1. Introduction

2. Related works

3. Methods

3.3. Self-Guided Pseudo-Label Generation

As a general segmentation foundation model, SAM faces challenges when applied to medical image segmentation, particularly in colonoscopy polyp segmentation. The inherent similarity between polyps and surrounding skin tissues, coupled with their indistinct boundaries, exacerbates the complexity of the segmentation task. Consequently, when weak annotation $Y_i$ guides pretrained SAM in generating segmentation mask as pseudo-labels for the next precise segmentation, the resulting segmentation mask lacks precision due to these complexities. Employing imprecise segmentation masks as pseudo-labels not only fails to provide additional information but also may mislead the model, and therefore could impede SAM’s fine-tuning. To address this challenge, we designed three techniques, i.e., multi-augmentation fusion, pixel-level weighting and mask post-processing to enhance the accuracy of ’pseudo-labels’, especially in complex scenarios. For a detailed overview of the model architecture, refer to Figure 2.

Figure 2. WSPoly-SAM polyp framework under bounding box guidance. Note that the masks corresponding to flipping augmentation should be flipped back before engaging in the fusion process.

Figure 2. WSPoly-SAM polyp framework under bounding box guidance. Note that the masks corresponding to flipping augmentation should be flipped back before engaging in the fusion process.

Multi-enhancement fusion (MEF). The information content of a single segmentation mask is often insufficient and inaccurate. In order to enhance the information capacity and accuracy of the segmentation mask, we adopt multiple enhancement fusion techniques. For each image in the training set $(X_i,Y_i)\in \mathcal{S}$, random augmentation is applied, including horizontal flipping, vertical flipping, gaussian blurring, and color jittering. Prompt information also changes accordingly after flipping operations, resulting in the generation of K augmented images and corresponding prompts ${\{X_i^k,Y_i^k\}}_{k=1}^K$. Specifically, color jittering follows the method proposed in [44], where brightness factor is uniformly sampled from the range [0.6, 1.6], contrast is set to 0.2, saturation factor to 0.1, and hue factor to 0.01. Subsequently, we feed ${\{X_i^k,Y_i^k\}}_{k=1}^K$ into SAM to generate the corresponding segmentation masks ${\left\{{\mathcal{M}}_i^k\right\}}_{k=1}^K$, where

$${\mathcal{M}}_i^k=SAM(X_i^k,Y_i^k).$$

(1)

Note that ${\mathcal{M}}_i^k$ and $X_i^k$ have the same shape but differ from the shape of $X_i$. To ensure consistency between the mask and the original image, ${\mathcal{M}}_i^k$ obtained after flipping will be flipped back.

Due to using different augmented images for segmentation, the generated masks ${\mathcal{M}}_i^k$ may exhibit some shape differences, but they often overlap in certain regions. SAM reliably predicts these overlapping regions, remains unaffected by image transformations, and usually corresponds to the correct foreground areas. Considering this complementarity, we propose a method to fuse the segmentation masks of different augmented images. The fusion process is described as follows:

$${\tilde{\mathcal{M}}}_i=\underset{k}{max}{\mathcal{M}}_i^k,$$

(2)

where ${\tilde{\mathcal{M}}}_i$ represents the fused mask. We anticipate the fused mask ${\tilde{\mathcal{M}}}_i$ to be more dependable than individual masks due to its ability to comprehensively reflect the features of the colonoscopy image.

Pixel-level weighting (PLW). The confidence of predictions may fluctuate across pixels. To accentuate the most reliable predictions, we suggest employing entropy for weighting. We compute the entropy of each pixel to generate an entropy map:

$${\tilde{E}}_i=-{\tilde{\mathcal{M}}}_{i}log{\tilde{\mathcal{M}}}_i-(1-{\tilde{\mathcal{M}}}_i)log(1-{\tilde{\mathcal{M}}}_i).$$

(3)

