Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection
Manual detection of newly formed lesions in multiple sclerosis is an important but tedious and difficult task. Several approaches for automating the detection of new lesions have recently been proposed, but they tend to either overestimate the actual amount of new lesions or to miss many lesions. In...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-09-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2022.981523/full |
| _version_ | 1798037422174371840 |
|---|---|
| author | Julia Andresen Hristina Uzunova Jan Ehrhardt Jan Ehrhardt Timo Kepp Heinz Handels Heinz Handels |
| author_facet | Julia Andresen Hristina Uzunova Jan Ehrhardt Jan Ehrhardt Timo Kepp Heinz Handels Heinz Handels |
| author_sort | Julia Andresen |
| collection | DOAJ |
| description | Manual detection of newly formed lesions in multiple sclerosis is an important but tedious and difficult task. Several approaches for automating the detection of new lesions have recently been proposed, but they tend to either overestimate the actual amount of new lesions or to miss many lesions. In this paper, an image registration convolutional neural network (CNN) that adapts the baseline image to the follow-up image by spatial deformations and simulation of new lesions is proposed. Simultaneously, segmentations of new lesions are generated, which are shown to reliably estimate the real new lesion load and to separate stable and progressive patients. Several applications of the proposed network emerge: image registration, detection and segmentation of new lesions, and modeling of new MS lesions. The modeled lesions offer the possibility to investigate the intensity profile of new lesions. |
| first_indexed | 2024-04-11T21:26:19Z |
| format | Article |
| id | doaj.art-f790bc68037e4c9381335b7d55eb1f71 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-04-11T21:26:19Z |
| publishDate | 2022-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-f790bc68037e4c9381335b7d55eb1f712022-12-22T04:02:24ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2022-09-011610.3389/fnins.2022.981523981523Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detectionJulia Andresen0Hristina Uzunova1Jan Ehrhardt2Jan Ehrhardt3Timo Kepp4Heinz Handels5Heinz Handels6Institute of Medical Informatics, University of Lübeck, Lübeck, GermanyGerman Research Center for Artificial Intelligence, Lübeck, GermanyInstitute of Medical Informatics, University of Lübeck, Lübeck, GermanyGerman Research Center for Artificial Intelligence, Lübeck, GermanyInstitute of Medical Informatics, University of Lübeck, Lübeck, GermanyInstitute of Medical Informatics, University of Lübeck, Lübeck, GermanyGerman Research Center for Artificial Intelligence, Lübeck, GermanyManual detection of newly formed lesions in multiple sclerosis is an important but tedious and difficult task. Several approaches for automating the detection of new lesions have recently been proposed, but they tend to either overestimate the actual amount of new lesions or to miss many lesions. In this paper, an image registration convolutional neural network (CNN) that adapts the baseline image to the follow-up image by spatial deformations and simulation of new lesions is proposed. Simultaneously, segmentations of new lesions are generated, which are shown to reliably estimate the real new lesion load and to separate stable and progressive patients. Several applications of the proposed network emerge: image registration, detection and segmentation of new lesions, and modeling of new MS lesions. The modeled lesions offer the possibility to investigate the intensity profile of new lesions.https://www.frontiersin.org/articles/10.3389/fnins.2022.981523/fullconvolutional neural networksnon-correspondencesimage registrationshape and appearance adaptationmultiple sclerosisnew lesions |
| spellingShingle | Julia Andresen Hristina Uzunova Jan Ehrhardt Jan Ehrhardt Timo Kepp Heinz Handels Heinz Handels Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection Frontiers in Neuroscience convolutional neural networks non-correspondences image registration shape and appearance adaptation multiple sclerosis new lesions |
| title | Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection |
| title_full | Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection |
| title_fullStr | Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection |
| title_full_unstemmed | Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection |
| title_short | Image registration and appearance adaptation in non-correspondent image regions for new MS lesions detection |
| title_sort | image registration and appearance adaptation in non correspondent image regions for new ms lesions detection |
| topic | convolutional neural networks non-correspondences image registration shape and appearance adaptation multiple sclerosis new lesions |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2022.981523/full |
| work_keys_str_mv | AT juliaandresen imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection AT hristinauzunova imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection AT janehrhardt imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection AT janehrhardt imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection AT timokepp imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection AT heinzhandels imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection AT heinzhandels imageregistrationandappearanceadaptationinnoncorrespondentimageregionsfornewmslesionsdetection |