Automatic pectoral muscle removal in mammograms
| Autor: | Taye Girma Debelee, Samuel Rahimeto, Dereje Yohannes, Friedhelm Schwenker |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
03 medical and health sciences
0302 clinical medicine Control and Optimization Control and Systems Engineering Computer science Modeling and Simulation Pectoral muscle 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing 02 engineering and technology Anatomy 030218 nuclear medicine & medical imaging Computer Science Applications |
| Zdroj: | Evolving Systems. 12:519-526 |
| ISSN: | 1868-6486 1868-6478 |
| DOI: | 10.1007/s12530-019-09310-8 |
| Popis: | The pectoral muscle is the high-intensity region in most mediolateral oblique (MLO) views of mammograms. Since it appears at the same intensity as most abnormalities it should be removed for successful classification. Removal of pectoral muscle is often a challenging task since its position, size and shape are different for different patients and it may not occur at all. In this paper, an efficient technique for the detection and removal of pectoral muscle is proposed. The algorithm is tested and proved efficient over a wide range of pectoral muscle types and datasets based on IOU and RMSE value. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink