GENERIC MODEL FOR GONIOPHOTOMETER GEOMETRIES

In this work, we present a method to describe the model of a goniophotometer for uncertainty analysis by state-of-the-art Universal Robotic Description Format (URDF). The parameters of the kinematic chain model are determined by measurements of the geometric properties of the goniophotometer. The uncertainties of the pose are determined using Monte Carlo (MC) simulations of the kinematic chain. The measured geometric uncertainties are input the MC simulations. The proposed framework enables high level description of kinematic chains for MC simulations of measurement systems. Furthermore, the uncertainty of the total system is demonstrated over the MC trials to prove a sufficient amount of MC trials. The results of this generic approach are evaluated against an existing model and the uncertainty determination of the same goniophotometer.

Strut Diameter Uncertainty Prediction by Deep Neural Network for Additively Manufactured Lattice Structures

Additive manufacturing (AM) introduces geometric uncertainties on the fabricated strut members of lattice structures. These uncertainties result in deviations between the modeled and fabricated geometries of struts. The use of deep neural networks (DNNs) to accurately predict the statistical parameters of the effective strut diameters to account for the AM-introduced geometric uncertainties with a small training dataset for constant process parameters is studied in this research. For the training data, struts with certain angle and diameter values are fabricated by the material extrusion process. The geometric uncertainties are quantified using the random field theory based on the spatial strut radius measurements obtained from the microscope images of the fabricated struts. The uncertainties are propagated to the effective diameters of the struts using a stochastic upscaling technique. The relationship between the modeled strut diameter and the characterized statistical parameters of the effective diameters are used as the training data to establish a DNN model. The validation results show that the DNN model can predict the statistical parameters of the effective diameters of the struts modeled with angle and diameters different from the ones used in the training data with good accuracy even if the training data set is small. Developing such a DNN model with a small data will allow designers to use the fabricated results in the design optimization processes without requiring additional experimentations.

Task-level Dexterous Manipulation with Multifingered Hand Under Modeling Uncertainties

This article presents an approach to efficiently control grippers/multifingered hands for dexterous manipulation according to a task, i.e. a predefined trajectory in the object space. The object motion is decomposed using a basis of predefined object motions equivalent to object-level coordinates couplings and leading to the definition of the task-level space. In the proposed approach, the decomposition of the motion in the task space is associated with a robust control design based on Linear Matrix Inequalities (LMIs) and Bilinear Matrix Inequalities (BMI). Eigenvalue placement ensures the robustness of the system to geometric uncertainties and eigenvector placement decouples the system according to task specifications. A practical evaluation of the proposed control strategy is provided with a two-fingers and six-DoFs robotic system manipulating an object in the horizontal plane. Results show a better trajectory tracking and the robustness of the control law according to geometric uncertainties and the manipulation of various objects.

Strut Diameter Uncertainty Prediction by Deep Neural Network for Additively Manufactured Lattice Structures

Additive manufacturing introduces geometric uncertainties on the fabricated strut members of lattice structures. These uncertainties lead to deviations between the simulation result and the fabricated mechanical performance. Although these uncertainties can be characterized and quantified in the existing literature, the generation of a high number of samples for the quantified uncertainties to use in the computer-aided design of lattice structures for different strut diameters and angles requires high experimental effort and computational cost. The use of deep neural network models to accurately predict the samples of uncertainties is studied in this research to address this issue. For the training data, the geometric uncertainties on the fabricated struts introduced by the material extrusion process are characterized from microscope measurements using random field theory. These uncertainties are propagated to effective diameters of the strut members using a stochastic upscaling technique. The relationship between the deterministic strut model parameters, namely the model diameter and angle, and the effective diameter with propagated uncertainties is established through a deep neural network model. The validation data results show accurate predictions for the effective diameter when model parameters are given as inputs. Thus, the proposed model has the potential to use the fabricated results in the design optimization processes without requiring computationally expensive repetitive simulations.

An Evaluation of the Impact of MR Image Slice Thickness on the Accuracy of High-Dose-Rate Brachytherapy for Gynaecological Cancers

High-dose-rate (HDR) Magnetic Resonance (MR) guided brachytherapy (BT) is rapidly becoming the standard for treatment of locally advanced cervical cancer, globally. MR is an integral aspect of this treatment, enabling the level of soft tissue visualization required for precise delineation of organ and target contours with respect to the BT applicator or needles during treatment planning. The optimal slice thickness for MR datasets, and the role of super-resolved datasets are questions yet to be investigated. A digital phantom-based study assessed the impact of slice thickness on volumetric and geometric uncertainties in traditional MR datasets and estimated the resultant dosimetric uncertainty. Datasets with traditional slice thicknesses produced uncertainties up to 27% of the imaged structure volume, and contour uncertainty up to one third of the slice thickness This resulted in the exceeding of the American Association of Physicists in Medicine’s (AAPM) recommended dosimetric uncertainty in HDR BT. Trilinearly interpolated datasets reduced these uncertainties substantially, allowing imaging with 2.7 mm coarser slices while conferring an imaging time reduction of 6 minutes. The results of this thesis demonstrate that the recommended range of slice thicknesses introduces uncertainties on a level known to impact dosimetry more than 9%. Trilinearly interpolated datasets may thus confer benefit in this clinical setting.

An Evaluation of the Impact of MR Image Slice Thickness on the Accuracy of High-Dose-Rate Brachytherapy for Gynaecological Cancers

High-dose-rate (HDR) Magnetic Resonance (MR) guided brachytherapy (BT) is rapidly becoming the standard for treatment of locally advanced cervical cancer, globally. MR is an integral aspect of this treatment, enabling the level of soft tissue visualization required for precise delineation of organ and target contours with respect to the BT applicator or needles during treatment planning. The optimal slice thickness for MR datasets, and the role of super-resolved datasets are questions yet to be investigated. A digital phantom-based study assessed the impact of slice thickness on volumetric and geometric uncertainties in traditional MR datasets and estimated the resultant dosimetric uncertainty. Datasets with traditional slice thicknesses produced uncertainties up to 27% of the imaged structure volume, and contour uncertainty up to one third of the slice thickness This resulted in the exceeding of the American Association of Physicists in Medicine’s (AAPM) recommended dosimetric uncertainty in HDR BT. Trilinearly interpolated datasets reduced these uncertainties substantially, allowing imaging with 2.7 mm coarser slices while conferring an imaging time reduction of 6 minutes. The results of this thesis demonstrate that the recommended range of slice thicknesses introduces uncertainties on a level known to impact dosimetry more than 9%. Trilinearly interpolated datasets may thus confer benefit in this clinical setting.