Purpose

Osteoporosis is a highly prevalent skeletal disease that frequently entails vertebral fractures. Areal bone mineral density (BMD) derived from dual-energy X-ray absorptiometry (DXA) is the reference standard, but has well-known limitations. Texture analysis can provide surrogate markers of tissue microstructure based on computed tomography (CT) or magnetic resonance imaging (MRI) data of the spine, thus potentially improving fracture risk estimation beyond areal BMD. However, it is largely unknown whether MRI-derived texture analysis can predict volumetric BMD (vBMD), or whether a model incorporating texture analysis based on CT and MRI may be capable of differentiating between patients with and without osteoporotic vertebral fractures.

Materials and methods

Twenty-six patients (15 females, median age: 73 years, 11 patients showing at least one osteoporotic vertebral fracture) who had CT and 3-Tesla chemical shift encoding-based water-fat MRI (CSE-MRI) available were analyzed. In total, 171 vertebral bodies of the thoracolumbar spine were segmented using an automatic convolutional neural network (CNN)-based framework, followed by extraction of integral and trabecular vBMD using CT data. For CSE-MRI, manual segmentation of vertebral bodies and consecutive extraction of the mean proton density fat fraction (PDFF) and T2* was performed. First-order, second-order, and higher-order texture features were derived from texture analysis using CT and CSE-MRI data. Stepwise multivariate linear regression models were computed using integral vBMD and fracture status as dependent variables.

Results

Patients with osteoporotic vertebral fractures showed significantly lower integral and trabecular vBMD when compared to patients without fractures (p<0.001). For the model with integral vBMD as the dependent variable, T2* combined with three PDFF-based texture features explained 40% of the variance (adjusted R² [Ra2] = 0.40; p<0.001). Furthermore, regarding the differentiation between patients with and without osteoporotic vertebral fractures, a model including texture features from CT and CSE-MRI data showed better performance than a model based on integral vBMD and PDFF only ( Ra2 = 0.47 vs. Ra2  = 0.81; included texture features in the final model: integral vBMD, CT_Short-run_emphasis, CT_Varianceglobal, and PDFF_Variance).

Conclusion

Using texture analysis for spine CT and CSE-MRI can facilitate the differentiation between patients with and without osteoporotic vertebral fractures, implicating that future fracture prediction in osteoporosis may be improved.

Accurate lesion segmentation is critical in stroke rehabilitation research for the quantification of lesion burden and accurate image processing. Current automated lesion segmentation methods for T1-weighted (T1w) MRIs, commonly used in stroke research, lack accuracy and reliability. Manual segmentation remains the gold standard, but it is time-consuming, subjective, and requires neuroanatomical expertise. We previously released an open-source dataset of stroke T1w MRIs and manually-segmented lesion masks (ATLAS v1.2, N = 304) to encourage the development of better algorithms. However, many methods developed with ATLAS v1.2 report low accuracy, are not publicly accessible or are improperly validated, limiting their utility to the field. Here we present ATLAS v2.0 (N = 1271), a larger dataset of T1w MRIs and manually segmented lesion masks that includes training (n = 655), test (hidden masks, n = 300), and generalizability (hidden MRIs and masks, n = 316) datasets. Algorithm development using this larger sample should lead to more robust solutions; the hidden datasets allow for unbiased performance evaluation via segmentation challenges. We anticipate that ATLAS v2.0 will lead to improved algorithms, facilitating large-scale stroke research.

In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.