Ir al contenido

Documat

Exploring the Limits of Foundation Models in Medical Image Segmentation: A Case Study With SAM and Genetic Algorithms

  • Juan [1] ; Nuria Lozano-García [2] ; Emilio Delgado [2] ; Álvaro Rubio Largo [2] ; Roberto Rodríguez-Echeverría [2] Árbol académico
    1. [1] Universidade de Santiago de Compostela

      Universidade de Santiago de Compostela

      Santiago de Compostela, España

    2. [2] Universidad de Extremadura

      Universidad de Extremadura

      Badajoz, España

  • Localización: IJIMAI, ISSN-e 1989-1660, Vol. 9, Nº. 7, 2026, págs. 16-29
  • Idioma: inglés
  • DOI: 10.9781/ijimai.2026.2223
  • Enlaces
  • Resumen

    • This paper investigates the limits of foundation models in medical image segmentation, mainly focusing on SAM by Meta. While previous research demonstrated SAM’s potential for cost-efficient segmentation, this study explores its performance enhancement through integration with prompt enhancement optimization and genetic algorithms, aiming to minimize user input further. As a proof of concept, we apply this novel approach to lung segmentation tasks using public axial lung CT scans, frontal chest X-ray datasets, and spleen MRIs. Our findings reveal that the genetic algorithm optimization significantly improves SAM’s segmentation accuracy, bringing it closer to the state-of-the-art performance achieved by specifically trained models. In particular, when compared with our previous approach, this technique reaches a 94.85 % Jaccard Index (+3.77 delta) and a 97.17 % Dice Score (+2.50 delta) for lung CT scans, a 93.39 % Jaccard Index (+5.95 delta) and a 96.57 %Dice Score (+3.38 delta) for chest X-rays, and a 91.00 % Jaccard Index (+6.51 delta) and a 95.07 % Dice Score (+4.12 delta) for spleen MRIs. Notably, this improvement is achieved without retraining or modifying SAM’s architecture. However, our analysis also identifies an inherent limitation in this optimization approach, revealing a performance ceiling that cannot be surpassed despite further genetic algorithm iterations. The implications of these findings emphasize the potential of combining foundation models with non-intrusive optimization techniques for cost-effective and accessible medical image segmentation. While dataset-related limitations may affect generalizability, validating the approach across broader clinical scenarios remains essential. Future work should explore applications to additional organs, diverse datasets, and the integration of expert-in-the-loop strategies to enhance clinical utility.

