Systems and methods for prediction of osteoporotic fracture risk
There is provided a method for predicting risk of osteoporotic fracture, comprising: receiving imaging data of a computed tomography (CT) scan of a body of a patient containing at least a bone portion, the CT scan being performed with settings selected for imaging of non-osteoporosis related pathology; processing the imaging data to identify the bone portion; automatically extracting features based on the imaging data denoting the identified bone portion; computing an osteoporotic fracture predictive factor indicative of the risk of developing at least one osteoporotic fracture in the patient, or the risk of the patient having at least one severe osteoporotic fracture, based on the extracted features, the predictive factor calculated by applying a trained osteoporotic fracture classifier to the extracted features, the osteoporotic fracture classifier trained from data from a plurality of CT scans performed with settings selected for imaging non-osteoporosis related pathology; and providing the predictive factor.
1. A computer-implemented method for predicting risk of osteoporotic fracture for treatment of a patient, comprising:
receiving imaging data of a computed tomography (CT) scan of a body of a patient containing at least a bone portion, the CT scan being performed with settings selected for imaging of non-osteoporosis related pathology;
processing the imaging data to identify the bone portion;
automatically extracting features based on the imaging data denoting the identified bone portion;
computing an osteoporotic fracture predictive factor indicative of the risk of developing at least one osteoporotic fracture in the patient, or the risk of the patient having at least one severe osteoporotic fracture, based on the extracted features, the osteoporotic fracture predictive factor calculated by applying a trained osteoporotic fracture classifier to the extracted features, the osteoporotic fracture classifier trained from data from a plurality of CT scans performed with settings selected for imaging non-osteoporosis related pathology; and
providing the osteoporotic fracture predictive factor for treatment of the patient.
2. The method of claim 1 , further comprising computing a current bone state rating for the patient based on the extracted features.
3. The method of claim 1 , further comprising computing an estimate of the probability of the patient being in a certain bone grade state within a predefined time period based on the trained osteoporotic fracture classifier applied to the extracted features.
4. The method of claim 1 , wherein the osteoporotic fracture predictive factor includes a time frame for a risk of developing or having at least one osteoporotic fracture.
5. The method of claim 1 , wherein the osteoporotic fracture predictive factor is indicative of the risk of developing an osteoporotic vertebral body fracture.
6. The method of claim 1 , wherein processing the imaging data to identify the bone portion comprises segmenting the imaging data to extract at least one vertebrae bone, and wherein automatically extracting features comprises automatically extracting features from at least a portion of at least one segmented vertebrae bone.
7. The method of claim 6 , wherein the extracted feature denotes a trabecular texture characteristic of the vertebral trabecular bone structure associated with degradation of the micro-architecture of bone tissue.
8. The method of claim 6 , wherein the extracted feature denotes one or both of vertebra cortical width and vertebral cortical regularity.
9. The method of claim 6 , wherein the extracted features comprise height of the at least one vertebrae bone along at least a frontal plane, a medial plane, and a back plane.
10. The method of claim 6 , wherein the extracted features comprise a loss of height of the at least one vertebrae bone.
11. The method of claim 10 , further comprising correlating the loss of height of the at least one vertebra bone with at least one score representing a relation to risk of compression fracture in a population.
12. The method of claim 1 , wherein automatically extracting features comprises automatically extracting features associated with a risk of osteoporotic fractures.
13. The method of claim 12 , wherein the extracted features are associated with the risk of osteoporotic fractures include at least one member of a group consisting of: vertebra upper edge flatness, vertebra upper and lower edge angles, vertebrae height changes, and spinal column similarity to healthy model.
14. The method of claim 12 , wherein the automatically extracted features associated with osteoporotic fractures are based on an analysis of the structure of the spine based on identified vertebrae and relation between vertebrae sizes.
15. The method of claim 1 , wherein the osteoporotic fracture predictive factor includes a confidence grade of the risk factor.
16. The method of claim 1 , wherein the CT scan is ordered for a conventional clinical indication including at least one member of a group consisting of: low dose CT scan of the chest to screen for lung cancer, CT scan to screen for colon cancer, standard non-contrast CT scan of the chest, IV contrast CT scan of the chest, standard non-contrast CT scan of the abdomen, IV contrast CT scan of the abdomen, oral contrast CT scan of the abdomen, pelvic CT scan.
17. The method of claim 1 , further comprising detection of osteoporosis fractures and an indication for the position of the detected fracture.
18. The method of claim 1 , wherein the bone portion comprises a vertebra, and the osteoporotic fracture comprises a compression fracture of the vertebra.
19. A computer-implemented method of training an osteoporotic fracture classifier for use in a process to predict risk of osteoporotic fracture for treatment of a patient, comprising:
receiving a corpus of training image files, the training image files comprising data of a CT scan of a body image of at least one patient containing at least one bone portion, the CT scan having being performed with settings selected for imaging of non-osteoporosis related pathology, each respective training image being associated with either a patient diagnosed osteoporosis, or a patient without an osteoporosis diagnosis;
extracting features based on the imaging data denoting the identified bone portion in each respective CT scan; and
training an osteoporotic fracture classifier based on the extracted features, to provide an osteoporotic fracture predictive factor indicative of osteoporotic fracture risk in the patient for treatment of the patient.
20. A system for predicting risk of osteoporotic fracture for treatment of a patient, comprising:
a hardware processor;
an interface for receiving imaging data of a computed tomography (CT) scan of a body of the patient containing at least a bone portion, the CT scan being performed with settings selected for imaging of non-osteoporosis related pathology;
a memory having stored thereon program modules for instruction execution by the processor, comprising:
a bone identification module including instructions to process the imaging data to identify the bone portion;
a feature extraction module including instructions to extract features based on the imaging data denoting the identified bone portion;
a classifier module including instructions to compute an osteoporotic fracture predictive factor indicative of the risk of developing or having at least one osteoporotic fracture in the patient based on the extracted features, the osteoporotic fracture predictive factor calculated by applying a trained osteoporotic fracture classifier to the extracted features, the osteoporotic fracture classifier trained from data from a plurality of CT scans performed with settings selected for imaging non-osteoporosis related pathology; and
an output unit for providing a the osteoporotic fracture predictive factor for treatment of the patient.