{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T18:34:04Z","timestamp":1777055644545,"version":"3.51.4"},"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"4","license":[{"start":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T00:00:00Z","timestamp":1776988800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Cross-modal fusion of medical imaging and laboratory data is a key pathway for accurate diagnosis of diseases, yet it is constrained by issues such as the modal heterogeneity gap, accumulation of feature redundancy, and efficiency imbalance. Existing methods struggle to balance precision and clinical adaptability, and some rely on simulated data leading to limited generalization ability. To address these challenges, we propose the Cross-Modal Alignment-Pruning Fusion model (CMAP-Fusion), which achieves optimization through modular collaboration of \u201cencoding alignment\u2009\u2192\u2009redundant pruning\u2009\u2192\u2009fusion prediction\u201d: ViT-B\/16 is used to complete imaging feature extraction and dimension alignment, the SmartTrim dynamic pruning module screens key features and reduces redundancy, and the Cross-Modal Transformer (CMT) mines deep associations between dual modalities. Experiments on the COVID-19 Radiography Dataset, ISIC Skin Cancer Dataset, and ChestX-ray14 Dataset demonstrate that the model achieves accuracies of 95.3%, 89.7%, and 93.6% respectively, representing an improvement of 3.1% to 4.1% compared with optimal baselines. Meanwhile, the number of parameters is reduced by 44.2%, computational complexity is decreased by more than 43%, and cross-modal similarity and feature sparsity are significantly superior to baselines. This model realizes the synergistic optimization of \u201cprecision-efficiency-generalization,\u201d providing an efficient solution for medical cross-modal fusion. In the future, we will expand to multi-source modalities and multi-disease scenarios, strengthen clinical multi-center validation, further improve the model\u2019s interpretability and clinical acceptance, and facilitate the lightweight deployment of medical AI.<\/jats:p>","DOI":"10.1371\/journal.pone.0346875","type":"journal-article","created":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T17:34:50Z","timestamp":1777052090000},"page":"e0346875","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["CMAP-Fusion: A cross-modal feature selection and model pruning framework for laboratory and imaging data"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-0604-3545","authenticated-orcid":true,"given":"Chong","family":"Liu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lei","family":"Yang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinmeng","family":"Lei","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,4,24]]},"reference":[{"issue":"2","key":"pone.0346875.ref001","doi-asserted-by":"crossref","first-page":"770","DOI":"10.1109\/TCSVT.2023.3289170","article-title":"Cross-Modal Transformers for Infrared and Visible Image Fusion","volume":"34","author":"S Park","year":"2024","journal-title":"IEEE Trans Circuits Syst Video Technol"},{"issue":"1","key":"pone.0346875.ref002","doi-asserted-by":"crossref","first-page":"15","DOI":"10.62762\/TSCC.2025.862776","article-title":"Intelligent fire recognition for surveillance control using cascaded multi-scale attention framework","volume":"3","author":"IM Khan","year":"2026","journal-title":"ICCK Trans Sens Commun Control"},{"issue":"2","key":"pone.0346875.ref003","doi-asserted-by":"crossref","first-page":"761","DOI":"10.1109\/TSTE.2024.3482360","article-title":"SolarFusionNet: Enhanced Solar Irradiance Forecasting via Automated Multi-Modal Feature Selection and Cross-Modal Fusion","volume":"16","author":"T Jing","year":"2025","journal-title":"IEEE Trans Sustain Energy"},{"issue":"9","key":"pone.0346875.ref004","doi-asserted-by":"crossref","first-page":"7931","DOI":"10.1109\/TCOMM.2025.3548021","article-title":"Advancing Multi-Modal Beam Prediction With Cross-Modal Feature Enhancement and Dynamic Fusion Mechanism","volume":"73","author":"Q Zhu","year":"2025","journal-title":"IEEE Trans Commun"},{"key":"pone.0346875.ref005","doi-asserted-by":"crossref","first-page":"108042","DOI":"10.1016\/j.patcog.2021.108042","article-title":"Non-linear and selective fusion of cross-modal