Multi-structure bone segmentation in pediatric MR images with combined regularization from shape priors and adversarial network.

Morphological and diagnostic evaluation of pediatric musculoskeletal system is crucial in clinical practice. However, most segmentation models do not perform well on scarce pediatric imaging data. We propose a new pre-trained regularized convolutional encoder-decoder network for the challenging task of segmenting heterogeneous pediatric magnetic resonance (MR) images. To this end, we have conceived a novel optimization scheme for the segmentation network which comprises additional regularization terms to the loss function. In order to obtain globally consistent predictions, we incorporate a shape priors based regularization, derived from a non-linear shape representation learnt by an auto-encoder. Additionally, an adversarial regularization computed by a discriminator is integrated to encourage precise delineations. The proposed method is evaluated for the task of multi-bone segmentation on two scarce pediatric imaging datasets from ankle and shoulder joints, comprising pathological as well as healthy examinations. The proposed method performed either better or at par with previously proposed approaches for Dice, sensitivity, specificity, maximum symmetric surface distance, average symmetric surface distance, and relative absolute volume difference metrics. We illustrate that the proposed approach can be easily integrated into various bone segmentation strategies and can improve the prediction accuracy of models pre-trained on large non-medical images databases. The obtained results bring new perspectives for the management of pediatric musculoskeletal disorders.

Generalizable multi-task, multi-domain deep segmentation of sparse pediatric imaging datasets via multi-scale contrastive regularization and multi-joint anatomical priors.

Clinical diagnosis of the pediatric musculoskeletal system relies on the analysis of medical imaging examinations. In the medical image processing pipeline, semantic segmentation using deep learning algorithms enables an automatic generation of patient-specific three-dimensional anatomical models which are crucial for morphological evaluation. However, the scarcity of pediatric imaging resources may result in reduced accuracy and generalization performance of individual deep segmentation models. In this study, we propose to design a novel multi-task, multi-domain learning framework in which a single segmentation network is optimized over the union of multiple datasets arising from distinct parts of the anatomy. Unlike previous approaches, we simultaneously consider multiple intensity domains and segmentation tasks to overcome the inherent scarcity of pediatric data while leveraging shared features between imaging datasets. To further improve generalization capabilities, we employ a transfer learning scheme from natural image classification, along with a multi-scale contrastive regularization aimed at promoting domain-specific clusters in the shared representations, and multi-joint anatomical priors to enforce anatomically consistent predictions. We evaluate our contributions for performing bone segmentation using three scarce and pediatric imaging datasets of the ankle, knee, and shoulder joints. Our results demonstrate that the proposed approach outperforms individual, transfer, and shared segmentation schemes in Dice metric with statistically sufficient margins. The proposed model brings new perspectives towards intelligent use of imaging resources and better management of pediatric musculoskeletal disorders.

Anatomically Parameterized Statistical Shape Model: Explaining Morphometry Through Statistical Learning.

OBJECTIVE: Statistical shape models (SSMs) are a popular tool to conduct morphological analysis of anatomical structures which is a crucial step in clinical practices. However, shape representations through SSMs are based on shape coefficients and lack an explicit one-to-one relationship with anatomical measures of clinical relevance. While a shape coefficient embeds a combination of anatomical measures, a formalized approach to find the relationship between them remains elusive in the literature. This limits the use of SSMs to subjective evaluations in clinical practices. We propose a novel SSM controlled by anatomical parameters derived from morphometric analysis. METHODS: The proposed anatomically parameterized SSM (ANAT[Formula: see text]) is based on learning a linear mapping between shape coefficients (latent space) and selected anatomical parameters (anatomical space). This mapping is learned from a synthetic population generated by the standard SSM. Determining the pseudo-inverse of the mapping allows us to build the ANAT[Formula: see text]. We further impose orthogonality constraints to the anatomical parameterization (OC-ANAT[Formula: see text]) to obtain independent shape variation patterns. The proposed contribution was evaluated on two skeletal databases of femoral and scapular bone shapes using clinically relevant anatomical parameters within each (five for femoral and six for scapular bone). RESULTS: Anatomical measures of the synthetically generated shapes exhibited realistic statistics. The learned matrices corroborated well with the obtained statistical relationship, while the two SSMs achieved moderate to excellent performance in predicting anatomical parameters on unseen shapes. CONCLUSION: This study demonstrates the use of anatomical representation for creating anatomically parameterized SSMs and as a result, removes the limited clinical interpretability of standard SSMs. SIGNIFICANCE: The proposed models could help analyze differences in relevant bone morphometry between populations, and be integrated in patient-specific pre-surgery planning or in-surgery assessment.

Statistical Shape Modeling Approach to Predict Missing Scapular Bone.

Prediction of complete and premorbid scapular anatomy is an important aspect of successful shoulder arthroplasty surgeries to treat glenohumeral arthritis and which remains elusive in the current literature. We proposed to build a statistical shape model (SSM) of the scapula and use it to build a framework to predict a complete scapular shape from virtually created scapular bone defects. The bone defects were synthetically created to imitate bone loss in the glenoid region and missing bony part in inferior and superior scapular regions. Sixty seven dry scapulae were used to build the SSM while ten external scapular shapes (not used in SSM building) were selected to map scapular shape variability using its anatomical classification. For each external scapula, four virtual bone defects were created in the superior, inferior and glenoid regions by manually removing a part of the original mesh. Using these defective shapes as prior knowledge, original shapes were reconstructed using scapula SSM and Gaussian process regression. Robustness of the scapula SSM was excellent (generality = 0.79 mm, specificity = 1.74 mm, first 15 principal modes of variations accounted for 95% variability). The validity and quality of the reconstruction of complete scapular bone were evaluated using two methods (1) mesh distances in terms of mean and RMS values and (2) four anatomical measures (three angles: glenoid version, glenoid inclination, and critical shoulder angle, and glenoid center location). The prediction error in the angle measures ranged from 1.0° to 2.2°. For mesh distances, highest mean and RMS error was 0.97 mm and 1.30 respectively. DICE similarity coefficient between the original and predicted shapes was excellent (≥ 0.81). This framework provided high reconstruction accuracy and can be effectively embedded in the pre-surgical planning of shoulder arthroplasty or in morphology-based shoulder biomechanics modeling pipelines.