{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T18:24:27Z","timestamp":1787163867060,"version":"build-2736575974"},"reference-count":36,"publisher":"Public Library of Science (PLoS)","issue":"8","license":[{"start":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T00:00:00Z","timestamp":1787097600000},"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>Accurate segmentation of gastrointestinal stromal tumour (GIST) is always challenging task due to tissues intestines may have similar intensity values, which makes boundary delineation harder with only one source of contrast. Presently, clinicians make use of CT scans for manually annotate and diagnosis of GIST, which is ineffective and susceptible to subjectivity. To improve tumour delineation and effective segmentation of GIST, here we proposed two stage deep learning model for this purpose. In the first stage an DIF-Net is used to fuse CT and PET images, helps to improve quality of image for accurate delineation. In further stage the tumour region is effectively segmented by employing a Segformer architecture supplemented with a Mix Vision Transformer (MViT) encoder. The proposed methodology combines the powerful fusion characteristics of DIF-Net with the efficient segmentation characteristics of Segformer, allowing correctly capturing both local and global features, ultimately achieving superior segmentation results. The superiority of model performance is evaluated with multiple performance parameters for segmentation like mean Intersection Over Union (IoU), Pixel Accuracy, and dice coefficient, and for fusion it is PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index), MI (Mutual Information), UIQI (Universal Image Quality Index) and Edge Preservation. Experimental results show that the proposed method achieves better segmentation and fusion quality than existing models, enhancing the accurate diagnosis.<\/jats:p>","DOI":"10.1371\/journal.pone.0355109","type":"journal-article","created":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T17:28:25Z","timestamp":1787160505000},"page":"e0355109","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["An segformer-mix vision transformer-based GIST segmentation in fused CT-PET images"],"prefix":"10.1371","volume":"21","author":[{"given":"Venu","family":"Allapakam","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4203-5554","authenticated-orcid":true,"given":"Yepuganti","family":"Karuna","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,8,19]]},"reference":[{"key":"pone.0355109.ref001","doi-asserted-by":"crossref","DOI":"10.1053\/j.semnuclmed.2024.04.001","article-title":"The Role of 18-Fluorodeoxyglucose Positron Emission Tomography\/Computed Tomography (18F-FDG PET\/CT) in the Diagnosis and Evaluation of Spondylodiscitis","volume-title":"Seminars in Nuclear Medicine","author":"D Paez","year":"2024"},{"issue":"1","key":"pone.0355109.ref002","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/s10120-015-0526-8","article-title":"The standard diagnosis, treatment, and follow-up of gastrointestinal stromal tumors based on guidelines","volume":"19","author":"T Nishida","year":"2016","journal-title":"Gastric Cancer"},{"key":"pone.0355109.ref003","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1016\/j.canep.2015.10.031","article-title":"Global epidemiology of gastrointestinal stromal tumours (GIST): A systematic review of population-based cohort studies","volume":"40","author":"K S\u00f8reide","year":"2016","journal-title":"Cancer Epidemiol"},{"key":"pone.0355109.ref004","doi-asserted-by":"crossref","first-page":"853","DOI":"10.6004\/jnccn.2014.0080","article-title":"Gastrointestinal stromal Tumours, version 2.2014: Featured updates to the NCCN guidelines","volume":"12","author":"M Mehren","year":"2014","journal-title":"Journal of The National Comprehensive Cancer Network"},{"issue":"3","key":"pone.0355109.ref005","article-title":"Clinical significance of positron emission tomography-computed tomography in the classification of thymic tumors","volume":"40","author":"J Gao","year":"2025","journal-title":"Interdiscip Cardiovasc Thorac Surg"},{"key":"pone.0355109.ref006","volume-title":"DeVita, Hellman, and Rosenbergs Cancer: Principles & Practice of Oncology","author":"VT DeVita","year":"2008","edition":"10"},{"issue":"10","key":"pone.0355109.ref007","doi-asserted-by":"crossref","first-page":"1635","DOI":"10.1007\/s11548-019-01993-3","article-title":"Computer-aided diagnosis of gastrointestinal stromal tumors: a radiomics method on endoscopic ultrasound image","volume":"14","author":"X Li","year":"2019","journal-title":"Int J Comput Assist Radiol Surg"},{"issue":"1","key":"pone.0355109.ref008","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1186\/s13550-023-01059-1","article-title":"Performance and application of the total-body PET\/CT scanner: a literature