  • Referencias bibliográficas
    • [1] A. Kirillov, K. He, R. Girshick, C. Rother, P. Dollar, “Panoptic Segmentation,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern...
    • [2] G. Litjens, T. Kooi, B. E. Bejnordi, A. A. A. Setio, F. Ciompi, M. Ghafoorian, J. A. Van Der Laak, B. Van Ginneken, C. I. Sánchez, “A...
    • [3] S. Minaee, Y. Boykov, F. Porikli, A. Plaza, N. Kehtarnavaz, D. Terzopoulos, “Image Segmentation Using Deep Learning: A Survey,” IEEE Transactions...
    • [4] A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y. Lo, P. Dollár, R. Girshick,...
    • [5] Y. Huang, X. Yang, L. Liu, H. Zhou, A. Chang, X. Zhou, R. Chen, J. Yu, J. Chen, C. Chen, S. Liu, H. Chi, X. Hu, K. Yue, L. Li, V. Grau,...
    • [6] J. D. Gutiérrez, R. Rodriguez-Echeverria, E. Delgado, M. Á. S. Rodrigo, F. Sanchez-Figueroa, “No More Training: SAM’s Zero-Shot Transfer...
    • [7] S. N. Kumar, A. L. Fred, P. S. Varghese, “An Overview of Segmentation Algorithms for the Analysis of Anomalies on Medical Images,” Journal...
    • [8] J. H. Holland, Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial...
    • [9] K. K. Verma, B. M. Singh, “Deep Multi-Model Fusion for Human Activity Recognition Using Evolutionary Algorithms,” International Journal...
    • [10] T. Hui-Ye Chiu, C. Wu, C.-H. Chen, “A Generalized Wine Quality Prediction Framework by Evolutionary Algorithms,” International Journal...
    • [11] L. Ali, F. Alnajjar, M. Swavaf, O. Elharrouss, A. Abd-alrazaq, R. Damseh, “Evaluating segment anything model (SAM) on MRI scans of brain...
    • [12] T. Cai, H. Yan, K. Ding, Y. Zhang, Y. Zhou, “WSPolyp-SAM: Weakly Supervised and Self-Guided Fine-Tuning of SAM for Colonoscopy Polyp...
    • [13] C. Chen, J. Miao, D. Wu, A. Zhong, Z. Yan, S. Kim, J. Hu, Z. Liu, L. Sun, X. Li, T. Liu, P.- A. Heng, Q. Li, “MA-SAM: Modality-agnostic...
    • [14] G. Dong, Z. Wang, Y. Chen, Y. Sun, H. Song, L. Liu, H. Cui, “An efficient segment anything model for the segmentation of medical images,”...
    • [15] S. Gong, Y. Zhong, W. Ma, J. Li, Z. Wang, J. Zhang, P.-A. Heng, Q. Dou, “3DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for...
    • [16] Y. Gu, Q. Wu, H. Tang, X. Mai, H. Shu, B. Li, Y. Chen, “LeSAM: Adapt Segment Anything Model for medical lesion segmentation,” IEEE Journal...
    • [17] X. Liu, Y. Zhao, S. Wang, J. Wei, “G-SAM: GMM-based segment anything model for medical image classification and segmentation,” Cluster...
    • [18] Z. Ren, Y. Zhang, S. Wang, “Large Foundation Model for Cancer Segmentation,” Technology in Cancer Research & Treatment, vol. 23,...
    • [19] P. Shi, J. Qiu, S. M. D. Abaxi, H. Wei, F. P.- W. Lo, W. Yuan, “Generalist Vision Foundation Models for Medical Imaging: A Case Study...
    • [20] N. Ndipenoch, A. Miron, Y. Li, “Performance Evaluation of Retinal OCT Fluid Segmentation, Detection, and Generalization Over Variations...
    • [21] D. He, Z. Ma, C. Li, Y. Li, “Dual-Branch Fully Convolutional Segment Anything Model for Lesion Segmentation in Endoscopic Images,” IEEE...
    • [22] Z. Morton Colbert, D. Arrington, M. Foote, J. Gårding, D. Fay, M. Huo, M. Pinkham, P. Ramachandran, “Repurposing traditional U-Net predictions...
    • [23] C. Wang, H. Chen, X. Zhou, M. Wang, Q. Zhang, “SAM-IE: SAM-based image enhancement for facilitating medical image diagnosis with segmentation...
    • [24] M. A. JiMing, D. HongYu, W. YuFan, W. LiNa, “Medical image segmentation based on simulated annealing and opposition-based learning island...
    • [25] K. M. Hosny, A. M. Khalid, H. M. Hamza, Mirjalili, “Multilevel segmentation of 2D and volumetric medical images using hybrid Coronavirus...
