Integrating model explanations and hybrid priors into deep stacked networks for the "safe zone" prediction of acetabular cup.

BACKGROUND: Existing state-of-the-art "safe zone" prediction methods are statistics-based methods, image-matching techniques, and machine learning methods. Yet, those methods bring a tension between accuracy and interpretability. PURPOSE: To explore the model explanations and estimator consensus for "safe zone" prediction. MATERIAL AND METHODS: We collected the pelvic datasets from Orthopaedic Hospital, and a novel acetabular cup detection method is proposed for automatic ROI segmentation. Hybrid priors comprising both specific priors from data and general priors from experts are constructed. Specifically, specific priors are constructed based on the fine-tuned ResNet-101 convolutional neural networks (CNN) model, and general priors are constructed based on expert knowledge. Our method considers the model explanations and dynamic consensus through appending a SHapley Additive exPlanations (SHAP) module and a dynamic estimator stacking. RESULTS: The proposed method achieves an accuracy of 99.40% and an area under the curve of 0.9998. Experimental results show that our model achieves superior results to the state-of-the-art conventional ensemble classifiers and deep CNN models. CONCLUSION: This new screening model provides a new option for the "safe zone" prediction of acetabular cup.

Automatic Real-Time Space Registration Application for Simulating Dental and Maxillofacial Surgery.

ABSTRACT: Real-time surgical navigation systems are important for preoperative planning and intraoperative navigation. Automatic preoperative multimodal data registration and postoperative spatial registration are extremely crucial in such surgical navigation systems. However, existing automatic multimodal data registration methods have extremely limited application scope due to the lack of accuracy and speed. In addition, the registration results obtained by existing methods are practically lacking and are rarely applied in clinics. To address the above issues, this paper proposes a novel real-time teeth registration algorithm with computed tomography (CT) data and optical tracking scanning data. The proposed method is based on the weighted iterative closest point (ICP) algorithm with 3 improvements: (1) the multilayer spherical point set is generated inside the laser scanning marker sphere, (2) the weight decreases from inside to outside layer by layer, and (3) the weight of the voxel center point set is combined with the CT data of the marker sphere. Specifically, the proposed iCP registration method can overcome the limitation of surface point set registration and tackle the problem of high surface deformity of laser scanning marker spheres. For the registration result of CT and scanning data, the authors employ the real-time spatial registration algorithm based on optical tracking to complete the navigation of the simulated surgical instruments on the multimodal fusion image. The experimental results show that the proposed ICP algorithm reduces the mean square error by 1 order of magnitude and that our method has strong practical value.

Elastic parameter identification of three-dimensional soft tissue based on deep neural network.

In the field of virtual surgery and deformation simulation, the identification of elastic parameters of human soft tissues is a critical technology that directly affects the accuracy of deformation simulation. Current research on soft tissue deformation simulation predominantly assumes that the elasticity of tissues is fixed and already known, leading to the difficulty in populating with the elasticity measured or identified from specific tissues of real patients. Existing elasticity modeling efforts struggle to be implemented on irregularly structured soft tissues, failing to adapt to clinical surgical practices. Therefore, this paper proposes a new method for identifying human soft tissue elastic parameters based on the finite element method and the deep neural network, UNet. This method requires only the full-field displacement data of soft tissues under external loads to predict their elastic distribution. The performance and validity of the algorithm are assessed using test data and clinical data from rhinoplasty surgeries. Experiments demonstrate that the method proposed in this paper can achieve an accuracy of over 99% in predicting elastic parameters. Clinical data validation shows that the predicted elastic distribution can reduce the error in finite element deformation simulations by more than 80% at the maximum compared to the error with traditional uniform elastic parameters, effectively enhancing the computational accuracy in virtual surgery simulations and soft tissue deformation modeling.