Jihun Kim

Distance-preserving rigidity penalty on deformable image registration of multiple skeletal components in the neck

PURPOSE This study aims at developing and testing a novel rigidity penalty suitable for the deformable registration of tightly located skeletal components in the head and neck from planning computed tomography (CT) and daily cone-beam CT (CBCT) scans of patients undergoing radiotherapy. METHODS The proposed rigidity penalty is designed to preserve intervoxel distances within each bony structure. This penalty was tested in the intensity-based B-spline deformable registration of five cervical vertebral bodies (C1-C5). The displacement vector fields (DVFs) from the registrations were compared to the DVFs generated by using rigid body motions of the cervical vertebrae, measured by the surface registration of vertebrae delineated on CT and CBCT images. Twenty five pairs of planning CT (reference) and treatment CBCTs (target) from five patients were aligned without and with the penalty. An existing penalty based on the orthonormality of the deformation gradient tensor was also tested and the effects of the penalties compared. RESULTS The mean magnitude of the maximum registration error with the proposed distance-preserving penalty was (0.86, 1.12, 1.33) mm compared to (2.11, 2.49, 2.46) without penalty and (1.53, 1.64, 1.64) with the existing orthonormality-based penalty. The improvement in the accuracy of the deformable image registration was also verified by comparing the Procrustes distance between the DVFs. With the proposed penalty, the average distance was 0.11 (σ 0.03 mm) which is smaller than 0.53 (0.1 mm) without penalty and 0.28 (0.04 mm) with the orthonormality-based penalty. CONCLUSIONS The accuracy of aligning multiple bony elements was improved by using the proposed distance-preserving rigidity penalty. The voxel-based statistical analysis of the registration error shows that the proposed penalty improved the integrity of the DVFs within the vertebral bodies.

A finite element head and neck model as a supportive tool for deformable image registration.

PurposeA finite element (FE) head and neck model was developed as a tool to aid investigations and development of deformable image registration and patient modeling in radiation oncology. Useful aspects of a FE model for these purposes include ability to produce realistic deformations (similar to those seen in patients over the course of treatment) and a rational means of generating new configurations, e.g., via the application of force and/or displacement boundary conditions.MethodsThe model was constructed based on a cone-beam computed tomography image of a head and neck cancer patient. The three-node triangular surface meshes created for the bony elements (skull, mandible, and cervical spine) and joint elements were integrated into a skeletal system and combined with the exterior surface. Nodes were additionally created inside the surface structures which were composed of the three-node triangular surface meshes, so that four-node tetrahedral FE elements were created over the whole region of the model. The bony elements were modeled as a homogeneous linear elastic material connected by intervertebral disks. The surrounding tissues were modeled as a homogeneous linear elastic material. Under force or displacement boundary conditions, FE analysis on the model calculates approximate solutions of the displacement vector field.ResultsA FE head and neck model was constructed that skull, mandible, and cervical vertebrae were mechanically connected by disks. The developed FE model is capable of generating realistic deformations that are strain-free for the bony elements and of creating new configurations of the skeletal system with the surrounding tissues reasonably deformed.ConclusionsThe FE model can generate realistic deformations for skeletal elements. In addition, the model provides a way of evaluating the accuracy of image alignment methods by producing a ground truth deformation and correspondingly simulated images. The ability to combine force and displacement conditions provides flexibility for simulating realistic anatomic configurations.

Design Optimization of a Solar-Powered Reverse Osmosis Desalination System for Small Communities

Fresh water availability is essential for the economic development in small communities in remote areas. In desert climate, where naturally occurring fresh water is scarce, seawater or brackish water from wells is often more abundant. Since water desalination approaches are energy intensive, a strong motivation exists for the design of cost-effective desalination systems that utilize the abundant renewable energy resource; solar energy. This paper presents an optimization model of a solar-powered reverse osmosis (RO) desalination system. RO systems rely on pumping salty water at high pressure through semi-permeable membrane modules. Under sufficient pressure, water molecules will flow through the membranes, leaving salt ions behind, and are collected in a fresh water stream. Since RO system are primarily powered via electricity, the system model incorporates photovoltaic (PV) panels, and battery storage for smoothing out fluctuations in the PV power output, as well as allowing system operation for a number of hours after sunset. Design variables include sizing of the PV solar collectors, battery storage capacity, as well as the sizing of the RO system membrane module and power elements. The objective is to minimize the cost of unit volume produced fresh water, subject to constraints on production capacity. A genetic algorithm is used to generate and compare optimal designs for two different locations near the Red Sea and Sinai.Copyright

Design optimization of batteryless photovoltaic-powered reverse osmosis water desalination in remote areas

Reverse osmosis (RO) is one of the main technologies for water desalination, which can be used in locations with water resources that have high salinity content (such as saline ground water or seawater) to produce fresh water. Energy requirement for RO is less than other desalination processes, but is in the form of electric power, which can be scarce as fresh water in in remote areas not connected to the grid. Fortunately, many areas with fresh water shortage due to lack of rainfall have abundant sunshine. The combination of solar power and RO desalination is attractive, but remote areas usually requires small modular units, which favors photovoltaic (PV) solar energy harvesting. It is important to consider the net cost-effectiveness of the system when designing the PV-RO desalination plant. Adding battery storage to a PV-RO system has the advantage of steadier operation, but is an additional cost whose real benefit is only realized with a larger PV array that can harvest more energy during daytime. This paper compares the net unit cost of fresh water for realistic scenarios of PV-RO systems with and without battery storage. A multi-level optimization approach previously developed by the authors for time-variant power PV-RO systems is adopted; a “sub-loop” optimization determines the operating pressure and flow rate given a fixed system configuration and instantaneous power input, while an “outer loop” optimizes the configuration of the desalination plant. The sub-loop optimization is done via an enumeration approach, while the outer loop is optimized via a mixed real-coded genetic algorithm (GA). A demonstration study shows a batteryless system being approx. 30% more expensive per unit fresh water production than a fully optimized battery-backed system. However, most of the cost of a batteryless system is in initial investment, which with 7% less annual operating cost, can present a plausible design choice for remote areas.Copyright

An improved rigidity penalty for deformable registration of head and neck images in intensity-modulated radiation therapy

A new rigidity penalty was developed to improve the accuracy of an intensity-based B-spline deformable image registration, and its impact on the displacement vector field (DVF) within cervical vertebrae was investigated. The developed rigidity penalty was tested on ten (10) pairs of cone-beam computed tomography (CBCT) scans. The results were compared to those with a conventional rigidity penalty based on the orthonormality of the deformation gradient tensor. The registration error calculated on the DVFs with both penalties was defined as the distance from the ground-truth DVFs obtained by aligning surface models of the pairs of the image data sets. The ability of the image registration methods to cover multiple rigid body motions was evaluated by using orthogonal Procrustes analysis. The results show that the new rigidity penalty better discouraged local deformation within the skeletal components than the conventional orthonormality penalty.

Cost Optimization of a Solar Humidification-Dehumidification Desalination System Augmented by Thermal Energy Storage

Solar-powered water desalination is one of the promising approaches for addressing fresh water scarcity in the Middle-East, North Africa, and areas of similar climate around the world. Humidification-dehumidification (HDH) is a scalable, commercially-viable technology that primarily utilizes thermal energy in order to extract fresh water from a high salinity water source. Because of inherent variability and uncertainty in solar energy availability due to daily and seasonal cycles, solar-powered HDH desalination systems may benefit from installing thermal energy storage (TES). TES can allow higher utilization of the installed system components and thus reduce the overall lifecycle cost of fresh water production. This work presents a configuration for a HDH desalination system augmented by TES. The system is optimized using Genetic Algorithms (GA) for minimum total annual cost (TAC) per unit volume of produced potable water while satisfying a preset potable water demand. The optimum results for the same location and cost function are compared with results from a previous system which does not have TES. The comparison shows a considerable reduction in potable water production cost when TES is utilized in addition to the benefit of smaller variation in water production across the day.Copyright

Multi-Level Design Optimization of Reverse Osmosis Water Desalination Powered via Photovoltaic Panels With Battery Storage

Reverse osmosis (RO) is one of the main commercial technologies for desalination of water with salinity content too high for human consumption in order to produce fresh water. RO may hold promise for remote areas with scarce fresh water resources, however, RO energy requirements being in the form of electric power have few options in such areas. Fortunately, scarce rainfall is often associated with abundant sunshine, which makes solar photovoltaic (PV) power an attractive option. Equipping a photovoltaic powered reverse osmosis (PV-RO) desalination plant with battery storage has an advantage of steadier and longer hours of operation, thereby making better use of the investments in RO system components, but the additional cost from including batteries may end up increasing the overall cost of fresh water. It is therefore of paramount importance to consider the overall cost-effectiveness of the PV-RO system when designing the desalination plant. Recent work by the authors has generalized the steady operation model of RO systems to hourly adjusted power-dispatch via a proportional-derivative (PD) controller that depends on the state of charge (SOC) of the battery, yet the operating conditions; namely pressure and flow for a given power dispatch were only empirically selected. This paper considers a multi-level optimization model for PV-RO systems with battery storage by considering a “sub-loop” optimization of the feed pressure and flow given power dispatch for a fixed RO system configuration, as well as a “top-level” optimization where the system configuration itself is adjusted by the design variables. Effect of the sub-loop optimization is assessed by comparing the obtained cost of fresh water with the previous empirically adjusted system for locations and weather conditions near the city of Hurgada on the Red Sea.© 2014 ASME

