Improvement in Disease Diagnosis in Computed Tomography Images by Correlating Organ Volumes with Disease Occurrences in Humans
- In this paper, we investigate the influence of different disease groups on the size of different 1 anatomical structures. To this end, we first modify and improve an existing anatomical segmentation 2 model. Then, we use this model to segment 104 anatomical structures from computed tomography 3 (CT) scans and compute their volumes from the segmentation. After correlating the results with each 4 other, we find no new significant correlations. After correlating the volume data with known diseases 5 for each case, we find two weak correlations, one of which has not been described before and for 6 which we present a possible explanation.
| Author: | Timo van MeegdenburgORCiD, Jens KleesiekORCiD, Jan EggerORCiD, Sören PerreyORCiD |
|---|---|
| URN: | urn:nbn:de:hbz:1010-opus4-42927 |
| DOI: | https://doi.org/10.3390/biomedinformatics3030036 |
| ISSN: | 2673-7426 |
| Parent Title (English): | BioMedInformatics |
| Publisher: | MDPI |
| Document Type: | Article |
| Language: | English |
| Date of Publication (online): | 2023/12/19 |
| Date of first Publication: | 2023/07/05 |
| Publishing Institution: | Westfälische Hochschule Gelsenkirchen Bocholt Recklinghausen |
| Release Date: | 2024/01/26 |
| Tag: | Convolutional Neural Networks; Correlation; Diseases; Segmentation |
| Volume: | 2023 |
| Issue: | 3 (3) |
| Page Number: | 17 |
| First Page: | 526 |
| Last Page: | 542 |
| Institutes: | Institute / Institut für biologische und chemische Informatik |
| Licence (German): | Creative Commons - Namensnennung |



