Explainable Diagnosis of COVID-19 from Chest X-ray Images via CNNs
| dc.contributor.author | Arkan, Emre | |
| dc.contributor.author | Beckert, Jan Malte | |
| dc.contributor.editor | Gesellschaft für Informatik | |
| dc.date.accessioned | 2021-12-15T10:17:08Z | |
| dc.date.available | 2021-12-15T10:17:08Z | |
| dc.date.issued | 2021 | |
| dc.description.abstract | This work demonstrates how Convolutional Neural Networks ( CNN s) can be used to identify signs of COVID-19 from Chest X-rays (CXR s) and discusses the challenges of deep learning with small datasets. In order to validate the model’s performance, two novel explanation methods LIME and Grad-CAM are explored. Additionally, they serve to further increase users’ confidence in specific classifications. Since the explanation results revealed model biases, additional preprocessing mechanisms were explored: A U-Net-based lung segmenter is introduced to the preprocessing pipeline, which masks all non-lung parts of the CXRs images. Subsequently, the segmentation and non-segmentation results were evaluated with regard to both their performance metrics and interpreted explanation results. | en |
| dc.identifier.isbn | 978-3-88579-751-7 | |
| dc.identifier.pissn | 1614-3213 | |
| dc.identifier.uri | https://dl.gi.de/handle/20.500.12116/37773 | |
| dc.language.iso | en | |
| dc.publisher | Gesellschaft für Informatik, Bonn | |
| dc.relation.ispartof | SKILL 2021 | |
| dc.relation.ispartofseries | Lecture Notes in Informatics (LNI) - Seminars, Volume S-17 | |
| dc.subject | COVID-19 | |
| dc.subject | Chest X-ray | |
| dc.subject | CNN | |
| dc.subject | Grad-CAM | |
| dc.subject | LIME | |
| dc.subject | Explainable AI | |
| dc.title | Explainable Diagnosis of COVID-19 from Chest X-ray Images via CNNs | en |
| gi.citation.endPage | 150 | |
| gi.citation.startPage | 139 | |
| gi.conference.date | 28. September und 01. Oktober 2021 | |
| gi.conference.location | Berlin | |
| gi.conference.sessiontitle | SKILL 2021 |
Dateien
Originalbündel
1 - 1 von 1
Vorschaubild nicht verfügbar
- Name:
- A4-2.pdf
- Größe:
- 975.59 KB
- Format:
- Adobe Portable Document Format