Proximal Femur Fractures Classification

Proximal femur fractures are a significant problem especially of the elderly population in the western world. Starting at an age of 65 the incidence of femoral fractures increases exponentially and is almost doubled every five years. The consequences of proximal femur fractures have a significant socioeconomic impact since the mortality rate one year after the accident ranges between 14 and 36%. In almost all cases, surgical treatment has to be considered the gold standard. If surgical treatment is decelerated, several complications, as well as an increase in mortality rates, may result. Early detection and classification of proximal femur fractures are crucial for the indication of surgery and, if so, to choose the adequate surgical implant. In this context, the Arbeitsgemeinschaft für Osteosynthesefragen (AO-Foundation) established a generally applicable and valid classification system for fractures of all bones of the skeleton based on X-rays including the proximal femurs.

We aim to develop a computer-aided diagnosis (CAD) tool based on radiographs to automatically identify proximal femur fractures in a first step, and consecutively classify them according to the AO standard. Such a CAD system can not only help in the correct classification of fractures but also be effective in planning the optimal therapy for the individual patient since the adequate treatment plan arises from the initial classification.

Publications

Curriculum Learning for Improved Femur Fracture Classification: Scheduling Data with Prior Knowledge and Uncertainty
Amelia Jiménez-Sánchez, Diana Mateus, Sonja Kirchhoff, Chlodwig Kirchhoff, Peter Biberthaler, Nassir Navab, Miguel A. González Ballester, Gemma Piella
Medical Image Analysis 2021
PDF   Bibtex   Code

Precise Proximal Femur Fracture Classification for Interactive Training and Surgical Planning
Amelia Jiménez-Sánchez, Anees Kazi, Shadi Albarqouni, Chlodwig Kirchhoff, Peter Biberthaler, Nassir Navab, Sonja Kirchhoff, Diana Mateus
IPCAI 2020 – IJCARS
PDF   Bibtex   Slides   Video

Medical-based Deep Curriculum Learning for Improved Fracture Classification
Amelia Jiménez-Sánchez, Diana Mateus, Sonja Kirchhoff, Chlodwig Kirchhoff, Peter Biberthaler, Nassir Navab, Miguel A. González Ballester, Gemma Piella
MICCAI 2019
PDF   Bibtex   Slides   Poster