Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution
Abstract Colorectal cancer ranks third in global malignancy incidence, and serrated adenoma is a precursor to colon cancer. However, current studies primarily focus on polyp detection, neglecting the crucial discrimination of polyp nature, hindering effective cancer prevention. This study establishe...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-04-01
|
| Series: | International Journal of Computational Intelligence Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44196-024-00441-8 |
| _version_ | 1827248457154297856 |
|---|---|
| author | Jiading Xu Shuheng Tao Chiye Ma |
| author_facet | Jiading Xu Shuheng Tao Chiye Ma |
| author_sort | Jiading Xu |
| collection | DOAJ |
| description | Abstract Colorectal cancer ranks third in global malignancy incidence, and serrated adenoma is a precursor to colon cancer. However, current studies primarily focus on polyp detection, neglecting the crucial discrimination of polyp nature, hindering effective cancer prevention. This study established a static image dataset for serrated adenoma (SA) and developed a deep learning SA detection model. The proposed MSSDet (Multi-Scale Sub-pixel Detection) innovatively modifies each layer of the original feature pyramid’s structure to retain high-resolution polyp features. Additionally, feature fusion and optimization modules were incorporated to enhance multi-scale information utilization, leveraging the narrow-band imaging endoscope’s ability to provide clearer colonoscopy capillary and texture images. This paper utilized 639 cases of colonic NBI endoscopic images to construct the model, achieving a mean average precision (mAP) of 86.3% for SA in the test set. The SA detection rate via this approach has significantly surpassed conventional object detection methods. |
| first_indexed | 2024-04-24T07:12:28Z |
| format | Article |
| id | doaj.art-fbe02fceadc247279c690bfa39c06a94 |
| institution | Directory Open Access Journal |
| issn | 1875-6883 |
| language | English |
| last_indexed | 2025-03-21T23:32:03Z |
| publishDate | 2024-04-01 |
| publisher | Springer |
| record_format | Article |
| series | International Journal of Computational Intelligence Systems |
| spelling | doaj.art-fbe02fceadc247279c690bfa39c06a942024-05-19T11:32:51ZengSpringerInternational Journal of Computational Intelligence Systems1875-68832024-04-0117111110.1007/s44196-024-00441-8Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel ConvolutionJiading Xu0Shuheng Tao1Chiye Ma2Medical Big Data Center, Shanghai Institute of Computing TechnologyMedical Big Data Center, Shanghai Institute of Computing TechnologyMedical Big Data Center, Shanghai Institute of Computing TechnologyAbstract Colorectal cancer ranks third in global malignancy incidence, and serrated adenoma is a precursor to colon cancer. However, current studies primarily focus on polyp detection, neglecting the crucial discrimination of polyp nature, hindering effective cancer prevention. This study established a static image dataset for serrated adenoma (SA) and developed a deep learning SA detection model. The proposed MSSDet (Multi-Scale Sub-pixel Detection) innovatively modifies each layer of the original feature pyramid’s structure to retain high-resolution polyp features. Additionally, feature fusion and optimization modules were incorporated to enhance multi-scale information utilization, leveraging the narrow-band imaging endoscope’s ability to provide clearer colonoscopy capillary and texture images. This paper utilized 639 cases of colonic NBI endoscopic images to construct the model, achieving a mean average precision (mAP) of 86.3% for SA in the test set. The SA detection rate via this approach has significantly surpassed conventional object detection methods.https://doi.org/10.1007/s44196-024-00441-8NBISerrated adenoma (SA)Sub-pixel-CNNMulti-scale |
| spellingShingle | Jiading Xu Shuheng Tao Chiye Ma Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution International Journal of Computational Intelligence Systems NBI Serrated adenoma (SA) Sub-pixel-CNN Multi-scale |
| title | Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution |
| title_full | Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution |
| title_fullStr | Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution |
| title_full_unstemmed | Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution |
| title_short | Detection of Serrated Adenoma in NBI Based on Multi-Scale Sub-Pixel Convolution |
| title_sort | detection of serrated adenoma in nbi based on multi scale sub pixel convolution |
| topic | NBI Serrated adenoma (SA) Sub-pixel-CNN Multi-scale |
| url | https://doi.org/10.1007/s44196-024-00441-8 |
| work_keys_str_mv | AT jiadingxu detectionofserratedadenomainnbibasedonmultiscalesubpixelconvolution AT shuhengtao detectionofserratedadenomainnbibasedonmultiscalesubpixelconvolution AT chiyema detectionofserratedadenomainnbibasedonmultiscalesubpixelconvolution |