The entropy map ${\tilde{E}}_i$ is computed from the fused mask ${\tilde{\mathcal{M}}}_i$ and quantifies the prediction uncertainty for each pixel across all augmented images. Pixels with high confidence and consistent predictions across all augmented images exhibit low entropy. Thus, we can leverage this entropy map to weight the fused mask, assigning greater weights to more reliable pixels. By applying entropy weights on the fused mask, we obtain a pixel-level weighted mask ${\tilde{M}}_i$, as follows:

$${\tilde{M}}_i=(1-{\tilde{E}}_i)\times {\tilde{\mathcal{M}}}_i.$$

(4)

This strategy considers the prediction uncertainty of pixels, allowing more reliable pixels in the fused mask ${\tilde{\mathcal{M}}}_i$ to receive higher weights. As a result, it enhances the segmentation accuracy of the SAM model in critical regions.

Mask post-processing (MPP). The pixel-level weighted mask ${\tilde{M}}_i$ may have some issues, such as insufficiently smooth edges or disconnected pathological areas. To address these problems, we perform some post-processing operations. First, the pixel-level weighted mask ${\tilde{M}}_i$ is binarized as follows:

$${\tilde{M}}_i^b=Binar{y}_{\alpha }\left({\tilde{M}}_i\right).$$

(5)

${\tilde{M}}_i^b$ represents the binarized mask, with $\alpha$ representing the threshold which is set to 0.2. Next, we employ gaussian filtering techniques to smooth images and diminish noise, along with morphological closing to connect adjacent regions, thereby ensuring the contiguity of pathological areas and smoother edges. Through these post-processing operations, we reduce noise interference and extraneous information, yielding high-quality pseudo-labels ${\tilde{Y}}_i$, as follows:

$${\tilde{Y}}_i=Smooth\left({\tilde{M}}_i^b\right).$$

(6)

Note that, as shown in Figure 4, multi-enhancement fusion, pixel-level weighting and mask post-processing operations are applied sequentially to make the generated pseudo-labels more accurate.

4. Experiments

4.3. Results

4.3.1. Quantitative Results

In this section, we assessed our model’s learning capacity using the Kvasir-SEG and ClinicDB datasets. To ascertain the model’s ability to generalize, we conducted evaluations on three additional datasets: ColonDB, CVC-300, and ETIS. We compared our model with the current mainstream fully-supervised polyp segmentation networks.

Table 1 reports the comparison results of our approach with fully-supervised networks (including UNet [15], UNet++ [16], SFA [22], PraNet [54], C2FNet [55], Polyp-PVT [52], and HSNet [53]). From the comparison in Table 1, WSPoly-SAM exhibits superior performance over existing mainstream baseline networks across the Kvasir-SEG, ColonDB, CVC-300, and ETIS datasets. Specifically, our approach demonstrates a 0.7% enhancement in mDice score and a 0.2% boost in mIoU score compared to HSNet on the Kvasir-SEG dataset. In ColonDB, it achieves a notable 9.4% improvement in mDice score and a substantial 9.6% increase in mIoU score relative to HSNet. Furthermore, on the CVC-300 dataset, it delivers a 3% increase in mDice score and a 3.9% elevation in mIoU score compared to HSNet. On ETIS, it achieves an 8% improvement in mDice score and a 7.9% improvement in mIoU score compared to HSNet. Compared to the original SAM, our method demonstrates significant improvements after fine-tuning on all five datasets, demonstrating that SAM is more adaptable to specific domains after fine-tuning.

4.3.2. Qualitative Results

To visually illustrate the superiority of our proposed method, Figure 3 shows the predicted results of our model compared to the competing models. From Figure 3, our method more accurately locates and identifies the contours of polyps, generating prediction masks that closely resemble ground-truth labels compared to fully-supervised models. This improvement can be attributed to the enhanced learning of complex polyp features in SAM after fine-tuning, with the incorporation of prompts helping to reduce misjudgments in erroneous areas. Additionally, we provide visualizations of the original prediction masks by SAM, revealing issues such as misjudgments, jagged contours, and unclear edges before fine-tuning. Overall, our method effectively captures global contextual information and restores detailed appearance features, thereby better delineating the polyp regions.