review","volume":"14","author":"Y Sun","year":"2024","journal-title":"EJNMMI Res"},{"issue":"13","key":"pone.0355109.ref009","doi-asserted-by":"crossref","first-page":"4083","DOI":"10.1007\/s00259-024-06819-2","article-title":"Repeated dynamic 18F.FDG PET\/CT imaging using a high-sensitivity PET\/CT scanner for assessing non-small cell lung cancer patients undergoing induction immuno-chemotherapy followed by hypo-fractionated chemoradiotherapy and consolidative immunotherapy: report from a prospective observational study (GASTO-1067)","volume":"51","author":"D Wang","year":"2024","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"pone.0355109.ref010","doi-asserted-by":"crossref","first-page":"108827","DOI":"10.1016\/j.compbiomed.2024.108827","article-title":"Challenges and limitations in applying radiomics to PET imaging: Possible opportunities and avenues for research","volume":"179","author":"A Stefano","year":"2024","journal-title":"Comput Biol Med"},{"key":"pone.0355109.ref011","doi-asserted-by":"crossref","first-page":"107939","DOI":"10.1016\/j.cmpb.2023.107939","article-title":"Deep learning-based outcome prediction using PET\/CT and automatically predicted probability maps of primary tumor in patients with oropharyngeal cancer","volume":"244","author":"A De Biase","year":"2024","journal-title":"Comput Methods Programs Biomed"},{"key":"pone.0355109.ref012","doi-asserted-by":"crossref","first-page":"110852","DOI":"10.1016\/j.radonc.2025.110852","article-title":"PET and CT based DenseNet outperforms advanced deep learning models for outcome prediction of oropharyngeal cancer","volume":"207","author":"B Ma","year":"2025","journal-title":"Radiother Oncol"},{"key":"pone.0355109.ref013","doi-asserted-by":"crossref","first-page":"103958","DOI":"10.1016\/j.micpro.2021.103958","article-title":"An efficient image segmentation and classification of lung lesions in pet and CT image fusion using DTWT incorporated SVM","volume":"82","author":"R Mohana Priya","year":"2021","journal-title":"Microprocessors and Microsystems"},{"issue":"8","key":"pone.0355109.ref014","doi-asserted-by":"crossref","first-page":"549","DOI":"10.1049\/cvi2.12051","article-title":"Enhanced three\u2010dimensional U\u2010Net with graph\u2010based refining for segmentation of gastrointestinal stromal tumours","volume":"15","author":"Q Wang","year":"2021","journal-title":"IET Computer Vision"},{"key":"pone.0355109.ref015","doi-asserted-by":"crossref","first-page":"140","DOI":"10.1016\/j.procs.2017.11.221","article-title":"Lung tumor segmentation algorithm","volume":"120","author":"S Uzelaltinbulat","year":"2017","journal-title":"Procedia Computer Science"},{"key":"pone.0355109.ref016","doi-asserted-by":"crossref","first-page":"5805","DOI":"10.1109\/ACCESS.2019.2963472","article-title":"Multi-Task Refined Boundary-Supervision U-Net (MRBSU-Net) for Gastrointestinal Stromal Tumor Segmentation in Endoscopic Ultrasound (EUS) Images","volume":"8","author":"X Li","year":"2020","journal-title":"IEEE Access"},{"issue":"16","key":"pone.0355109.ref017","doi-asserted-by":"crossref","first-page":"3589","DOI":"10.3390\/jcm10163589","article-title":"Automatic Segmentation of Pancreatic Tumors Using Deep Learning on a Video Image of Contrast-Enhanced Endoscopic Ultrasound","volume":"10","author":"Y Iwasa","year":"2021","journal-title":"J Clin Med"},{"key":"pone.0355109.ref018","unstructured":"Zhang Y, et al. Mutual consistency dual-stream tumour segmentation network based on semi-supervised training. 2025."},{"issue":"1","key":"pone.0355109.ref019","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1186\/s12880-025-01587-3","article-title":"Comparative analysis of intestinal tumor segmentation in PET CT scans using organ based and whole body deep learning","volume":"25","author":"M Torkaman","year":"2025","journal-title":"BMC Med Imaging"},{"key":"pone.0355109.ref020","doi-asserted-by":"crossref","unstructured":"Torkaman M, Jemaa S, Fredrickson J, Coimbra AF, Carano RAD. Gastro-intestinal lesion segmentation using deep learning: organ-based versus whole-body training. In: Medical Imaging 2024: Clinical and Biomedical Imaging, 2024. 30. https:\/\/doi.org\/10.1117\/12.3006489","DOI":"10.1117\/12.3006489"},{"issue":"36","key":"pone.0355109.ref021","first-page":"111137","article-title":"Application of deep learning-based convolutional neural networks in gastrointestinal disease endoscopic examination","volume":"31","author":"Y-Y Wang","year":"2025","journal-title":"World J Gastroenterol"},{"key":"pone.0355109.ref022","doi-asserted-by":"crossref","first-page":"107326","DOI":"10.1016\/j.compbiomed.2023.107326","article-title":"BiFTransNet: A unified and simultaneous segmentation network for gastrointestinal images of CT & MRI","volume":"165","author":"X Jiang","year":"2023","journal-title":"Comput Biol