration enhances Clinical Efficacy metrics, yielding an F1 score improvement of\n \n \n +<\/mml:mo>\n \u0394<\/mml:mi>\n 14.78<\/mml:mn>\n %<\/mml:mi>\n <\/mml:math>\n <\/jats:inline-formula>\n for the CT2Rep extension and\n \n \n +<\/mml:mo>\n \u0394<\/mml:mi>\n 6.69<\/mml:mn>\n %<\/mml:mi>\n <\/mml:math>\n <\/jats:inline-formula>\n for the CT-AGRG extension. Our findings suggest that incorporating patient demographics and clinical information into deep learning frameworks can significantly improve automated CT scan analysis. This approach has the potential to enhance radiological workflows and facilitate more comprehensive and accurate abnormality detection in clinical practice.\n <\/jats:p>","DOI":"10.3389\/fradi.2025.1672364","type":"journal-article","created":{"date-parts":[[2025,10,24]],"date-time":"2025-10-24T05:27:34Z","timestamp":1761283654000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Integrating clinical indications and patient demographics for multilabel abnormality classification and automated report generation in 3D chest CT scans"],"prefix":"10.3389","volume":"5","author":[{"given":"Theo","family":"Di Piazza","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Carole","family":"Lazarus","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Olivier","family":"Nempont","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Loic","family":"Boussel","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,10,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"5097","DOI":"10.3390\/s20185097","article-title":"3D deep learning on medical images: a review","volume":"20","author":"Singh","year":"2020","journal-title":"Sensors (Basel)"},{"key":"B2","doi-asserted-by":"publisher","first-page":"377","DOI":"10.3238\/arztebl.2019.0377","article-title":"Pleural effusion in adults\u2014etiology, diagnosis, and treatment","volume":"116","author":"Jany","year":"2019","journal-title":"Deutsches \u00c4rzteblatt Int"},{"key":"B3","doi-asserted-by":"publisher","first-page":"605","DOI":"10.1016\/j.ccm.2011.09.001","article-title":"Lung cancer: epidemiology, etiology, and prevention","volume":"32","author":"Dela Cruz","year":"2011","journal-title":"Clin Chest Med"},{"article-title":"Cardiomegaly","year":"","author":"Amin","key":"B4"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1111\/jmiro.2013.57.issue-1","article-title":"Evidence-based guideline for the written radiology report: methods, recommendations and implementation challenges","volume":"57","author":"Goergen","year":"2013","journal-title":"J Med Imaging Radiat Oncol"},{"key":"B6","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1007\/s10278-016-9937-2","article-title":"Improving patient safety: avoiding unread imaging exams in the national VA enterprise electronic health record","volume":"30","author":"Bastawrous","year":"2017","journal-title":"J Digit Imaging"},{"key":"B7","doi-asserted-by":"publisher","first-page":"j4683","DOI":"10.1136\/bmj.j4683","article-title":"Radiologist shortage leaves patient care at risk, warns royal college","volume":"359","author":"Rimmer","year":"2017","journal-title":"BMJ (Clin Res Ed.)"},{"key":"B8","doi-asserted-by":"publisher","first-page":"8272","DOI":"10.1002\/mp.v51.11","article-title":"Weakly-supervised learning-based pathology detection and localization in 3D chest CT scans","volume":"51","author":"Djahnine","year":"2024","journal-title":"Med Phys"},{"key":"B9","doi-asserted-by":"publisher","first-page":"101857","DOI":"10.1016\/j.media.2020.101857","article-title":"Machine-learning-based multiple abnormality prediction with large-scale chest computed tomography volumes","volume":"67","author":"Draelos","year":"2021","journal-title":"Med Image Anal"},{"article-title":"Data from: A foundation model utilizing chest CT volumes and radiology reports for supervised-level zero-shot detection of abnormalities (2024)","year":"","author":"Hamamci","key":"B10"},{"key":"B11","doi-asserted-by":"publisher","first-page":"588","DOI":"10.1007\/978-3-319-19992-4_46","article-title":"Multi-scale convolutional neural networks for lung nodule classification","volume":"24","author":"Shen","year":"2015","journal-title":"Inf Process Med Imaging: Proc Conf"},{"key":"B12","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1111\/jep.2005.11.issue-1","article-title":"Demographic and clinical characteristics of patients admitted to medical departments","volume":"11","author":"Raveh","year":"2005","journal-title":"J Eval Clin Pract"},{"key":"B13","doi-asserted-by":"publisher","first-page":"5454","DOI":"10.1007\/s00330-021-07700-8","article-title":"Impact of clinical information on CT diagnosis by radiologist and subsequent clinical management by physician in acute abdominal pain","volume":"31","author":"Hattori","year":"2021","journal-title":"Eur Radiol"},{"volume-title":"Leveraging Clinical Indications and Demographics to Improve Multi-Label Abnormality Classification in 3D Chest CT Scans","year":"2025","author":"Di Piazza","key":"B14"},{"key":"B15","doi-asserted-by":"publisher","first-page":"c17","DOI":"10.1159\/000286345","article-title":"Measures of disease frequency: prevalence and incidence","volume":"115","author":"Noordzij","year":"2010","journal-title":"Nephron Clin Pract"},{"article-title":"Prevalence","year":"","author":"Tenny","key":"B16"},{"key":"B17","doi-asserted-by":"crossref","DOI":"10.1109\/ISBI60581.2025.10981073","article-title":"CT-AGRG: Automated abnormality-guided report generation from 3D chest CT volumes","author":"Di Piazza","year":""},{"article-title":"Data from: CT2Rep: automated radiology report generation for 3D medical imaging (2024)","year":"","author":"Hamamci","key":"B18"},{"article-title":"Data from: GenerateCT: text-conditional generation of 3D chest CT volumes (2023)","year":"","author":"Hamamci","key":"B19"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e210258","DOI":"10.1148\/ryai.210258","article-title":"RadBERT: adapting transformer-based language models to radiology","volume":"4","author":"Yan","year":"2022","journal-title":"Radiol Artif Intell"},{"key":"B21","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med Image Anal"},{"key":"B22","doi-asserted-by":"publisher","first-page":"292","DOI":"10.1007\/s42979-022-01166-1","article-title":"Survey of supervised learning for medical image processing","volume":"3","author":"Aljuaid","year":"2022","journal-title":"SN Comput Sci"},{"key":"B23","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.369","article-title":"ChestX-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases","author":"Wang","year":""},{"key":"B24","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-020-0273-z","article-title":"Automated abnormality classification of chest radiographs using deep convolutional neural networks","volume":"3","author":"Tang","year":"2020","journal-title":"npj Digit Med"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1144","DOI":"10.1002\/jmri.v50.4","article-title":"Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images","volume":"50","author":"Zhou","year":"2019","journal-title":"J Magn Reson Imaging"},{"article-title":"Data from: DENTEX: an abnormal tooth detection with dental enumeration and diagnosis benchmark for panoramic x-rays (2023)","year":"","author":"Hamamci","key":"B26"},{"article-title":"Data from: CheXpert: a large chest radiograph dataset with uncertainty labels and expert comparison (2019)","year":"","author":"Irvin","key":"B27"},{"article-title":"Data from: VinDr-CXR: an open dataset of chest x-rays with radiologist\u2019s annotations (2022)","year":"","author":"Nguyen","key":"B28"},{"article-title":"Data from: CheXNet: radiologist-level pneumonia detection on chest x-rays with deep learning (2017)","year":"","author":"Rajpurkar","key":"B29"},{"key":"B30","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","year":"2017","journal-title":"Nature"},{"article-title":"Data from: The MIMIC-CXR database (2019)","year":"","author":"Johnson","key":"B31"},{"article-title":"Data from: Computer-aided abnormality detection in chest radiographs in a clinical setting via domain-adaptation (2020)","year":"","author":"Dubey","key":"B32"},{"key":"B33","doi-asserted-by":"publisher","first-page":"200","DOI":"10.1186\/s12859-020-3503-0","article-title":"A multi-label classification model for full slice brain computerised tomography image","volume":"21","author":"Li","year":"2020","journal-title":"BMC Bioinform"},{"key":"B34","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-16443-9_63","article-title":"Anatomy-guided weakly-supervised abnormality localization in chest x-rays","author":"Yu","year":""},{"key":"B35","doi-asserted-by":"publisher","first-page":"101987","DOI":"10.1016\/j.compmedimag.2021.101987","article-title":"3D multi-scale, multi-task, and multi-label deep learning for prediction of lymph node metastasis in T1 lung adenocarcinoma patients\u2019 CT images","volume":"93","author":"Zhao","year":"2021","journal-title":"Comput Med Imaging Graph"},{"key":"B36","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-62469-9_7","article-title":"Can we trust deep learning based diagnosis? The impact of domain shift in chest radiograph classification","author":"Pooch","year":""},{"article-title":"Data from: Automatic lung cancer prediction from chest x-ray images using deep learning approach (2018)","year":"","author":"Ausawalaithong","key":"B37"},{"key":"B38","doi-asserted-by":"publisher","first-page":"511","DOI":"10.1016\/j.measurement.2019.05.076","article-title":"Identifying pneumonia in chest x-rays: a deep learning approach","volume":"145","author":"Jaiswal","year":"2019","journal-title":"Measurement"},{"key":"B39","doi-asserted-by":"publisher","first-page":"e495","DOI":"10.7717\/peerj-cs.495","article-title":"AI-driven deep CNN approach for multi-label pathology classification using chest x-rays","volume":"7","author":"Albahli","year":"2021","journal-title":"PeerJ Comput Sci"},{"article-title":"Data from: Chest x-rays classification: a multi-label and fine-grained problem (2018)","year":"","author":"Ge","key":"B40"},{"key":"B41","doi-asserted-by":"publisher","first-page":"104348","DOI":"10.1016\/j.compbiomed.2021.104348","article-title":"Deep-chest: multi-classification deep learning model for diagnosing COVID-19, pneumonia, and lung cancer chest diseases","volume":"132","author":"Ibrahim","year":"2021","journal-title":"Comput Biol Med"},{"key":"B42","doi-asserted-by":"publisher","first-page":"121165","DOI":"10.1016\/j.eswa.2023.121165","article-title":"3D unsupervised anomaly detection through virtual multi-view projection and reconstruction: clinical validation on low-dose chest computed tomography","volume":"236","author":"Kim","year":"2024","journal-title":"Expert Syst Appl"},{"article-title":"Data from: Attention is all you need (2023)","year":"","author":"Vaswani","key":"B43"},{"key":"B44","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV48922.2021.00676","article-title":"ViViT: a video vision transformer","author":"Arnab","year":""},{"article-title":"Data from: An image is worth 16\u00d716 words: transformers for image recognition at scale (2021)","year":"","author":"Dosovitskiy","key":"B45"},{"article-title":"Data from: Phenaki: variable length video generation from open domain textual description (2022)","year":"","author":"Villegas","key":"B46"},{"key":"B47","doi-asserted-by":"publisher","first-page":"423","DOI":"10.1109\/TPAMI.2018.2798607","article-title":"Multimodal machine learning: a survey and taxonomy","volume":"41","author":"Baltrusaitis","year":"2019","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B48","doi-asserted-by":"publisher","first-page":"136","DOI":"10.1038\/s41746-020-00341-z","article-title":"Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines","volume":"3","author":"Huang","year":"2020","journal-title":"npj Digit Med"},{"key":"B49","doi-asserted-by":"publisher","first-page":"8294","DOI":"10.1007\/s11227-019-03101-3","article-title":"Multimodal deep learning for finance: integrating and forecasting international stock markets","volume":"76","author":"Lee","year":"2020","journal-title":"J Supercomput"},{"key":"B50","doi-asserted-by":"publisher","first-page":"615","DOI":"10.1016\/j.procir.2021.03.083","article-title":"Deep multimodal learning for manufacturing problem solving","volume":"99","author":"Lichtenwalter","year":"2021","journal-title":"Procedia CIRP"},{"volume-title":"Multimodal Deep Learning Approach for Event Detection in Sports Using Amazon SageMaker | AWS Machine Learning Blog","year":"2021","author":"Sarraf","key":"B51"},{"key":"B52","doi-asserted-by":"publisher","first-page":"5162","DOI":"10.1002\/mp.2017.44.issue-10","article-title":"A deep feature fusion methodology for breast cancer diagnosis demonstrated on three imaging modality datasets","volume":"44","author":"Antropova","year":"2017","journal-title":"Med Phys"},{"key":"B53","doi-asserted-by":"publisher","first-page":"569","DOI":"10.1016\/j.neuroimage.2014.06.077","article-title":"Hierarchical feature representation and multimodal fusion with deep learning for AD\/MCI diagnosis","volume":"101","author":"Suk","year":"2014","journal-title":"NeuroImage"},{"key":"B54","doi-asserted-by":"publisher","first-page":"102260","DOI":"10.1016\/j.artmed.2022.102260","article-title":"A multi-modal fusion framework based on multi-task correlation learning for cancer prognosis prediction","volume":"126","author":"Tan","year":"2022","journal-title":"Artif Intell Med"},{"key":"B55","doi-asserted-by":"publisher","first-page":"15873","DOI":"10.1038\/s41598-023-41463-0","article-title":"MDF-Net for abnormality detection by fusing x-rays with clinical data","volume":"13","author":"Hsieh","year":"2023","journal-title":"Sci Rep"},{"key":"B56","doi-asserted-by":"publisher","first-page":"22388","DOI":"10.1038\/s41598-023-49514-2","article-title":"Brain age prediction using combined deep convolutional neural