4.3.3. Ablation Study

Our method includes a key component, which is pseudo-label generation based on SAM. It consists of three parts: Multi-enhancement fusion (MEF), Pixel-level weighting (PLW), and Mask post-processing (MPP). In this section, we conducted comprehensive ablation experiments on five colonoscopy polyp datasets, with results shown in Table 2. The 1^st row presents the results of directly using SAM to generate segmentation masks on the five datasets. As for the baseline, we used SAM to generate a segmentation mask for training image and fine-tuned SAM (2^nd row) as the baseline without data augmentation. We then gradually added MEF (3^rd row), PLW (4^th row), and MPP (5^th row) on top of the baseline, ultimately achieving the best evaluation scores.

Additionally, we conducted visual analysis to assess the impact of each module on the quality of model predictions (refer to Figure 4). It can be observed that in the baseline scenario, due to unclear boundaries between polyps and surrounding mucosa and low contrast between polyp foreground and background information, ’Baseline’ exhibits issues such as unclear predicted edges and missed detections. After introducing ’MEF’, the fusion of multiple enhancement results provided additional predictive information, significantly reducing instances of missed detections. Subsequently, with the introduction of ’PLW’, assigning higher weights to reliable pixels further reduced the probability of error detection. Finally, with the inclusion of ’MPP’, better noise reduction during prediction and improvement in edge clarity were achieved, further enhancing the segmentation results. In summary, our method enhances the detection coverage of polyp regions, reduces segmentation errors, and minimizes missed detections.

Figure 4. Visualization of the ablation study results.

Figure 4. Visualization of the ablation study results.

4.3.4. Further Analysis

Different Visual Backbones. We also investigated the segmentation capabilities of the SAM model with different Vision Transformer (ViT) for encoding image features, as shown in Table 3. Table 3 (1^st row, 2^nd row, and 3^rd row) shows the predictions of the original SAM model without fine-tuning. It can be observed that on the ETIS dataset, although SAM-H and SAM-L are more computationally complex, their performance is not as good as SAM-B. To fully explore the transfer learning capabilities of SAM on different ViTs, we conducted fine-tuning. Considering memory limitations, in this work, we only fine-tuned SAM-B and SAM-L. From Table 3 (4^th row and 5^th row), it can be observed that under weakly-supervised strategies, SAM-B outperforms SAM-L on the Kvasir-SEG, ClinicDB, and ETIS datasets, while performing less effectively on ColonDB and CVC-300 compared to SAM-L. This indicates that the lightweight SAM-B demonstrates better learning capabilities compared to the higher-complexity SAM-L.

To verify this point, we performed full supervision fine-tuning on SAM-B and SAM-L using ground-truth labels. As shown in Table 3 (6^th row and 7^th row), under full supervision, the trends across datasets align with those observed under weak supervision. In general, although SAM-L entails significantly greater computational complexity, it does not surpass the lightweight SAM-B model on three out of five public datasets. This finding suggests that the lightweight SAM-B model is well-suited for adoption in medical image applications. Additionally, we observed that under the same ViT, weak supervision performs only 0.3% lower in average mDice than full supervision on the Kvasir-SEG, ColonDB, and CVC-300 datasets. This result indicates that while weakly-supervised methods reduce dependence on precise annotations, they can still maintain comparable segmentation precision to fully-supervised methods.

Pseudo-label Generation vs. WSPoly-SAM. We conducted a series of experiments (refer to Table 4) comparing our method WSPoly-SAM with directly using pseudo-label generation, aiming to evaluate whether our fine-tuning approach is superior to the direct adoption of pseudo-label generation. The experimental results from five test datasets indicate that the mDice and mIoU scores obtained through the pseudo-label generation strategy are between the original predictions of SAM and those after fine-tuning, and significantly lower than the results after fine-tuning. This indicates that while the pseudo-label generation method can optimize the predictions of the original SAM to some extent, there is still a certain gap compared to the predictions after fine-tuning. Additionally, the pseudo-label generation process requires multiple invocations of SAM, incurring high time costs, making it challenging for direct use in predictions. In comparison, using WSPoly-SAM is more feasible.

