Enabling Real-Time Colonoscopic Polyp Segmentation on Commodity CPUs via Ultra-Lightweight Architecture

• URL: http://arxiv.org/abs/2602.04381v1
• Date: Wed, 04 Feb 2026 10:04:30 GMT
• Title: Enabling Real-Time Colonoscopic Polyp Segmentation on Commodity CPUs via Ultra-Lightweight Architecture
• Abstract summary: We present the UltraSeg family, operating in an extreme-compression regime (0.3 M parameters)<n>Models achieve 90 FPS on a single CPU core without sacrificing accuracy.<n>This work provides not only a CPU-native solution for colonoscopy but also a reproducible blueprint for broader minimally invasive surgical vision applications.
• Score: 11.978087816729534
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Early detection of colorectal cancer hinges on real-time, accurate polyp identification and resection. Yet current high-precision segmentation models rely on GPUs, making them impractical to deploy in primary hospitals, mobile endoscopy units, or capsule robots. To bridge this gap, we present the UltraSeg family, operating in an extreme-compression regime (<0.3 M parameters). UltraSeg-108K (0.108 M parameters) is optimized for single-center data, while UltraSeg-130K (0.13 M parameters) generalizes to multi-center, multi-modal images. By jointly optimizing encoder-decoder widths, incorporating constrained dilated convolutions to enlarge receptive fields, and integrating a cross-layer lightweight fusion module, the models achieve 90 FPS on a single CPU core without sacrificing accuracy. Evaluated on seven public datasets, UltraSeg retains >94% of the Dice score of a 31 M-parameter U-Net while utilizing only 0.4% of its parameters, establishing a strong, clinically viable baseline for the extreme-compression domain and offering an immediately deployable solution for resource-constrained settings. This work provides not only a CPU-native solution for colonoscopy but also a reproducible blueprint for broader minimally invasive surgical vision applications. Source code is publicly available to ensure reproducibility and facilitate future benchmarking.

Related papers

• A Cloud-Edge System for Multimodal Clinical Screening in Resource-Constrained Rural Settings [19.58239117494568]
  Medical AI has demonstrated specialist-level diagnostic accuracy, yet these capabilities remain largely inaccessible in resource-constrained rural settings.<n>We introduce a cloud--edge collaborative architecture that addresses these constraints.<n>We evaluate on 20 multimodal clinical cases spanning cardiac, obstetric, trauma, and screening scenarios.
  arXiv  Detail & Related papers  (2026-08-13T02:45:40Z)
• Enabling Real-Time Point-of-Care Ultrasound Segmentation: A GPU-Free Deployment in Resource-Limited Settings [1.3059755992268272]
  UltraSeg is an ultra-lightweight architecture originally developed for colonoscopic polyp segmentation.<n>This work brings AI costs in line with ultrasound accessibility, making advanced diagnostics available in resource-limited settings.
  arXiv  Detail & Related papers  (2026-06-13T07:49:43Z)
• MonoUNet: A Robust Tiny Neural Network for Automated Knee Cartilage Segmentation on Point-of-Care Ultrasound Devices [3.64325868505768]
  We develop an ultra-compact deep learning model for knee cartilage segmentation on point-of-care ultrasound (POCUS) devices.<n>MonoUNet consists of (i) an aggressively reduced backbone with an asymmetric decoder, (ii) a trainable monogenic block that extracts multi-scale local phase features, and (iii) a gated feature injection mechanism.<n>MonoUNet was evaluated on a multi-site, multi-device knee cartilage ultrasound dataset.
  arXiv  Detail & Related papers  (2026-04-09T04:14:16Z)
• PRISM: Performer RS-IMLE for Single-pass Multisensory Imitation Learning [51.24484551729328]
  We introduce PRISM, a single-pass policy based on a batch-global rejection-sampling variant of IMLE.<n> PRISM couples a temporal multisensory encoder with a linear-attention generator using a Performer architecture.<n>We demonstrate the efficacy of PRISM on a diverse real-world hardware suite, including loco-manipulation using a Unitree Go2 with a 7-DoF arm D1 and tabletop manipulation with a UR5 manipulator.
  arXiv  Detail & Related papers  (2026-02-02T17:57:37Z)
• EndoCaver: Handling Fog, Blur and Glare in Endoscopic Images via Joint Deblurring-Segmentation [9.574713877383244]
  EndoCaver is a lightweight transformer with a unidirectional-guided dual-decoder architecture.<n>It enables joint multi-task capability for image deblurring and segmentation.<n>It is well-suited for on-device clinical deployment.
  arXiv  Detail & Related papers  (2026-01-30T04:18:04Z)
• DRBD-Mamba for Robust and Efficient Brain Tumor Segmentation with Analytical Insights [54.87947751720332]
  Accurate brain tumor segmentation is significant for clinical diagnosis and treatment.<n>Mamba-based State Space Models have demonstrated promising performance.<n>We propose a dual-resolution bi-directional Mamba that captures multi-scale long-range dependencies with minimal computational overhead.
  arXiv  Detail & Related papers  (2025-10-16T07:31:21Z)
• Automated MRI Tumor Segmentation using hybrid U-Net with Transformer and Efficient Attention [0.7456526005219317]
  Cancer is an abnormal growth with potential to invade locally and metastasize to distant organs.<n>Recent AI-based segmentation models are generally trained on large public datasets.<n>This study develops and integrates AI tumor segmentation models directly into hospital software for efficient and accurate oncology treatment planning and execution.
  arXiv  Detail & Related papers  (2025-06-18T15:36:37Z)
• Simple is what you need for efficient and accurate medical image segmentation [7.2109224546543675]
  This paper presents SimpleUNet, a scalable ultra-lightweight medical image segmentation model.<n>With a record-breaking 16 KB parameter configuration, SimpleUNet outperforms LBUNet and other lightweight benchmarks.<n>The 0.67 MB variant achieves superior efficiency (8.60 GFLOPs) and accuracy, attaining a mean DSC/IoU of 85.76%/75.60% on multi-center breast lesion datasets.
  arXiv  Detail & Related papers  (2025-06-16T12:31:48Z)
• Accelerating Large Language Models through Partially Linear Feed-Forward Network [6.916037014703829]
  Large language models (LLMs) demonstrate remarkable capabilities but face deployment challenges due to their massive parameter counts.<n>We present a novel perspective inspired by constant folding in compiler optimization.<n>We propose TARDIS, which enables optimization of LLMs with non-linear activations by partially approximating them with linear functions in frequently occurring input ranges.<n>Experiments demonstrate that TARDIS achieves 80% parameter reduction in feed-forward networks, while significantly outperforming state-of-the-art pruning methods Wanda and RIA with up to 65% higher accuracy.
  arXiv  Detail & Related papers  (2025-01-17T09:20:56Z)
• Towards a Benchmark for Colorectal Cancer Segmentation in Endorectal Ultrasound Videos: Dataset and Model Development [59.74920439478643]
  In this paper, we collect and annotated the first benchmark dataset that covers diverse ERUS scenarios. Our ERUS-10K dataset comprises 77 videos and 10,000 high-resolution annotated frames. We introduce a benchmark model for colorectal cancer segmentation, named the Adaptive Sparse-context TRansformer (ASTR)
  arXiv  Detail & Related papers  (2024-08-19T15:04:42Z)
• LATUP-Net: A Lightweight 3D Attention U-Net with Parallel Convolutions for Brain Tumor Segmentation [7.1789008189318455]
  LATUP-Net is a lightweight 3D ATtention U-Net with Parallel convolutions. It is specifically designed to reduce computational requirements significantly while maintaining high segmentation performance. It achieves promising segmentation performance: the average Dice scores for the whole tumor, tumor core, and enhancing tumor on the BraTS 2020 dataset are 88.41%, 83.82%, and 73.67%, and on the BraTS 2021 dataset, they are 90.29%, 89.54%, and 83.92%, respectively.
  arXiv  Detail & Related papers  (2024-04-09T00:05:45Z)
• 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)
• Revisiting Multi-Scale Feature Fusion for Semantic Segmentation [90.32746095413447]
  In this paper, we demonstrate that neither high internal resolution nor atrous convolutions are necessary for accurate semantic segmentation. We develop a simplified segmentation model, named ESeg, which has neither high internal resolution nor expensive atrous convolutions. Our simple method can achieve better accuracy with faster speed than prior art across multiple datasets.
  arXiv  Detail & Related papers  (2022-03-23T19:14:11Z)
• A TinyML Platform for On-Device Continual Learning with Quantized Latent Replays [66.62377866022221]
  Latent Replay-based Continual Learning (CL) techniques enable online, serverless adaptation in principle. We introduce a HW/SW platform for end-to-end CL based on a 10-core FP32-enabled parallel ultra-low-power processor. Our results show that by combining these techniques, continual learning can be achieved in practice using less than 64MB of memory.
  arXiv  Detail & Related papers  (2021-10-20T11:01:23Z)

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.