Fully automatic system to detect and segment the proximal femur in pelvic radiographic images for Legg–Calvé–Perthes disease (LCPD)
We used readily available computer vision methods for image processing in LCPD. This means better ways to classify, grade, and monitor patients in the future!
We achieved impressive results:
- Detection Accuracy: 99%
- Segmentation Accuracy: 91%
- Dice Coefficient: 0.75
- Binary IoU Score: 0.85
A huge shoutout to Sofie, Nicole, and Louise for their algorithm development, and a big thanks to Prof. Ole Rahbek and Prof. Harry Kim for their leadership. And thanks to Rikke V. Boel for her help with image classification.
Opdateret