images","volume":"119","author":"A Fang","year":"2021","journal-title":"Pattern Recognition"},{"issue":"2","key":"pone.0346875.ref006","doi-asserted-by":"crossref","first-page":"1177","DOI":"10.1109\/TAFFC.2024.3498443","article-title":"A Low-Rank Matching Attention Based Cross-Modal Feature Fusion Method for Conversational Emotion Recognition","volume":"16","author":"Y Shou","year":"2025","journal-title":"IEEE Trans Affective Comput"},{"issue":"4","key":"pone.0346875.ref007","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3458281","article-title":"Cross-Modal Hybrid Feature Fusion for Image-Sentence Matching","volume":"17","author":"X Xu","year":"2021","journal-title":"ACM Trans Multimedia Comput Commun Appl"},{"issue":"7","key":"pone.0346875.ref008","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0298102","article-title":"Brain tumor classification in VIT-B\/16 based on relative position encoding and residual MLP","volume":"19","author":"S Hong","year":"2024","journal-title":"PLoS One"},{"key":"pone.0346875.ref009","first-page":"727","article-title":"An Improved CNN-LSTM Model Compression Pruning Algorithm","volume-title":"Lecture Notes on Data Engineering and Communications Technologies","author":"J Wu","year":"2022"},{"issue":"1","key":"pone.0346875.ref010","doi-asserted-by":"crossref","first-page":"5473","DOI":"10.1038\/s41598-025-89202-x","article-title":"MemoCMT: multimodal emotion recognition using cross-modal transformer-based feature fusion","volume":"15","author":"M Khan","year":"2025","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0346875.ref011","doi-asserted-by":"crossref","first-page":"379","DOI":"10.1186\/s12880-025-01901-z","article-title":"Enhanced CoAtNet based hybrid deep learning architecture for automated tuberculosis detection in human chest X-rays","volume":"25","author":"G Siddharth","year":"2025","journal-title":"BMC Med Imaging"},{"key":"pone.0346875.ref012","doi-asserted-by":"crossref","first-page":"170126","DOI":"10.1109\/ACCESS.2025.3615654","article-title":"HFT-Net: Hybrid Fusion Transformer Network for Multi-Source Breast Cancer Classification","volume":"13","author":"K Guzel","year":"2025","journal-title":"IEEE Access"},{"key":"pone.0346875.ref013","doi-asserted-by":"crossref","unstructured":"Chandrasiri MDN, Talagala PD. Cross-ViT: Cross-attention Vision Transformer for Image Duplicate Detection. In: 2023 8th International Conference on Information Technology Research (ICITR), 2023. 1\u20136. https:\/\/doi.org\/10.1109\/icitr61062.2023.10382916","DOI":"10.1109\/ICITR61062.2023.10382916"},{"key":"pone.0346875.ref014","doi-asserted-by":"crossref","first-page":"20552076231207197","DOI":"10.1177\/20552076231207197","article-title":"HI-MViT: A lightweight model for explainable skin disease classification based on modified MobileViT","volume":"9","author":"Y Ding","year":"2023","journal-title":"Digit Health"},{"key":"pone.0346875.ref015","doi-asserted-by":"crossref","unstructured":"Wang Z, Chen J, Zhou W, Zhu H, Liang J, Shan L, et al. SmartTrim: Adaptive Tokens and Attention Pruning for Efficient Vision-Language Models. In: Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), 2024. 14937\u201353. https:\/\/doi.org\/10.63317\/23hifgwadz5j","DOI":"10.63317\/23hifgwadz5j"},{"key":"pone.0346875.ref016","doi-asserted-by":"crossref","unstructured":"Chen Z, Zhong F, Luo Q, Zhang X, Zheng Y. Edgevit: Efficient visual modeling for edge computing. In: International conference on wireless algorithms, systems, and applications, 2022. 393\u2013405.","DOI":"10.1007\/978-3-031-19211-1_33"},{"issue":"4","key":"pone.0346875.ref017","article-title":"FEF-Net: feature enhanced fusion network with crossmodal attention for multimodal humor prediction","volume":"30","author":"P Gao","year":"2024","journal-title":"Multimedia Systems"},{"key":"pone.0346875.ref018","doi-asserted-by":"crossref","unstructured":"Yu J, Zheng Y, Wang L, Wang Y, Xu S. Cross-Modal Facial Expression Recognition with Global Channel-Spatial Attention: Modal Enhancement and Proportional Criterion Fusion. In: 2025 IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2025. 