FRPNet: An improved Faster-ResNet with PASPP for real-time semantic segmentation in the unstructured field scene

Computers and Electronics in Agriculture, Journal Year: 2024, Volume and Issue: 217, P. 108623 - 108623

Published: Jan. 18, 2024

Language: Английский

---

Light-weight Retinal Layer Segmentation with Global Reasoning

IEEE Transactions on Instrumentation and Measurement, Journal Year: 2024, Volume and Issue: 73, P. 1 - 14

Published: Jan. 1, 2024

Automatic retinal layer segmentation with medical images, such as optical coherence tomography (OCT) serves an important tool for diagnosing ophthalmic diseases. However, it is challenging to achieve accurate due low contrast and blood flow noises presented in the images. In addition, algorithm should be light-weight deployed practical clinical applications. Therefore, desired design a network high performance segmentation. this paper, we propose LightReSeg which can applied OCT Specifically, our approach follows encoder-decoder structure, where encoder part employs multi-scale feature extraction Transformer block fully exploiting semantic information of maps at all scales making features have better global reasoning capabilities, while decoder part, asymmetric attention (MAA) module preserving each scale. The experiments show that achieves compared current state-of-the-art method TransUnet 105.7M parameters on both collected dataset two other public datasets, only 3.3M parameters.

Language: Английский

---

CrossU-Net: Dual-modality cross-attention U-Net for segmentation of precancerous lesions in gastric cancer

Computerized Medical Imaging and Graphics, Journal Year: 2024, Volume and Issue: 112, P. 102339 - 102339

Published: Jan. 19, 2024

Language: Английский

---

A Hierarchical Full-Resolution Fusion Network and Topology-aware Connectivity Booster for Retinal Vessel Segmentation

IEEE Transactions on Instrumentation and Measurement, Journal Year: 2024, Volume and Issue: 73, P. 1 - 16

Published: Jan. 1, 2024

Language: Английский

---

Enhancing ROP plus form diagnosis: An automatic blood vessel segmentation approach for newborn fundus images

Results in Engineering, Journal Year: 2024, Volume and Issue: 24, P. 103054 - 103054

Published: Oct. 8, 2024

Language: Английский

---

LTDA‐Mamba：基于CNN‐Mamba混合网络的视网膜血管分割算法

Acta Optica Sinica, Journal Year: 2025, Volume and Issue: 45(7), P. 0717001 - 0717001

Published: Jan. 1, 2025

---

MSPDD-net: Mamba semantic perception dual decoding network for retinal image vessel segmentation

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 193, P. 110370 - 110370

Published: May 19, 2025

Language: Английский

---

DBMA-Net: A Dual-Branch Multiattention Network for Polyp Segmentation

IEEE Transactions on Instrumentation and Measurement, Journal Year: 2024, Volume and Issue: 73, P. 1 - 16

Published: Jan. 1, 2024

In the early prevention stage of colorectal cancer, utilization automatic polyp segmentation techniques from colonoscopy images has demonstrated efficacy in mitigating misdiagnosis rate. Nonetheless, accurate is always against with various challenges, including presence inconsistent size and morphological changes within classes, limited inter-class contrast, high levels interference. recent years, much methodologies based on convolutional neural networks (CNNs) have been widely introduced to enhance precision segmentation. However, two significant hurdles persist: (1) These methods frequently suffer an inadequate acquisition contextual features, causing insufficient feature representation. (2) There a deficiency recognizing intricate information, such as precise boundaries. Addressing these issues, this paper introduces novel dual-branch multi-attention network, denoted DBMA-Net. Specifically, proposed DBMA-Net primarily dual-encoding path that combines CNN Transformer-based approaches enrich Additionally, attention-based fusion module (AFM) incorporated between path, aimed at optimizing features by supplementing local information global insights. Subsequently, distinct attention mechanisms are features: enhancement (AEM) multi-view (MAM), acquire stronger features. modules serve finer details while extensively exploring enhancing lesion region, thereby further elevating accuracy. Following above optimization, enhanced maps hierarchically integrated across multiple scales multi-scale integration (MFIM) for reconstruction. This strategy not only curtails loss but also aids restoring resolution. Ultimately, comprehensive experiments, comparative ablation studies datasets, validate superior performance network compared most state-of-the-art (SOTA) models.

Language: Английский

---

Cross-patch feature interactive net with edge refinement for retinal vessel segmentation

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 174, P. 108443 - 108443

Published: April 9, 2024

Language: Английский

---

Curve-Like Structure Detection Using Multiscale and Boundary-Assisted Segmentation Network

IEEE Transactions on Instrumentation and Measurement, Journal Year: 2024, Volume and Issue: 73, P. 1 - 15

Published: Jan. 1, 2024

Due to the curve-like structure being fine, its contrast with image background weak, and it is often contaminated noise, accurately effectively detecting a major challenge. Furthermore, because of diverse intersecting shapes structure, most existing detection methods are unable obtain continuous complete structure. Therefore, this article proposes robust network based on multiscale boundary assistance. In our work, we initially extract features different sizes by module, then input extracted into triple attention module which learns more representative Finally, acquired fed assistance provide additional information, guiding distinguish background. We conducted experiments various datasets structures, experimental results showed that achieved best $F1$ -score performance across all six datasets. Our model has 67% fewer parameters compared U-Net. Moreover, exhibits excellent adaptability images afflicted noise low brightness levels. It can solve problem fracture during segmentation realize precise addition, yarn hairiness show proposed directional constraint skeleton detect cross structures.

Language: Английский

---