Med"},{"issue":"3","key":"pone.0355109.ref023","doi-asserted-by":"crossref","first-page":"309","DOI":"10.3390\/bioengineering12030309","article-title":"Encoder-Decoder Variant Analysis for Semantic Segmentation of Gastrointestinal Tract Using UW-Madison Dataset","volume":"12","author":"N Sharma","year":"2025","journal-title":"Bioengineering (Basel)"},{"issue":"1","key":"pone.0355109.ref024","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1038\/s41746-025-02287-6","article-title":"Geometric multi-instance learning for weakly supervised gastric cancer segmentation","volume":"9","author":"C Huang","year":"2026","journal-title":"NPJ Digit Med"},{"key":"pone.0355109.ref025","unstructured":"Gupta P, et al. Performance of a deep learning-based segmentation model for pancreatic tumors on public endoscopic ultrasound datasets. arXiv preprint. 2026. https:\/\/doi.org\/arXiv:2601.05937"},{"key":"pone.0355109.ref026","doi-asserted-by":"crossref","first-page":"108377","DOI":"10.1016\/j.bspc.2025.108377","article-title":"Optimal feature selection for medical image fusion using deep learning with transformer","volume":"111","author":"DM Pathak","year":"2026","journal-title":"Biomedical Signal Processing and Control"},{"issue":"10","key":"pone.0355109.ref027","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0309651","article-title":"An ensemble deep learning model for medical image fusion with Siamese neural networks and VGG-19","volume":"19","author":"V Allapakam","year":"2024","journal-title":"PLoS One"},{"key":"pone.0355109.ref028","unstructured":"Khan AR, Khan A. MaxViT-UNet: Multi-axis attention for medical image segmentation. arXiv preprint. 2023. https:\/\/doi.org\/arXiv:2305.08396"},{"key":"pone.0355109.ref029","doi-asserted-by":"crossref","unstructured":"Li Y, Wu C-Y, Fan H, Mangalam K, Xiong B, Malik J, et al. MViTv2: Improved Multiscale Vision Transformers for Classification and Detection. In: 2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022. 4794\u2013804. https:\/\/doi.org\/10.1109\/cvpr52688.2022.00476","DOI":"10.1109\/CVPR52688.2022.00476"},{"issue":"1","key":"pone.0355109.ref030","doi-asserted-by":"crossref","first-page":"20098","DOI":"10.1038\/s41598-023-46580-4","article-title":"MANet: a multi-attention network for automatic liver tumor segmentation in computed tomography (CT) imaging","volume":"13","author":"K Hettihewa","year":"2023","journal-title":"Sci Rep"},{"key":"pone.0355109.ref031","doi-asserted-by":"crossref","first-page":"5805","DOI":"10.1109\/ACCESS.2019.2963472","article-title":"Multi-Task Refined Boundary-Supervision U-Net (MRBSU-Net) for Gastrointestinal Stromal Tumor Segmentation in Endoscopic Ultrasound (EUS) Images","volume":"8","author":"X Li","year":"2020","journal-title":"IEEE Access"},{"issue":"9","key":"pone.0355109.ref032","doi-asserted-by":"crossref","first-page":"1661","DOI":"10.1002\/jum.16489","article-title":"Deep Learning-Based Segmentation and Risk Stratification for Gastrointestinal Stromal Tumors in Transabdominal Ultrasound Imaging","volume":"43","author":"M Zhuo","year":"2024","journal-title":"J Ultrasound Med"},{"issue":"6","key":"pone.0355109.ref033","first-page":"1","article-title":"Analyzing the evaluation metrics of detecting gastrointestinal tumor using segmentation techniques in endoscopic images","volume":"18","author":"K Pooja","year":"2024","journal-title":"Onkologia i Radioterapia"},{"key":"pone.0355109.ref034","doi-asserted-by":"crossref","first-page":"109617","DOI":"10.1016\/j.compbiomed.2024.109617","article-title":"FocusU2Net: Pioneering dual attention with gated U-Net for colonoscopic polyp segmentation","volume":"186","author":"TB Ovi","year":"2025","journal-title":"Comput Biol Med"},{"key":"pone.0355109.ref035","doi-asserted-by":"crossref","first-page":"1749675","DOI":"10.3389\/fonc.2026.1749675","article-title":"A multi-task and explainable swin transformer framework for cross-scale computational pathology in gastrointestinal cancer","volume":"16","author":"Q-C Feng","year":"2026","journal-title":"Front Oncol"},{"issue":"2","key":"pone.0355109.ref036","doi-asserted-by":"crossref","first-page":"910","DOI":"10.1007\/s00259-025-07450-5","article-title":"Fully automated 3D multi-modal deep learning model for preoperative T-stage prediction of colorectal cancer using 18F-FDG PET\/CT","volume":"53","author":"M Zhang","year":"2026","journal-title":"Eur J Nucl Med Mol Imaging"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0355109","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T17:28:30Z","timestamp":1787160510000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0355109"}},"subtitle":[],"editor":[{"given":"Yufeng","family":"Zhou","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,8,19]]},"references-count":36,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,8,19]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0355109","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,19]]}}}