Academic Journal
Multi-window scanning method for region-of-interest selection in pancreatic endoscopic ultrasound images.
| Τίτλος: | Multi-window scanning method for region-of-interest selection in pancreatic endoscopic ultrasound images. |
|---|---|
| Συγγραφείς: | Filist S; South-West State University, 50 Let Oktyabrya St. 94, Kursk, 305040, Russia., Al-Kasasbeh RT; Department of Mechatronics Engineering, The University of Jordan, Amman, 11942, Jordan. r.al-kasasbeh@ju.edu.jo., Gevorkyan T; Blokhin National Medical Research Center of Oncology of the Russian Ministry of Health, Moscow, Russia., Kondrashov DS; South-West State University, 50 Let Oktyabrya St. 94, Kursk, 305040, Russia., Shatalova O; South-West State University, 50 Let Oktyabrya St. 94, Kursk, 305040, Russia., Korenevskiy NA; South-West State University, 50 Let Oktyabrya St. 94, Kursk, 305040, Russia., Telfah A; Fachhochschule Dortmund University of Applied Sciences and Arts, 44139, Dortmund, Germany.; Cell Therapy Center, The University of Jordan, Amman, 11942, Jordan. |
| Πηγή: | Journal of ultrasound [J Ultrasound] 2026 Sep; Vol. 29 (3), pp. 581-594. Date of Electronic Publication: 2026 Jul 22. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Springer Country of Publication: Italy NLM ID: 101315005 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1876-7931 (Electronic) Linking ISSN: 18767931 NLM ISO Abbreviation: J Ultrasound Subsets: MEDLINE |
| Imprint Name(s): | Publication: 2013: [Milan] : Springer Original Publication: Milano : Elsevier Masson |
| Ιατρικοί όροι (MeSH): | Endosonography*/methods , Pancreas*/diagnostic imaging , Image Processing, Computer-Assisted*/methods , Image Interpretation, Computer-Assisted*/methods , Pancreatic Diseases*/diagnostic imaging, Humans ; Neural Networks, Computer ; Sensitivity and Specificity |
| Περίληψη: | This study presents a method for semantic segmentation of pancreatic endoscopic ultrasound (EUS) images. We developed a hybrid method to automatically select relevant frames from EUS video sequences, employing two nested dynamic windows: a smaller window scans local regions of each frame, while a larger window defines the overall scanning area, capturing spatial context. Texture features, including mode, mean, and standard deviation, are extracted from both windows to construct a feature vector for a fully connected neural network classifier (NNC), which predicts the class of each pixel. Following pixel-wise classification, a heatmap is generated for each frame to highlight regions of interest (ROI), allowing specialists to identify echotextural features. The method uses hierarchical image decomposition to improve ROI identification. Custom software was implemented to enable interactive segmentation, division into local windows, ROI classification, and structured database formation. Based on expert endoscopist evaluation, the optimal window size was determined to be 32 × 32 pixels. A curated dataset of local windows containing normal and pathological pancreatic echotextures was compiled from the selected frames. Experimental evaluation on 114 test frames demonstrated that the proposed method achieves an overall accuracy of 95.6% (PPV 96.3%; sensitivity 97.5%; specificity 91.2%), distinguishing relevant frames for subsequent semantic segmentation. This framework supports clinical assessment and automated ROI extraction in pancreatic EUS imaging. (© 2026. Società Italiana di Ultrasonologia in Medicina e Biologia (SIUMB).) |
| Competing Interests: | Declarations. Conflict of Interest: The authors declare no apparent or potential conflicts of interest related to the publication of this article. |
| References: | Halbrook CJ et al (2023) Pancreatic cancer: advances and challenges. Cell 186(8):1729–1754. (PMID: 10.1016/j.cell.2023.02.0143705907010182830) Mizrahi JD et al (2020) Pancreatic cancer. Lancet 395(10242):2008–2020. (PMID: 10.1016/S0140-6736(20)30974-032593337) Bilreiro C et al (2024) Imaging of pancreatic ductal adenocarcinoma – an update for all stages of patient management. Eur J Radiol Open 12:100553. (PMID: 10.1016/j.ejro.2024.1005533835738510864763) Overbeek KA, Cahen DL, Bruno MJ (2024) The role of endoscopic ultrasound in the detection of pancreatic lesions in high-risk individuals. Fam Cancer 23(3):279–293. (PMID: 10.1007/s10689-024-00380-53857339911255057) Duron L et al (2021) Can we use radiomics in ultrasound imaging? Impact of preprocessing on feature repeatability. Diagn Interv Imaging 102(11):659–667. (PMID: 