Effectiveness of weakly-supervised fine-tuning in reducing annotation costs. We further investigated the impact of different amounts of training data for supervision masks on the results of fine-tuned SAM, fully demonstrating SAM’s efficiency in utilizing fine-tuning data. Figure 5 illustrates the comparison between weakly-supervised and fully-supervised approaches when the training data amounts from 10% to 100%, using SAM-B model for fine-tuning. As the amount of data increases, the overall accuracy gradually improves. On the CVC-300 dataset, weakly-supervised strategy with 100% training data achieves results comparable to fully-supervised strategy with 90% to 100% training data. On the Kvasir-SEG dataset, weakly-supervised strategy with 100% training data achieves results comparable to fully-supervised strategy with 80% training data. On the ColonDB dataset, weakly-supervised strategy with 100% training data achieves results comparable to fully-supervised strategy with 50% training data. However, performance is relatively poorer on the other two datasets. On the ETIS dataset, weakly-supervised strategy with 100% training data achieves results comparable to fully-supervised strategy with 40% to 50% training data. On the ClinicDB dataset, weakly-supervised strategy with 100% training data only achieves results comparable to fully-supervised strategy with 20% to 30% training data. Overall, using 100% training data under the weakly-supervised strategy achieves 56% to 62% of the training results of fully-supervised learning. We simulated the scenario of annotating polyp data and found that the time ratio between bounding box annotation and rough pixel-level annotation is approximately 1:6 (which would be even higher if precise pixel-level annotation is done by professional annotators). Based on this calculation, for example, if bounding box annotation takes 1 second and pixel-level annotation takes 6 seconds, to achieve the training data results of 56% to 62% under fully-supervised learning with 100 images, the annotation time ratio would be approximately 27% to 30%, saving 70% to 73% of the time cost. This demonstrates the effectiveness of our method in reducing annotation costs.