5707\u201314. https:\/\/doi.org\/10.1109\/cvprw67362.2025.00569","DOI":"10.1109\/CVPRW67362.2025.00569"},{"issue":"15","key":"pone.0346875.ref019","article-title":"Multimodal prompt learning with selective feature fusion: towards robust cross-modal alignment","volume":"55","author":"J Han","year":"2025","journal-title":"Appl Intell"},{"key":"pone.0346875.ref020","doi-asserted-by":"crossref","first-page":"935","DOI":"10.1109\/TASLPRO.2025.3533359","article-title":"Cross-Modal Knowledge Distillation With Multi-Stage Adaptive Feature Fusion for Speech Separation","volume":"33","author":"C Fan","year":"2025","journal-title":"IEEE Trans Audio, Speech Lang Process"},{"key":"pone.0346875.ref021","doi-asserted-by":"crossref","first-page":"922","DOI":"10.1109\/JRFID.2022.3213753","article-title":"CNN-BiLSTM Short-Term Wind Power Forecasting Method Based on Feature Selection","volume":"6","author":"Y Chen","year":"2022","journal-title":"IEEE J Radio Freq Identif"},{"issue":"5","key":"pone.0346875.ref022","doi-asserted-by":"crossref","first-page":"5457","DOI":"10.1007\/s10489-021-02524-x","article-title":"Feature selection based on mutual information with correlation coefficient","volume":"52","author":"H Zhou","year":"2021","journal-title":"Appl Intell"},{"issue":"20","key":"pone.0346875.ref023","doi-asserted-by":"crossref","first-page":"17647","DOI":"10.1007\/s00521-022-07402-2","article-title":"A novel method for intrusion detection in computer networks by identifying multivariate outliers and ReliefF feature selection","volume":"34","author":"B Uzun","year":"2022","journal-title":"Neural Comput & Applic"},{"key":"pone.0346875.ref024","doi-asserted-by":"crossref","unstructured":"Liao R, Zhao C, Li J, Feng W, Lyu Y, Chen B, et al. CATP: Cross-Attention Token Pruning for Accuracy Preserved Multimodal Model Inference. In: 2025 IEEE Conference on Artificial Intelligence (CAI), 2025. 1100\u20134. https:\/\/doi.org\/10.1109\/cai64502.2025.00191","DOI":"10.1109\/CAI64502.2025.00191"},{"issue":"16","key":"pone.0346875.ref025","doi-asserted-by":"crossref","first-page":"11259","DOI":"10.1007\/s00500-023-08587-x","article-title":"A high-quality feature selection method based on frequent and correlated items for text classification","volume":"27","author":"H Mamdouh Farghaly","year":"2023","journal-title":"Soft Comput"},{"issue":"1","key":"pone.0346875.ref026","doi-asserted-by":"crossref","first-page":"54","DOI":"10.62762\/TSCC.2025.553018","article-title":"A safety-critical control scheme for spacecraft relative motion tracking based on the fully actuated system approach and offline QP solutions","volume":"3","author":"X Yang","year":"2026","journal-title":"ICCK Trans Sens Commun Control"},{"key":"pone.0346875.ref027","doi-asserted-by":"crossref","unstructured":"Cao Y, Xing S, Yu Z, Wu C, Weng Z, Du J. An optimal feature selection fusion method of visual models for CLIP. In: 2025 44th Chinese Control Conference (CCC), 2025. 8833\u20138.","DOI":"10.23919\/CCC64809.2025.11179370"},{"key":"pone.0346875.ref028","doi-asserted-by":"crossref","unstructured":"Machado R, Rodrigues R, Neto L, Barbosa L, Bessa M, Melo M. Immersive Virtual Reality Training Platforms Powered by Digital Twin Technologies: The Smartcut Case Study. In: 2023 International Conference on Graphics and Interaction (ICGI), 2023. 1\u20138. https:\/\/doi.org\/10.1109\/icgi60907.2023.10452724","DOI":"10.1109\/ICGI60907.2023.10452724"},{"key":"pone.0346875.ref029","doi-asserted-by":"crossref","first-page":"8166","DOI":"10.1109\/JSTARS.2023.3294624","article-title":"Vehicle Detection Based on Adaptive Multimodal Feature Fusion and Cross-Modal Vehicle Index Using RGB-T Images","volume":"16","author":"Y Wu","year":"2023","journal-title":"IEEE J Sel Top Appl Earth Observations Remote Sensing"},{"key":"pone.0346875.ref030","doi-asserted-by":"crossref","first-page":"102577","DOI":"10.1016\/j.compmedimag.2025.102577","article-title":"CMT-FFNet: A CMT-based feature-fusion network for predicting TACE treatment response in hepatocellular carcinoma","volume":"124","author":"S Wang","year":"2025","journal-title":"Comput Med Imaging Graph"},{"key":"pone.0346875.ref031","doi-asserted-by":"crossref","first-page":"131671","DOI":"10.1109\/ACCESS.2022.3219200","article-title":"CMJRT: Cross-Modal Joint Representation Transformer for Multimodal Sentiment Analysis","volume":"10","author":"M Xu","year":"2022","journal-title":"IEEE Access"},{"issue":"1","key":"pone.0346875.ref032","article-title":"The RSNA International COVID-19 Open Radiology Database (RICORD)","volume":"299","author":"EB Tsai","year":"2021","journal-title":"Radiology"},{"key":"pone.0346875.ref033","doi-asserted-by":"crossref","unstructured":"Wang X, Peng Y, Lu L, Lu Z, Bagheri M, Summers RM. ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. 3462\u201371. https:\/\/doi.org\/10.1109\/cvpr.2017.369","DOI":"10.1109\/CVPR.2017.369"},{"key":"pone.0346875.ref034","doi-asserted-by":"crossref","first-page":"102305","DOI":"10.1016\/j.media.2021.102305","article-title":"Analysis of the ISIC image datasets: Usage, benchmarks and recommendations","volume":"75","author":"B Cassidy","year":"2022","journal-title":"Med Image Anal"},{"issue":"11","key":"pone.0346875.ref035","doi-asserted-by":"crossref","first-page":"1403","DOI":"10.3390\/pharmaceutics16111403","article-title":"Skin Structure, Physiology, and Pathology in Topical and Transdermal Drug Delivery","volume":"16","author":"S Brito","year":"2024","journal-title":"Pharmaceutics"},{"issue":"3","key":"pone.0346875.ref036","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0298305","article-title":"Analysis of dermoscopy images of multi-class for early detection of skin lesions by hybrid systems based on integrating features of CNN models","volume":"19","author":"M Alshahrani","year":"2024","journal-title":"Plos one"},{"issue":"05","key":"pone.0346875.ref037","doi-asserted-by":"crossref","DOI":"10.1117\/1.JEI.33.5.053049","article-title":"DIMF-Nets: depth-informed cross-modal fusion in three-stream networks for enhanced unsupervised video object segmentation","volume":"33","author":"Y Zhang","year":"2024","journal-title":"J Electron Imag"},{"issue":"4","key":"pone.0346875.ref038","doi-asserted-by":"crossref","first-page":"263","DOI":"10.62762\/TSCC.2025.675097","article-title":"Strip pooling coordinate attention with directional learning for intelligent fire recognition in smart cities","volume":"2","author":"AU Haider","year":"2025","journal-title":"ICCK Trans Sens Commun Control"},{"key":"pone.0346875.ref039","doi-asserted-by":"crossref","first-page":"110749","DOI":"10.1016\/j.chaos.2021.110749","article-title":"Using handpicked features in conjunction with ResNet-50 for improved detection of COVID-19 from chest X-ray images","volume":"145","author":"S Rajpal","year":"2021","journal-title":"Chaos Solitons Fractals"},{"key":"pone.0346875.ref040","article-title":"Breast Cancer Classification Using Graph Convolutional Networks and DenseNet121 with Pruning","author":"SJ Kavitha","year":"2025","journal-title":"J Shanghai Jiaotong Univ (Sci)"},{"key":"pone.0346875.ref041","doi-asserted-by":"crossref","first-page":"35716","DOI":"10.1109\/ACCESS.2024.3371151","article-title":"Classification of Paediatric Pneumonia Using Modified DenseNet-121 Deep-Learning Model","volume":"12","author":"TS Arulananth","year":"2024","journal-title":"IEEE Access"},{"key":"pone.0346875.ref042","article-title":"Cross-modal retrieval based on multi-dimensional feature fusion hashing","volume":"12","author":"D Ren","year":"2024","journal-title":"Front Phys"},{"issue":"6","key":"pone.0346875.ref043","doi-asserted-by":"crossref","DOI":"10.1002\/ima.23220","article-title":"CATNet: A Cross Attention and Texture-Aware Network for Polyp Segmentation","volume":"34","author":"Z Deng","year":"2024","journal-title":"International Journal of Imaging Systems and Technology"},{"key":"pone.0346875.ref044","first-page":"1","article-title":"Multiscale Feature Fusion Transformer With Hybrid Attention for Insulator Defect Detection","volume":"74","author":"J Xu","year":"2025","journal-title":"IEEE Trans Instrum Meas"},{"key":"pone.0346875.ref045","first-page":"294","volume-title":"Edgevits: Competing light-weight cnns on mobile devices with vision transformers. In: European conference on computer vision","author":"J Pan","year":"2022"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346875","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T17:35:00Z","timestamp":1777052100000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346875"}},"subtitle":[],"editor":[{"given":"Musa","family":"Aydin","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,4,24]]},"references-count":45,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4,24]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0346875","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,24]]}}}