| Title: |
Transfer learning for endoscopy disease detection and segmentation with mask-RCNN benchmark architecture |
| Authors: |
Rezvy, S.; Zebin, T.; Pang, W.; Taylor, S.; Gao, X. |
| Publication Year: |
2020 |
| Collection: |
Middlesex University London: Research Repository |
| Subject Terms: |
deep learning; computer vision; endoscopy; gastrointestinal |
| Description: |
We proposed and implemented a disease detection and semantic segmentation pipeline using a modified mask-RCNN infrastructure model on the EDD2020 dataset1. On the images provided for the phase-I test dataset, for ’BE’, we achieved an average precision of 51.14%, for ’HGD’ and ’polyp’ it is 50%. However, the detection score for ’suspicious’ and ’cancer’ were low. For phase-I, we achieved a dice coefficient of 0.4562 and an F2 score of 0.4508. We noticed the missed and mis-classification was due to the imbalance between classes. Hence, we applied a selective and balanced augmentation stage in our architecture to provide more accurate detection and segmentation. We observed an increase in detection score to 0.29 on phase-II images after balancing the dataset from our phase-I detection score of 0.24. We achieved an improved semantic segmentation score of 0.62 from our phase-I score of 0.52. |
| Document Type: |
conference object |
| File Description: |
application/pdf |
| Language: |
unknown |
| Relation: |
https://repository.mdx.ac.uk/download/f148ad257a2a94b7953cd074e10b8f6808169e0ab52c8b4758eb3d8b1d05c915/2218503/endoCV2020_paper_id_17.pdf; Rezvy, S., Zebin, T., Pang, W., Taylor, S. and Gao, X. 2020. Transfer learning for endoscopy disease detection and segmentation with mask-RCNN benchmark architecture. 2nd International Workshop and Challenge on Computer Vision in Endoscopy. Iowa City, United States 03 Apr 2020 pp. 68-72 |
| Availability: |
https://repository.mdx.ac.uk/item/891q7; https://repository.mdx.ac.uk/download/f148ad257a2a94b7953cd074e10b8f6808169e0ab52c8b4758eb3d8b1d05c915/2218503/endoCV2020_paper_id_17.pdf |
| Rights: |
CC BY 4.0 |
| Accession Number: |
edsbas.8D94FDF1 |
| Database: |
BASE |