5. Conclusions

References

1. Isensee, F.; Jaeger, P.F.; Kohl, S.A.; Petersen, J.; Maier-Hein, K.H. nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation. Nature Methods 2021, 18, 203–211. [Google Scholar] [CrossRef] [PubMed]
2. Ouyang, D.; He, B.; Ghorbani, A.; Yuan, N.; Ebinger, J.; Langlotz, C.P.; Heidenreich, P.A.; Harrington, R.A.; Liang, D.H.; Ashley, E.A.; others. Video-based AI for beat-to-beat assessment of cardiac function. Nature 2020, 580, 252–256. [Google Scholar] [CrossRef] [PubMed]
3. Zhao, H.; Shi, J.; Qi, X.; Wang, X.; Jia, J. Pyramid scene parsing network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 2881–2890.
4. Chen, L.C.; Papandreou, G.; Kokkinos, I.; Murphy, K.; Yuille, A.L. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE Transactions on Pattern Analysis and Machine Intelligence 2017, 40, 834–848. [Google Scholar] [CrossRef] [PubMed]
5. Litjens, G.; Kooi, T.; Bejnordi, B.E.; Setio, A.A.A.; Ciompi, F.; Ghafoorian, M.; Van Der Laak, J.A.; Van Ginneken, B.; Sánchez, C.I. A survey on deep learning in medical image analysis. Medical Image Analysis 2017, 42, 60–88. [Google Scholar] [CrossRef] [PubMed]
6. Asgari Taghanaki, S.; Abhishek, K.; Cohen, J.P.; Cohen-Adad, J.; Hamarneh, G. Deep semantic segmentation of natural and medical images: A review. Artificial Intelligence Review 2021, 54, 137–178. [Google Scholar] [CrossRef]
7. Zhang, R.; Lai, P.; Wan, X.; Fan, D.J.; Gao, F.; Wu, X.J.; Li, G. Lesion-aware dynamic kernel for polyp segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 2022, pp. 99–109.
8. Zhou, T.; Zhou, Y.; Gong, C.; Yang, J.; Zhang, Y. Feature aggregation and propagation network for camouflaged object detection. IEEE Transactions on Image Processing 2022, 31, 7036–7047. [Google Scholar] [CrossRef] [PubMed]
9. Ma, J.; He, Y.; Li, F.; Han, L.; You, C.; Wang, B. Segment anything in medical images. Nature Communications 2024, 15, 654. [Google Scholar] [CrossRef] [PubMed]
10. Ding, N.; Qin, Y.; Yang, G.; Wei, F.; Yang, Z.; Su, Y.; Hu, S.; Chen, Y.; Chan, C.M.; Chen, W.; others. Parameter-efficient fine-tuning of large-scale pre-trained language models. Nature Machine Intelligence 2023, 5, 220–235. [Google Scholar] [CrossRef]
11. Kirillov, A.; Mintun, E.; Ravi, N.; Mao, H.; Rolland, C.; Gustafson, L.; Xiao, T.; Whitehead, S.; Berg, A.C.; Lo, W.Y. ; others. Segment anything. Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 4015–4026.
12. Hu, M.; Li, Y.; Yang, X. Skinsam: Empowering skin cancer segmentation with segment anything model. arXiv 2023, arXiv:2304.13973. [Google Scholar]
13. Wu, J.; Fu, R.; Fang, H.; Liu, Y.; Wang, Z.; Xu, Y.; Jin, Y.; Arbel, T. Medical sam adapter: Adapting segment anything model for medical image segmentation. arXiv 2023, arXiv:2304.12620. [Google Scholar]
14. Long, J.; Shelhamer, E.; Darrell, T. Fully convolutional networks for semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 3431–3440.
15. Ronneberger, O.; Fischer, P.; Brox, T. U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, 5-9, October 2015, Proceedings, Part III 18. Springer, 2015, pp. 234–241.
16. Zhou, Z.; Siddiquee, M.M.R.; Tajbakhsh, N.; Liang, J. Unet++: Redesigning skip connections to exploit multiscale features in image segmentation. IEEE Transactions on Medical Imaging 2019, 39, 1856–1867. [Google Scholar] [CrossRef]
17. Yang, X.; Li, X.; Ye, Y.; Lau, R.Y.; Zhang, X.; Huang, X. Road detection and centerline extraction via deep recurrent convolutional neural network U-Net. IEEE Transactions on Geoscience and Remote Sensing 2019, 57, 7209–7220. [Google Scholar] [CrossRef]
18. Qin, X.; Zhang, Z.; Huang, C.; Dehghan, M.; Zaiane, O.R.; Jagersand, M. U2-Net: Going deeper with nested U-structure for salient object detection. Pattern Recognition 2020, 106, 107404. [Google Scholar] [CrossRef]
