Citation :

Sumitha B S, Mohammed Tajuddin, Vikram Patil, Shweta R Poojary, Aarathi Santhosh, "An Agentic AI Model for MRI-Based Staging and Grading of Endometrial Cancer using MedSAM and ViT," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 2, pp. 125-150, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I2P109

Abstract

Endometrial Cancer (EC) is one of the most prevalent gynecological cancers globally. Accurate preoperative staging and the subsequent tumor grading are crucial for prognosis, surgical decisions, and treatment plans. Generally, for MRI evaluation, the radiologists depend totally on their own interpretation, which will be a mixture of subjectivity, a lengthy process, and inconsistency among different observers. The problems outlined above are addressed by creating a highly capable Artificial Intelligence (AI) framework that relies on multiple MRI modalities and clinical attributes for the precise identification of Deep Myometrial Invasion (DMI) and tumor grading. For accurate anatomical segmentation, the framework makes use of the Medical Segment Anything Model (MedSAM). After that, a Uterine Cavity Line Generation Algorithm (UCLGA) is employed to determine the depth of myometrial invasion. The Vision Transformer (ViT) model, fine-tuned via Low-Rank Adoption (LoRA), is used for feature representation and learning. Simultaneously, the agentic reasoning module successively enhances prediction through self-reflection and clinical knowledge. Furthermore, multiple smart agents are applied for segmentation visualization, report generation, compliance monitoring, and scheduling to provide a modular and interpretable system. The model is tested on the EC-MRI dataset, and its overall accuracy reached 96.82%, sensitivity 95.85%, precision 96.77%, F1 score 95.31%, and specificity 96.64%. The model surpassed the other current models in accuracy. Overall, the findings imply that this model is a clinically significant, elucidative, and efficient AI system that could support oncologists and radiologists in the preoperative EC evaluation.

Keywords

Agentic AI, Endometrial Cancer, LoRA, MRI, Myometrial Invasion, MedSAM, Tumor Grading, UCL, Vision Transformer.

References

[1] Junmei He et al., “Accuracy of Radiomics in the Diagnosis and Preoperative High-Risk Assessment of Endometrial Cancer: A Systematic Review and Meta-Analysis,” Frontiers in Oncology, vol. 14, pp. 1-11, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[2] Chenwen Sun et al., “Analysis of the Correlation between MRI Imaging Signs and Lymphovascular Space Invasion in Endometrial Cancer,” Current Medical Imaging, vol. 21, no. 1, pp. 1-10, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[3] Nabiha Midhat Ansari et al., “AI-Augmented Advances in the Diagnostic Approaches to Endometrial Cancer,” Cancers, vol. 17, no. 11, pp. 1-14, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Gengshen Bai et al., “Artificial Intelligence Radiomics in the Diagnosis, Treatment, and Prognosis of Gynecological Cancer a Literature Review,” Translational Cancers Research, vol. 14, no. 4, pp. 2508-2532, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[5] Xuji Jiang et al., “Enhancing Clinical Decision-Making in Endometrial Cancer through Deep Learning Technology: A Review of Current Research,” Digital Health, vol. 10, pp. 1-12, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[6] Longyun Wang et al., “Diagnosis Test Accuracy of Artificial Intelligence for Endometrial Cancer Systematic Review and Meta-Analysis,” Journal of Medical Internet Research, vol. 27, pp. 1-15, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[7]  István Madár et al., “Diagnostic Accuracy of Transvaginal Ultrasound and Magnetic Resonance Imaging for the Detection of Myometrial Infiltration in Endometrial Cancer: A Systematic Review and Meta-Analysis,” Cancers, vol. 16, no. 5, pp. 1-16, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[8] Jing Yang et al., “MRI-based Intra Tumoral and Peritumoral Radiomic for Assessing Deep Myometrial Invasions in Patient with Early-Stages Endometrioid Adenocarcinoma,” Frontier in Oncology, vol. 14, pp. 1-11, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Marco Gennarini et al., “Multi-Models Quantitative MRI of Uterine Cancer in Precision Medicine Era-A Narrative Review,” Insight into Imaging, vol. 16, no. 1, pp. 1-19, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[10] Vipul Bhardwaj et al., “Machine Learning for Endometrial Cancers Predictions and Prognostications,” Frontier in Oncology, vol. 12, pp. 1-16, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Diogo Miguel Machado Pereira et al., “The Updated 2023 Staging of Endometrial Cancers: Tips for MRI Interpretations,” Abdominal Radiology, vol. 51, no. 1, pp. 261-273, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[12] Octavia Petrila et al., “The Applicability of Artificial Intelligences in Predicting the Depths of Myometrial Invasions on MRI Studies-A Systematic Review,” Diagnostics, vol. 13, no. 15, pp. 1-10, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Arnaldo Stanzione et al., “Training Radiology Resident to Evaluate Deep Myometrial Invasions in Endometrial Cancers Patient on MRI: A Learning Curve Study,” European Journals of Radiology, vol. 177, pp. 1-8, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[14] Lise Lecointre et al., “Artificial Intelligence-Enhanced Magnetic Resonance Imaging-based Pre-Operative Staging in Patients with Endometrial Cancer,” International Journal of Gynecological Cancers, vol. 35, no. 1, pp. 1-7, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[15] Jawaria Changhez et al., “Evaluating the Efficacy and Accuracy of AI Assisted Diagnostic Techniques in Endometrial Carcinoma: A Systematic Review,” Cureus, vol. 16, no. 5, pp. 1-11, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Liu Xiong et al., “A Computer-Aided Determining Method for the Myometrial Infiltrations Depths of Early Endometrial Cancers on MRI Image,” BioMedical Engineering OnLine, vol. 22, no. 1, pp. 1-17, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Wei Mao et al., “A Deep Learning-based Automatics Staging Methods for Early Endometrial Cancers on MRI Image,” Frontier in Physiology, vol. 13, pp. 1-12, 2022.

[CrossRef] [Google Scholar] [Publisher Link]

