Ferizi_et_al-2019-Journal_of_Magnetic_Resonance_Imaging.pdf...

ORIGINAL RESEARCH
Arti fi cial Intelligence Applied to
Osteoporosis: A Performance Comparison
of Machine Learning Algorithms in
Predicting Fragility Fractures From
MRI Data
Uran Ferizi, PhD,
1
*
Harrison Besser, MA,
1
Pirro Hysi, MD, PhD,
2
Joseph Jacobs, MSc,
3
Chamith S. Rajapakse, PhD,
4
Cheng Chen, PhD,
5
Punam K. Saha, PhD,
5
Stephen Honig, MD,
1
and Gregory Chang, MD
1
Background: A current challenge in osteoporosis is identifying patients at risk of bone fracture.
Purpose: To identify the machine learning classi fi ers that predict best osteoporotic bone fractures and, from the data, to
highlight the imaging features and the anatomical regions that contribute most to prediction performance.
Study Type: Prospective (cross-sectional) case – control study.
Population: Thirty-two women with prior fragility bone fractures, of mean age = 61.6 and body mass index
(BMI) = 22.7 kg/m
2
, and 60 women without fractures, of mean age = 62.3 and BMI = 21.4 kg/m
2
.
Field Strength/ Sequence: 3D FLASH at 3T.
Assessment: Quantitative MRI outcomes by software algorithms. Mechanical and topological microstructural parameters
of the trabecular bone were calculated for fi ve femoral regions, and added to the vector of features together with bone
mineral density measurement, fracture risk assessment tool (FRAX) score, and personal characteristics such as age, weight,
and height. We fi tted 15 classi fi ers using 200 randomized cross-validation datasets.
Statistical Tests: Data: Kolmogorov – Smirnov test for normality. Model Performance: sensitivity, speci fi city, precision, accu-
racy, F1-test, receiver operating characteristic curve (ROC). Two-sided t -test, with P < 0.05 for statistical signi fi cance.
Results: The top three performing classi fi ers are RUS-boosted trees (in particular, performing best with head data,
F1 = 0.64 ± 0.03), the logistic regression and the linear discriminant (both best with trochanteric datasets,
F1 = 0.65 ± 0.03 and F1 = 0.67 ± 0.03, respectively). A permutation of these classi fi ers comprised the best three per-
formers for four out of fi ve anatomical datasets. After averaging across all the anatomical datasets, the score for the best
performer, the boosted trees, was F1 = 0.63 ± 0.03 for All-features dataset, F1 = 0.52 ± 0.05 for the no-MRI dataset, and
F1 = 0.48 ± 0.06 for the no-FRAX dataset.
Data Conclusion: Of many classi fi ers, the RUS-boosted trees, the logistic regression, and the linear discriminant are best
for predicting osteoporotic fracture. Both MRI and FRAX independently add value in identifying osteoporotic fractures.
The femoral head, greater trochanter, and inter-trochanter anatomical regions within the proximal femur yielded better
F1-scores for the best three classi fi ers.
Level of Evidence: 2
Technical Ef fi cacy: Stage 2
J. MAGN. RESON. IMAGING 2019;49:1029 – 1038.
View this article online at wileyonlinelibrary.com. DOI: 10.1002/jmri.26280
Received May 28, 2018, Accepted for publication Jul 17, 2018.