    • [26] D. R. Reis, B. C. Santos, L. Bleicher, L. E. Zárate, C. N. Nobre, “Prediction of enzymatic function with high efficiency and a reduced...
    • [27] M. Chen, Z. Sun, F. Su, Y. Chen, D. Bu, Y. Lyu, “An Auxiliary Diagnostic System for Parkinson’s Disease Based on Wearable Sensors and...
    • [28] M. Ghosh, S. Adhikary, K. K. Ghosh, A. Sardar, Begum, R. Sarkar, “Genetic algorithm based cancerous gene identification from microarray...
    • [29] M. Sale, E. A. Sherer, “A genetic algorithm based global search strategy for population pharmacokinetic/pharmacodynamic model selection,”...
    • [30] M. Jun, G. Cheng, W. Yixin, A. Xingle, G. Jiantao, Y. Ziqi, Z. Minqing, L. Xin, D. Xueyuan, C. Shucheng, W. Hao, M. Sen, Y. Xiaoyu, N....
    • [31] M. Owais, N. R. Baek, K. R. Park, “DMDF-Net: Dual multiscale dilated fusion network for accurate segmentation of lesions related to COVID-19...
    • [32] S. Jaeger, S. Candemir, S. Antani, Y.-X. J. Wáng, P.-X. Lu, G. Thoma, “Two public chest X-ray datasets for computer-aided screening of...
    • [33] B. Chen, Y. Liu, Z. Zhang, G. Lu, A. W. K. Kong, “TransAttUnet: MultiLevel Attention-Guided U-Net With Transformer for Medical Image...
    • [34] Y. Ji, H. Bai, C. GE, J. Yang, Y. Zhu, R. Zhang, Z. Li, L. Zhanng, W. Ma, X. Wan, P. Luo, “AMOS: A large-scale abdominal multi-organ...
    • [35] D. Müller, I. Soto-Rey, F. Kramer, “Towards a guideline for evaluation metrics in medical image segmentation,” BMC Research Notes, vol....
    • [36] F. Kofler, I. Ezhov, F. Isensee, F. Balsiger, C. Berger, M. Koerner, B. Demiray, J. Rackerseder, J. Paetzold, H. Li, S. Shit, R. McKinley,...
    • [37] J. Ma, Y. He, F. Li, L. Han, C. You, B. Wang, “Segment anything in medical images,” Nature Communications, vol. 15, p. 654, Jan. 2024,...
    • [38] J. Ma, Y. Wang, X. An, C. Ge, Z. Yu, J. Chen, Q. Zhu, G. Dong, J. He, Z. He, T. Cao, Y. Zhu, Z. Nie, X. Yang, “Toward data-efficient...
    • [39] S. P. Morozov, A. E. Andreychenko, I. A. Blokhin, P. B. Gelezhe, A. P. Gonchar, A. E. Nikolaev, N. A. Pavlov, V. Y. Chernina, V. A. Gombolevskiy,...
    • [40] N. Codella, V. Rotemberg, P. Tschandl, M. E. Celebi, S. Dusza, D. Gutman, B. Helba, A. Kalloo, K. Liopyris, M. Marchetti, H. Kittler,...
    • [41] J. Shiraishi, S. Katsuragawa, J. Ikezoe, T. Matsumoto, T. Kobayashi, K.- i. Komatsu, M. Matsui, H. Fujita, Y. Kodera, K. Doi, “Development...
    • [42] Y.-B. Tang, Y.-X. Tang, J. Xiao, R. M. Summers, “XLSor: A Robust and Accurate Lung Segmentor on Chest X-Rays Using Criss-Cross Attention...
    • [43] X. He, S. Wang, X. Chu, S. Shi, J. Tang, X. Liu, C. Yan, J. Zhang, G. Ding, “Automated Model Design and Benchmarking of Deep Learning...
    • [44] J. C. Caicedo, A. Goodman, K. W. Karhohs, B. A. Cimini, J. Ackerman, M. Haghighi, C. Heng, T. Becker, M. Doan, C. McQuin, M. Rohban,...
    • [45] P. Malík, K. Knapová, Š. Krištofík, “Instance Segmentation Model Created from Three Semantic Segmentations of Mask, Boundary and Centroid...
    • [46] Q. Hu, L. F. De F. Souza, G. B. Holanda, S. S. Alves, F. H. Dos S. Silva, T. Han, P. P. Rebouças Filho, “An effective approach for CT...
    • [47] A. Khanna, N. D. Londhe, S. Gupta, A. Semwal, “A deep Residual U-Net convolutional neural network for automated lung segmentation in...
    • [48] R. D. Rudyanto et alii “Comparing algorithms for automated vessel segmentation in computed tomography scans of the lung: The VESSEL12...
    • [49] A. Depeursinge, A. Vargas, A. Platon, A. Geissbuhler, P.-A. Poletti, H. Müller, “Building a reference multimedia database for interstitial...
Fundación Dialnet

Mi Documat

Opciones de artículo

Opciones de compartir

Opciones de entorno