19. Sun, X.; Zhang, P.; Wang, D.; Cao, Y.; Liu, B. Colorectal polyp segmentation by U-Net with dilation convolution. 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA). IEEE, 2019, pp. 851–858.
20. Chen, L.C.; Zhu, Y.; Papandreou, G.; Schroff, F.; Adam, H. Encoder-decoder with atrous separable convolution for semantic image segmentation. Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 801–818.
21. Liu, S.; Huang, D. ; others. Receptive field block net for accurate and fast object detection. Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 385–400.
22. Fang, Y.; Chen, C.; Yuan, Y.; Tong, K.y. Selective feature aggregation network with area-boundary constraints for polyp segmentation. Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, 13–17, October 2019, Proceedings, Part I 22. Springer, 2019, pp. 302–310.
23. Zhang, R.; Li, G.; Li, Z.; Cui, S.; Qian, D.; Yu, Y. Adaptive context selection for polyp segmentation. Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, 4–8, October 2020, Proceedings, Part VI 23. Springer, 2020, pp. 253–262.
24. Nguyen, T.C.; Nguyen, T.P.; Diep, G.H.; Tran-Dinh, A.H.; Nguyen, T.V.; Tran, M.T. CCBANet: Cascading context and balancing attention for polyp segmentation. Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, 27 September–1 October 2021, Proceedings, Part I 24. Springer, 2021, pp. 633–643.
25. Deng, R.; Cui, C.; Liu, Q.; Yao, T.; Remedios, L.W.; Bao, S.; Landman, B.A.; Wheless, L.E.; Coburn, L.A.; Wilson, K.T. ; others. Segment anything model (sam) for digital pathology: Assess zero-shot segmentation on whole slide imaging. arXiv 2023, arXiv:2304.04155. [Google Scholar]
26. Hu, C.; Li, X. When sam meets medical images: An investigation of segment anything model (sam) on multi-phase liver tumor segmentation. arXiv 2023, arXiv:2304.08506. [Google Scholar]
27. He, S.; Bao, R.; Li, J.; Grant, P.E.; Ou, Y. Accuracy of segment-anything model (sam) in medical image segmentation tasks. arXiv 2023, arXiv:2304.09324. [Google Scholar]
28. Roy, S.; Wald, T.; Koehler, G.; Rokuss, M.R.; Disch, N.; Holzschuh, J.; Zimmerer, D.; Maier-Hein, K.H. Sam. md: Zero-shot medical image segmentation capabilities of the segment anything model. arXiv 2023, arXiv:2304.05396. [Google Scholar]
29. Zhou, T.; Zhang, Y.; Zhou, Y.; Wu, Y.; Gong, C. Can sam segment polyps? arXiv 2023, arXiv:2304.07583. [Google Scholar]
30. Mohapatra, S.; Gosai, A.; Schlaug, G. Brain extraction comparing segment anything model (sam) and fsl brain extraction tool. arXiv 2023, arXiv:2304.04738. [Google Scholar]
31. Mazurowski, M.A.; Dong, H.; Gu, H.; Yang, J.; Konz, N.; Zhang, Y. Segment anything model for medical image analysis: An experimental study. Medical Image Analysis 2023, 89, 102918. [Google Scholar] [CrossRef] [PubMed]
32. Chen, J.; Bai, X. Learning to “segment anything” in thermal infrared images through knowledge distillation with a large scale dataset satir. arXiv 2023, arXiv:2304.07969. [Google Scholar]
33. Tang, L.; Xiao, H.; Li, B. Can sam segment anything? when sam meets camouflaged object detection. arXiv 2023, arXiv:2304.04709. [Google Scholar]
34. Ji, G.P.; Fan, D.P.; Xu, P.; Cheng, M.M.; Zhou, B.; Van Gool, L. SAM Struggles in Concealed Scenes–Empirical Study on” Segment Anything“. arXiv 2023, arXiv:2304.06022. [Google Scholar]
35. Ji, W.; Li, J.; Bi, Q.; Li, W.; Cheng, L. Segment anything is not always perfect: An investigation of sam on different real-world applications. arXiv 2023, arXiv:2304.05750. [Google Scholar]
36. Ma, J.; He, Y.; Li, F.; Han, L.; You, C.; Wang, B. Segment anything in medical images. Nature Communications 2024, 15, 654. [Google Scholar] [CrossRef] [PubMed]
37. Cheng, J.; Ye, J.; Deng, Z.; Chen, J.; Li, T.; Wang, H.; Su, Y.; Huang, Z.; Chen, J.; Jiang, L. ; others. Sam-med2d. arXiv 2023, arXiv:2308.16184. [Google Scholar]