network and multi-layer perceptron algorithms","volume":"13","author":"Joo","year":"2023","journal-title":"Sci Rep"},{"article-title":"Multimodal deep learning","year":"","author":"Ngiam","key":"B57"},{"article-title":"Data from: What makes training multi-modal classification networks hard? (2020)","year":"","author":"Wang","key":"B58"},{"article-title":"Data from: Multimodal industrial anomaly detection by crossmodal feature mapping (2023)","year":"","author":"Costanzino","key":"B59"},{"key":"B60","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR42600.2020.01095","article-title":"Multi-modality cross attention network for image and sentence matching","author":"Wei","year":""},{"article-title":"Utilizing longitudinal chest x-rays and reports to pre-fill radiology reports (2023)","year":"","author":"Zhu","key":"B61"},{"article-title":"Data from: Cross-modality attention-based multimodal fusion for non-small cell lung cancer (NSCLC) patient survival prediction (2024)","year":"","author":"Deng","key":"B62"},{"key":"B63","doi-asserted-by":"publisher","first-page":"012045","DOI":"10.1088\/1742-6596\/1950\/1\/012045","article-title":"A survey on various deep learning models for automatic image captioning","volume":"1950","author":"Mathur","year":"2021","journal-title":"J Phys: Conf Ser"},{"article-title":"Data from: VizWiz grand challenge: answering visual questions from blind people (2018)","year":"","author":"Gurari","key":"B64"},{"article-title":"Data from: Show and tell: a neural image caption generator (2015)","year":"","author":"Vinyals","key":"B65"},{"article-title":"Data from: Multi-task prompt words learning for social media content generation (2024)","year":"","author":"Xue","key":"B66"},{"article-title":"Data from: A survey on biomedical image captioning (2019). arXiv:1905.13302 [cs] version: 1.","year":"","author":"Kougia","key":"B67"},{"article-title":"On the automatic generation of medical imaging reports. In: Gurevych I, Miyao Y, editors. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Melbourne, Australia: Association for Computational Linguistics (2018). p. 2577\u201386","year":"","author":"Jing","key":"B68"},{"article-title":"Data from: Generating radiology reports via memory-driven transformer","year":"","author":"Chen","key":"B69"},{"article-title":"Data from: Exploring and distilling posterior and prior knowledge for radiology report generation (2021)","year":"","author":"Liu","key":"B70"},{"article-title":"Data from: R2GenGPT: radiology report generation with frozen LLMs (2023)","year":"","author":"Wang","key":"B71"},{"article-title":"Data from: Advancing text-driven chest x-ray generation with policy-based reinforcement learning (2024)","year":"","author":"Han","key":"B72"},{"key":"B73","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52729.2023.00718","article-title":"Interactive and explainable region-guided radiology report generation","author":"Tanida","year":""},{"key":"B74","doi-asserted-by":"publisher","first-page":"368","DOI":"10.1109\/RBME.2024.3408456","article-title":"Automated radiology report generation: a review of recent advances","volume":"18","author":"Sloan","year":"2025","journal-title":"IEEE Rev Biomed Eng"},{"key":"B75","doi-asserted-by":"publisher","first-page":"106853","DOI":"10.1016\/j.cmpb.2022.106853","article-title":"Translating medical image to radiological report: adaptive multilevel multi-attention approach","volume":"221","author":"Gajbhiye","year":"2022","journal-title":"Comput Methods Programs Biomed"},{"article-title":"Data from: A self-guided framework for radiology report generation (2022)","year":"","author":"Li","key":"B76"},{"article-title":"Factual accuracy is not enough: planning consistent description order for radiology report generation. In: Goldberg Y, Kozareva Z, Zhang Y, editors. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. Abu Dhabi, United Arab Emirates: Association for Computational Linguistics (2022). p. 7123\u201338","year":"","author":"Nishino","key":"B77"},{"article-title":"Data from: Benchmarking and boosting radiology report generation for 3D high-resolution medical images (2024)","year":"","author":"Liu","key":"B78"},{"article-title":"Data from: MvKeTR: chest CT report generation with multi-view perception and knowledge enhancement (2025)","year":"","author":"Deng","key":"B79"},{"article-title":"Data from: Enhanced contrastive learning with multi-view longitudinal data for chest x-ray report generation (2025). arXiv: 2502.20056 [cs].","year":"","author":"Liu","key":"B80"},{"article-title":"Data from: Structural entities extraction and patient indications incorporation for chest