10.1016/j.diii.2021.10.00434690106) Elyan E et al (2022) Computer vision and machine learning for medical image analysis: recent advances, challenges, and way forward. Artif Intell Surg 2(1):24–45. Perona P, Malik J (1990) Scale-space and edge detection using anisotropic diffusion. IEEE Trans Pattern Anal Mach Intell 12(7):629–639. (PMID: 10.1109/34.56205) Gonzalez RC (2009) Digital image processing. Pearson Education India. Seo K et al (2022) Semantic segmentation of pancreatic cancer in endoscopic ultrasound images using deep learning approach. Cancers 14:5111. (PMID: 10.3390/cancers14205111362918959600976) Moglia A et al (2025) Deep learning for pancreas segmentation on computed tomography: a systematic review. Artif Intell Rev 58:220. (PMID: 10.1007/s10462-024-11050-4) Abdar M et al (2021) Uncertainty quantification in skin cancer classification using three-way decision-based Bayesian deep learning. Comput Biol Med 135:104418. (PMID: 10.1016/j.compbiomed.2021.10441834052016) Hossain MS et al (2023) Region of interest (ROI) selection using vision transformer for automatic analysis using whole slide images. Sci Rep 13(1):11314. (PMID: 10.1038/s41598-023-38109-63744318810344922) Samet H (1984) The quadtree and related hierarchical data structures. ACM Comput Surv 16(2):187–260. (PMID: 10.1145/356924.356930) Filist SA, Dabagov AR, Tomakova RA, Malyutina IA, Kondrashov DS (2019) Method of cascade segmentation of biomedical raster images. Bulletin of the South-West State University. Series: Management, computing, informatics. Medical instrument making 9(1):49–61. Haralick RM, Shanmugam K, Dinstein I (1973) Textural features for image classification. IEEE Trans Syst Man Cybern SMC-3(6):610–621. (PMID: 10.1109/TSMC.1973.4309314) Varghese BA et al (2019) Texture analysis of imaging: what radiologists need to know. Am J Roentgenol 212(3):520–528. (PMID: 10.2214/AJR.18.20624) Pitas I (2000) Digital image processing algorithms and applications. John Wiley & Sons. Filist SA, Kondrashov DS, Sukhomlinov AY, Shulga LV, Al-Darraji CH, Belozerov VA (2023) Automated system for classification of ultrasound images of the pancreas based on the method of segmental spectral analysis. Model Optim Inf Technol 11(4):1–19. https://doi.org/10.26102/2310-6018/2023.40.1.021. (PMID: 10.26102/2310-6018/2023.40.1.021) Jain AK, Mao J, Mohiuddin KM (1996) Artificial neural networks: a tutorial. Comput 29(3):31–44. Cireşan DC et al (2013) Mitosis detection in breast cancer histology images with deep neural networks. International conference on medical image computing and computer-assisted intervention. Springer. Rakhshan V (2014) Image resolution in the digital era: notion and clinical implications. J Dent 15(4):153. Al-Kasasbeh RT et al (2024) Intelligent system for classifying of acoustic endoscopic images of the pancreas based on the analysis of walsh spectra of local windows. In: 2024 25th international Arab conference on information technology (ACIT). IEEE. Suarez-Alvarez MM et al (2012) Statistical approach to normalization of feature vectors and clustering of mixed datasets. Proc R Soc A Math Phys Eng Sci 468(2145):2630–2651. Filist SA, Kondrashov DS, Kuz’min AA, Sukhomlinov AY, Al’-Darradzhi C-K (2024) Classification of medical images based on the spectra of local windows. Biomed Eng 57(5):321–324. https://doi.org/10.1007/s10527-023-10324-5. (PMID: 10.1007/s10527-023-10324-5) Tereikovskyi I et al (2022) The method of semantic image segmentation using neural networks. Int J Image Graph Signal Process 14(6):1. Waoo AA, Soni BK (2021) Performance analysis of sigmoid and relu activation functions in deep neural network. Intelligent systems: proceedings of SCIS 2021. Springer, pp 39–52. (PMID: 10.1007/978-981-16-2248-9_5) Ghosh J, Gupta S (2023) ADAM optimizer and Categorical Crossentropy loss function-based CNN method for diagnosing colorectal Cancer. In: 2023 international conference on computational intelligence and sustainable engineering solutions (CISES). IEEE. |
| Contributed Indexing: | Keywords: Endoscopic ultrasound; Heatmap visualization; Nested dynamic windows; Neural network classifier; Pancreas; Semantic segmentation; Texture analysis |
| Entry Date(s): | Date Created: 20260722 Date Completed: 20260918 Latest Revision: 20260921 |
| Update Code: | 20260921 |
| PubMed Central ID: | PMC13589020 |
| DOI: | 10.1007/s40477-026-01175-3 |
| PMID: | 42484989 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1876-7931 |
|---|---|
| DOI: | 10.1007/s40477-026-01175-3 |