2021
DOI: 10.48550/arxiv.2102.06982
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

DeepRA: Predicting Joint Damage From Radiographs Using CNN with Attention

Abstract: Joint damage in Rheumatoid Arthritis (RA) is assessed by manually inspecting and grading radiographs of hands and feet. This is a tedious task which requires trained experts whose subjective assessment leads to low inter-rater agreement. An algorithm which can automatically predict the joint level damage in hands and feet can help optimize this process, which will eventually aid the doctors in better patient care and research. In this paper, we propose a two-staged approach which amalgamates object detection a… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
3
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
3
1

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(3 citation statements)
references
References 19 publications
0
3
0
Order By: Relevance
“…Another two-staged model, which is proposed in [26], combines the use of the object detection method and convolution neural networks, which can predict the joint level narrowing and erosion SvH scores [13], as well as the overall RA damage, from patients' radiographs. At the first stage, the model performs object detection using the RetinaNet object recognition models.…”
Section: Deep Learning Methodsmentioning
confidence: 99%
“…Another two-staged model, which is proposed in [26], combines the use of the object detection method and convolution neural networks, which can predict the joint level narrowing and erosion SvH scores [13], as well as the overall RA damage, from patients' radiographs. At the first stage, the model performs object detection using the RetinaNet object recognition models.…”
Section: Deep Learning Methodsmentioning
confidence: 99%
“…We report the number of pipeline steps (each corresponding to a separately trained ML model or a group of models), narrowing and erosion results (both as RMSE and absolute rank on the final leaderboard), and whether the method relied on additional training labels. 3 (Stadler & Shi, 2020) 0.4132 (5) 0.4660 (6) 2 (Tran & Nguyen, 2020) 0.4409 (7) 0.4679 (7) 2 (Chaturvedi, 2021) 0.4424 (8) 0.4900 (8) 2 (Chilukri, 2020) 0.4813 (9) 0.5096 (9) 2 Ours 0.4075 (4) 0.4607 (5) 1…”
Section: B Comparison Of Top Submissions To the Ra2 Dream Challengementioning
confidence: 99%
“…Also, the proposed multi-task CNN-based DL model had an alternate system for label smoothing, which integrates classification and regression information into a single loss. Similarly, another algorithm (Chaturvedi, 2021) uses RA2 DREAM Challenge data to predict joint level narrowing and erosion of RA patients' X-ray images. Researchers combined object detection and CNN architecture and developed a two-stage model.…”
mentioning
confidence: 99%