x-ray report generation (2024). arXiv: 2405.14905 [eess].","year":"","author":"Liu","key":"B81"},{"article-title":"Data from: Pragmatic radiology report generation (2023). arXiv: 2311.17154 [cs].","year":"","author":"Nguyen","key":"B82"},{"article-title":"Language Models Are Unsupervised Multitask Learners (2019)","year":"","author":"Radford","key":"B83"},{"article-title":"Data from: DARE: data augmented relation extraction with GPT-2 (2020)","year":"","author":"Papanikolaou","key":"B84"},{"article-title":"Hounsfield unit. In: StatPearls. Treasure Island, FL: StatPearls Publishing (2024)","year":"","author":"DenOtter","key":"B85"},{"key":"B86","doi-asserted-by":"publisher","first-page":"20","DOI":"10.17148\/IARJSET","article-title":"Normalization: a preprocessing stage","author":"Patro","year":"2015","journal-title":"IARJSET"},{"article-title":"Data from: Deep residual learning for image recognition (2015)","year":"","author":"He","key":"B87"},{"article-title":"Data from: Learning spatiotemporal features with 3D convolutional networks (2015)","year":"","author":"Tran","key":"B88"},{"article-title":"Adapting pre-trained vision transformers from 2D to 3D through weight inflation improves medical image segmentation (2023)","year":"","author":"Zhang","key":"B89"},{"key":"B90","doi-asserted-by":"crossref","first-page":"248","DOI":"10.1109\/CVPR.2009.5206848","article-title":"ImageNet: A large-scale hierarchical image database","volume-title":"2009 IEEE Conference on Computer Vision and Pattern Recognition","author":"Deng","year":"2009"},{"article-title":"Data from: A simple framework for contrastive learning of visual representations (2020)","year":"","author":"Chen","key":"B91"},{"article-title":"Data from: BERT: pre-training of deep bidirectional transformers for language understanding (2019)","year":"","author":"Devlin","key":"B92"},{"article-title":"Data from: How to fine-tune BERT for text classification? (2020)","year":"","author":"Sun","key":"B93"},{"key":"B94","doi-asserted-by":"publisher","first-page":"829","DOI":"10.1162\/neco_a_01273","article-title":"A survey on deep learning for multimodal data fusion","volume":"32","author":"Gao","year":"2020","journal-title":"Neural Comput"},{"article-title":"Data from: Layer normalization (2016)","year":"","author":"Ba","key":"B95"},{"key":"B96","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1111\/j.2517-6161.1952.tb00104.x","article-title":"Rational decisions","volume":"14","author":"Good","year":"1952","journal-title":"J R Stat Soc Ser B (Methodol)"},{"article-title":"Data from: Adam: A method for stochastic optimization (2017)","year":"","author":"Kingma","key":"B97"},{"article-title":"Data from: Same task, more tokens: the impact of input length on the reasoning performance of large language models (2024)","year":"","author":"Levy","key":"B98"},{"article-title":"What would Elsa do? freezing layers during transformer fine-tuning","year":"","author":"Lee","key":"B99"},{"key":"B100","doi-asserted-by":"publisher","first-page":"6086","DOI":"10.1038\/s41598-024-56706-x","article-title":"Evaluation metrics and statistical tests for machine learning","volume":"14","author":"Rainio","year":"2024","journal-title":"Sci Rep"},{"article-title":"Data from: Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation (2020)","year":"","author":"Powers","key":"B101"},{"article-title":"Data from: BioMistral: a collection of open-source pretrained large language models for medical domains (2024)","year":"","author":"Labrak","key":"B102"},{"key":"B103","doi-asserted-by":"crossref","DOI":"10.3115\/1073083.1073135","article-title":"Bleu: a method for automatic evaluation of machine translation","author":"Papineni","year":""},{"article-title":"Data from: BERTScore: evaluating text generation with BERT (2020)","year":"","author":"Zhang","key":"B104"},{"article-title":"Data from: CRG score: a distribution-aware clinical metric for radiology report generation (2025)","year":"","author":"Hamamci","key":"B105"}],"container-title":["Frontiers in Radiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2025.1672364\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,24]],"date-time":"2025-10-24T05:27:48Z","timestamp":1761283668000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2025.1672364\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,24]]},"references-count":105,"alternative-id":["10.3389\/fradi.2025.1672364"],"URL":"https:\/\/doi.org\/10.3389\/fradi.2025.1672364","relation":{},"ISSN":["2673-8740"],"issn-type":[{"type":"electronic","value":"2673-8740"}],"subject":[],"published":{"date-parts":[[2025,10,24]]},"article-number":"1672364"}}