A Comparative Study in Surgical AI: Datasets, Foundation Models, and Barriers to Med-AGI

• URL: http://arxiv.org/abs/2603.27341v1
• Date: Sat, 28 Mar 2026 17:18:40 GMT
• Title: A Comparative Study in Surgical AI: Datasets, Foundation Models, and Barriers to Med-AGI
• Abstract summary: We show that even with multi-billion parameter models and extensive training, current Vision Language Models fall short in the seemingly simple task of tool detection in neurosurgery.<n>We discuss the main contributors to these constraints and advance potential solutions.
• Score: 0.6291991901087711
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Recent Artificial Intelligence (AI) models have matched or exceeded human experts in several benchmarks of biomedical task performance, but have lagged behind on surgical image-analysis benchmarks. Since surgery requires integrating disparate tasks -- including multimodal data integration, human interaction, and physical effects -- generally-capable AI models could be particularly attractive as a collaborative tool if performance could be improved. On the one hand, the canonical approach of scaling architecture size and training data is attractive, especially since there are millions of hours of surgical video data generated per year. On the other hand, preparing surgical data for AI training requires significantly higher levels of professional expertise, and training on that data requires expensive computational resources. These trade-offs paint an uncertain picture of whether and to-what-extent modern AI could aid surgical practice. In this paper, we explore this question through a case study of surgical tool detection using state-of-the-art AI methods available in 2026. We demonstrate that even with multi-billion parameter models and extensive training, current Vision Language Models fall short in the seemingly simple task of tool detection in neurosurgery. Additionally, we show scaling experiments indicating that increasing model size and training time only leads to diminishing improvements in relevant performance metrics. Thus, our experiments suggest that current models could still face significant obstacles in surgical use cases. Moreover, some obstacles cannot be simply ``scaled away'' with additional compute and persist across diverse model architectures, raising the question of whether data and label availability are the only limiting factors. We discuss the main contributors to these constraints and advance potential solutions.

Related papers

• Data-Efficient Learning for Generalizable Surgical Video Understanding [0.0]
  This research aims to bridge gap between deep learning-based surgical video analysis in research and its real-world clinical environments.<n>I benchmarked state-of-the-art neural network architectures to identify the most effective designs for each task.<n>We developed semi-driven frameworks that improve model performance across tasks by leveraging large amounts of unlabeled surgical video.
  arXiv  Detail & Related papers  (2025-08-13T22:00:23Z)
• Exploring Patient Data Requirements in Training Effective AI Models for MRI-based Breast Cancer Classification [1.6904868487513736]
  We show that medical institutions do not need a decade's worth of MRI images to train an AI model.<n>We observe that for patient counts greater than 50, the number of patients in the training set has a negligible impact on the performance of models.
  arXiv  Detail & Related papers  (2025-02-22T04:04:52Z)
• Structured Model Pruning for Efficient Inference in Computational Pathology [2.9687381456164004]
  We develop a methodology for pruning the widely used U-Net-style architectures in biomedical imaging. We empirically demonstrate that pruning can compress models by at least 70% with a negligible drop in performance.
  arXiv  Detail & Related papers  (2024-04-12T22:05:01Z)
• ORacle: Large Vision-Language Models for Knowledge-Guided Holistic OR Domain Modeling [41.30327565949726]
  We introduce ORacle, an advanced vision-language model designed for holistic OR domain modeling. It incorporates multi-view and temporal capabilities and can leverage external knowledge during inference, enabling it to adapt to previously unseen surgical scenarios. In rigorous testing, in scene graph generation, and downstream tasks on the 4D-OR dataset, ORacle not only demonstrates state-of-the-art performance but does so requiring less data than existing models.
  arXiv  Detail & Related papers  (2024-04-10T14:24:10Z)
• Towards a clinically accessible radiology foundation model: open-access and lightweight, with automated evaluation [113.5002649181103]
  Training open-source small multimodal models (SMMs) to bridge competency gaps for unmet clinical needs in radiology. For training, we assemble a large dataset of over 697 thousand radiology image-text pairs. For evaluation, we propose CheXprompt, a GPT-4-based metric for factuality evaluation, and demonstrate its parity with expert evaluation. The inference of LlaVA-Rad is fast and can be performed on a single V100 GPU in private settings, offering a promising state-of-the-art tool for real-world clinical applications.
  arXiv  Detail & Related papers  (2024-03-12T18:12:02Z)
• Incomplete Multimodal Learning for Complex Brain Disorders Prediction [65.95783479249745]
  We propose a new incomplete multimodal data integration approach that employs transformers and generative adversarial networks. We apply our new method to predict cognitive degeneration and disease outcomes using the multimodal imaging genetic data from Alzheimer's Disease Neuroimaging Initiative cohort.
  arXiv  Detail & Related papers  (2023-05-25T16:29:16Z)
• Intuitive Surgical SurgToolLoc Challenge Results: 2022-2023 [55.40111320730479]
  We have challenged the surgical data science community to solve difficult machine learning problems in the context of advanced RA applications.<n>Here we document the results of these challenges, focusing on surgical tool localization (SurgToolLoc)<n>The publicly released dataset that accompanies these challenges is detailed in a separate paper arXiv:2501.09209.
  arXiv  Detail & Related papers  (2023-05-11T21:44:39Z)
• ProcTHOR: Large-Scale Embodied AI Using Procedural Generation [55.485985317538194]
  ProcTHOR is a framework for procedural generation of Embodied AI environments. We demonstrate state-of-the-art results across 6 embodied AI benchmarks for navigation, rearrangement, and arm manipulation.
  arXiv  Detail & Related papers  (2022-06-14T17:09:35Z)
• SyntheX: Scaling Up Learning-based X-ray Image Analysis Through In Silico Experiments [12.019996672009375]
  We show that creating realistic simulated images from human models is a viable alternative to large-scale in situ data collection. Because synthetic generation of training data from human-based models scales easily, we find that our model transfer paradigm for X-ray image analysis, which we refer to as SyntheX, can even outperform real data-trained models.
  arXiv  Detail & Related papers  (2022-06-13T13:08:41Z)
• Robust and Efficient Medical Imaging with Self-Supervision [80.62711706785834]
  We present REMEDIS, a unified representation learning strategy to improve robustness and data-efficiency of medical imaging AI. We study a diverse range of medical imaging tasks and simulate three realistic application scenarios using retrospective data.
  arXiv  Detail & Related papers  (2022-05-19T17:34:18Z)
• DIME: Fine-grained Interpretations of Multimodal Models via Disentangled Local Explanations [119.1953397679783]
  We focus on advancing the state-of-the-art in interpreting multimodal models. Our proposed approach, DIME, enables accurate and fine-grained analysis of multimodal models.
  arXiv  Detail & Related papers  (2022-03-03T20:52:47Z)
• Self-Training with Improved Regularization for Sample-Efficient Chest X-Ray Classification [80.00316465793702]
  We present a deep learning framework that enables robust modeling in challenging scenarios. Our results show that using 85% lesser labeled data, we can build predictive models that match the performance of classifiers trained in a large-scale data setting.
  arXiv  Detail & Related papers  (2020-05-03T02:36:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.