[18] Jing Wang et al., “A Predictions Model based on Deep Learning and Radiomic Feature of DWI for the Assessments of Microsatellites Instability in Endometrial Cancers,” Cancer Medicines, vol. 13, no. 16, pp. 1-13, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[19] Xinyu Qi, “Artificial Intelligences-Assisted Magnetic Resonances Imaging Technology in the Differential Diagnoses and Prognoses Predictions of Endometrial Cancers,” Scientific Report, vol. 14, no. 1, pp. 1-12, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[20] Jin Yang et al., “Combined Deep-Learning MRI-based Radiomics Model for Preoperative Risks Classifications of Endometrial Endometrioid Adenocarcinoma,” Frontier in Oncology, vol. 13, pp. 1-10, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[21] Xueliang Zhu et al., “Detections of Deep Myometrial Invasions in Endometrial Cancers MR Imaging based on Multi-Features Fusions and Probabilistic Support Vector Machines Ensemble,” Computer in Biology and Medicines, vol. 134, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[22] Qiu Bi et al., “Different Multiparametric MRI-based Radiomic Model for Differentiating Stages IA Endometrial Cancers from Benign Endometrial Lesion: A Multicenter Study,” Frontier in Oncology, vol. 12, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[23] Tsukasa Saida et al., “Differentiations of Carcinosarcoma from Endometrial Carcinoma on Magnetic Resonances Imaging using Deep Learning,” Polish Journal of Radiology, vol. 87, no. 1, pp. 521-529, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[24] Sam Raja et al., “Enhancing Preoperative Assessments of Endometrial Cancers: The Roles of Diffusions-Weighted Magnetic Resonances Imaging in Evaluating Myometrial Invasions,” Cureus, vol. 16, no. 6, pp. 1-16, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[25] Jingxiong Tao et al., “Evaluations and Monitoring of Endometrial Cancers based on Magnetic Resonances Imaging Feature of Deep Learning,” Contrast Media and Molecular Imaging, vol. 2022, no. 1, pp. 1-9, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[26] Ziyu Zheng et al., “Multimodal MRI Images Fusions for Early Automatic Staging of Endometrial Cancers,” Sensors, vol. 25, no. 9, pp. 1-16, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[27] Xuxu Meng et al., “Multiparametric Magnetic Resonances Imaging-based Assessments of the Effects of Adenomyosis on Determining the Depths of Myometrial Invasions in Endometrial Cancers,” Quantitative Imaging in Medicines and Surgery, vol. 14, no. 5, pp. 3717-3730, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[28] Huan Meng et al., “Predicting Risks Stratifications in Early-Stages Endometrial Carcinoma: Significances of Multiparametric MRI Radiomic Models,” Journal of Imaging Informatics in Medicines, vol. 37, no. 1, pp. 81-91, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[29] Dong-Hoon Jang et al., “Predictions of Final Pathology Depending on Preoperative Myometrial Invasions and Grades Assessments in Low-Risks Endometrial Cancers Patient: A Korean Gynecologic Oncology Group Ancillary Study,” PloS One, vol. 19, no. 6, pp. 1-17, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[30] Aiko Urushibara et al., “The Efficacy of Deep Learning Model in the Diagnoses of Endometrial Cancers using MRI: A Comparison with Radiologist,” BMC Medical Imaging, vol. 22, no. 1, pp. 1-14, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[31] Yuan Chen et al., “Preoperative Discrimination of Absence or Presence of Myometrial Invasion in Endometrial Cancer with an MRI-based Multimodal Deep Learning Radiomics Model,” Abdominal Radiology, vol. 50, no. 7, pp. 3335-3346, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[32] Ran Guo et al., “Preoperative Prediction of Aggressive Endometrial Cancer using Multiparametric MRI-based Deep Transfer Learning Models,” Frontiers in Oncology, vol. 15, pp. 1-12, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[33] S. Manjunatha, and Malini M. Patil, “Interpolation Technique in Images Resampling,” International Journal of Engineering and Technology, vol. 7, no. 3.34, pp. 567-570, 2018.
[Google Scholar]

[34] Iza Sazanita Isa et al., “Evaluating Denoising Performance of Fundamental Filter for T2-Weighted MRI Image,” Procedia Computer Sciences, vol. 60, pp. 760-768, 2015.
[CrossRef] [Google Scholar] [Publisher Link]

[35] Nema Salem, Hebatullah Malik, and Asmaa Shams, “Medical Images Enhancements based on Histograms Algorithm,” Procedia Computer Sciences, vol. 163, pp. 300-311, 2019.
[CrossRef] [Google Scholar] [Publisher Link]

[36] Jun Ma et al., “Segment Anything in Medical Image,” Nature Communication, vol. 15, no. 1, pp. 1-9, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[37] Christian Chang et al., “Segment Anything Models (SAM) and Medical SAM (MedSAM) for Lumbar Spines MRI,” Sensors, vol. 25, no. 12, pp. 1-17, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[38] Zhenchen Hong et al., “Lightweight Low-Ranks Adaptations Vision Transformers Frameworks for Cervical Cancers Detections and Cervix Types Classifications,” Bioengineering, vol. 11, no. 5, pp. 1-23, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[39] Chenqi Kong, Haoliang Li, and Shiqi Wang, “Enhancing General Faces Forgery Detections via Vision Transformers with Low-Ranks Adaptations,” 2023 IEEE 6^th International Conference on Multimedia Information Processing and Retrieval (MIPR), Singapore, pp. 102-107, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[40] Dyke Ferber et al., “Developments and Validations of an Autonomous Artificial Intelligences Agents for Clinical Decisions-Making in Oncology,” Nature Cancer, vol. 6, no. 8, pp. 1337-1349, 2025.
       [CrossRef] [Google Scholar] [Publisher Link]