38. Jiang, P.T.; Yang, Y. Segment anything is a good pseudo-label generator for weakly supervised semantic segmentation. arXiv 2023, arXiv:2305.01275. [Google Scholar]
39. He, C.; Li, K.; Zhang, Y.; Xu, G.; Tang, L.; Zhang, Y.; Guo, Z.; Li, X. Weakly-supervised concealed object segmentation with sam-based pseudo labeling and multi-scale feature grouping. Advances in Neural Information Processing Systems 2024, 36. [Google Scholar]
40. Reedha, R.; Dericquebourg, E.; Canals, R.; Hafiane, A. Transformer neural network for weed and crop classification of high resolution UAV images. Remote Sensing 2022, 14, 592. [Google Scholar] [CrossRef]
41. Brown, T.; Mann, B.; Ryder, N.; Subbiah, M.; Kaplan, J.D.; Dhariwal, P.; Neelakantan, A.; Shyam, P.; Sastry, G.; Askell, A.; others. Language models are few-shot learners. Advances in Neural Information Processing Systems 2020, 33, 1877–1901. [Google Scholar]
42. Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S. ; others. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv 2020, arXiv:2010.11929. [Google Scholar]
43. He, K.; Chen, X.; Xie, S.; Li, Y.; Dollár, P.; Girshick, R. Masked autoencoders are scalable vision learners. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 16000–16009.
44. Zhang, W.; Fu, C.; Zheng, Y.; Zhang, F.; Zhao, Y.; Sham, C.W. HSNet: A hybrid semantic network for polyp segmentation. Computers in Biology and Medicine 2022, 150, 106173. [Google Scholar] [CrossRef] [PubMed]
45. Isensee, F.; Jaeger, P.F.; Kohl, S.A.; Petersen, J.; Maier-Hein, K.H. nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation. Nature Methods 2021, 18, 203–211. [Google Scholar] [CrossRef] [PubMed]
46. Ma, J.; Chen, J.; Ng, M.; Huang, R.; Li, Y.; Li, C.; Yang, X.; Martel, A.L. Loss odyssey in medical image segmentation. Medical Image Analysis 2021, 71, 102035. [Google Scholar] [CrossRef]
47. Jha, D.; Smedsrud, P.H.; Riegler, M.A.; Halvorsen, P.; De Lange, T.; Johansen, D.; Johansen, H.D. Kvasir-seg: A segmented polyp dataset. MultiMedia Modeling: 26th International Conference, MMM 2020, Daejeon, South Korea, 5–8, January 2020, Proceedings, Part II 26. Springer, 2020, pp. 451–462.
48. Bernal, J.; Sánchez, F.J.; Fernández-Esparrach, G.; Gil, D.; Rodríguez, C.; Vilariño, F. WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians. Computerized Medical Imaging and Graphics 2015, 43, 99–111. [Google Scholar] [CrossRef]
49. Tajbakhsh, N.; Gurudu, S.R.; Liang, J. Automated polyp detection in colonoscopy videos using shape and context information. IEEE Transactions on Medical Imaging 2015, 35, 630–644. [Google Scholar] [CrossRef] [PubMed]
50. Vázquez, D.; Bernal, J.; Sánchez, F.J.; Fernández-Esparrach, G.; López, A.M.; Romero, A.; Drozdzal, M.; Courville, A. A benchmark for endoluminal scene segmentation of colonoscopy images. Journal of Healthcare Engineering 2017, 2017, 4037190. [Google Scholar] [CrossRef] [PubMed]
51. Silva, J.; Histace, A.; Romain, O.; Dray, X.; Granado, B. Toward embedded detection of polyps in wce images for early diagnosis of colorectal cancer. International Journal of Computer Assisted Radiology and Surgery 2014, 9, 283–293. [Google Scholar] [CrossRef] [PubMed]
52. Dong, B.; Wang, W.; Fan, D.P.; Li, J.; Fu, H.; Shao, L. Polyp-pvt: Polyp segmentation with pyramid vision transformers. arXiv 2021, arXiv:2108.06932. [Google Scholar]
53. Zhang, W.; Fu, C.; Zheng, Y.; Zhang, F.; Zhao, Y.; Sham, C.W. HSNet: A hybrid semantic network for polyp segmentation. Computers in Biology and Medicine 2022, 150, 106173. [Google Scholar] [CrossRef]
54. Fan, D.P.; Ji, G.P.; Zhou, T.; Chen, G.; Fu, H.; Shen, J.; Shao, L. Pranet: Parallel reverse attention network for polyp segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 2020, pp. 263–273.
55. Sun, Y.; Chen, G.; Zhou, T.; Zhang, Y.; Liu, N. Context-aware cross-level fusion network for camouflaged object detection. arXiv 2021, arXiv:2105.12555. [Google Scholar]