MO‐F‐BRA‐01: A Biomechanical Constraint for Intensity‐Driven Deformable Alignment of Skeletal Components in the Head and Neck Region

PURPOSE To introduce a biomechanical constraint into an intensity-based deformable image registration (DIR) method in order to limit nonphysical deformations of skeletal components in the neck region. METHODS On the reference image, vertebral bodies were segmented. A penalty term, based on the differences in squared inter-voxel distances within each vertebra before and after deformation, was introduced into a routinely used (ITK) intensity-based B-spline alignment algorithm. To assess accuracy, deformable image registration was performed on five pairs of cone-beam CT scans of a head and neck cancer patient. Surface registrations of individual vertebrae established their true displacements (translations and rotations). Orthogonal Procrustes analysis of transformed points within each vertebra established the estimated rotations and translations from the resultant deformation vector fields with and without the penalty term. RESULTS The registration errors across all points within the vertebrae with the penalty term (0.2±0.2, 0.2±0.2, 0.3±0.2) [mm] were significantly lower than without (2.8±2.6, 3.2±2.9, 2.8±3.0) [mm], indicating that employing the penalty term successfully restricted local deformation in the region of the cervical vertebrae. The errors of the bulk translations and rotations of individual vertebrae were similarly reduced: (0.7±0.4, 0.9±0.7, 0.5±0.4) to (O.1±0.1, 0.l+0.1, 0.2±0.2) [mm] for translation and (3.4±2.6, 1.3±1.1, 1.4±1.1) to (0.7±0.6, 0.3±0.2, 0.3±0.3) [°] for rotation. CONCLUSIONS The introduction of a local rigidity penalty improved the integrity of skeletal alignment under neck articulation. Further research will explore biomechanical penalties that will more realistically constrain the changes of other tissues (e.g. muscles) in the neck region. Supported by NIHR01CA59827.

TH‐A‐220‐08: A Feasibility Study of a Rigidity Penalty Term for Deformable Alignment in the Head and Neck Region

Purpose: To investigate the feasibility and impact of a rigidity penalty (RP) on displacement vector field (DVF) for improvement of deformable alignment in the neck region. Methods: An existing B‐spline image registration method was modified to contain a weighted sum of image similarity measure and penalty terms. The penalty term was derived from the orthogonality condition of the deformation gradient tensor. This term was applied to a cervical vertebral body (C3) which was segmented from a reference image. Sum of squared difference (SSD) was used for the image similarity measure. The method was tested on cone‐beam CT scans taken for head and neck intensity‐modulated radiation therapy. To evaluate the accuracies of the existing and modified methods, the strain components of DVFs were calculated, and the measured versus transformed coordinates of 10 landmarks located on the vertebra were compared. Results: The RP value was minimized by the penalized image registration method although the SSD was slightly compromised. The strain components of the DVFs were significantly reduced. Moreover, the registration errors of the landmark pairs were decreased (−0.254 ± 2.176, −2.045 ± 3.894, 0.562 ± 3.179) [mm] to (− 0.381 ± 0.580, 0.236 ± 1.996, 0.437 ± 1.250) [mm]. Conclusions: Reduction in strain indicates that the RP term successfully discouraged local deformation. In addition, improvement in the registration accuracy of the landmark pairs shows the modified method outperforms unregularized alignment. The results also indicate that image alignment using only intensity metrics may result in nonphysical DVFs. Ongoing work will further explore optimizing the weighting of the penalty for clinical images in the neck region, as well as testing on a finite element modeled representation of the neck anatomy undergoing controlled physical deformations. This work was supported by NIHP01CA59827.

SU‐FF‐I‐52: Evaluation of the Contrast‐Detail Response of Digital Radiographic Systems Using the CDRAD Contrast‐Detail Phantom with the CDRAD Analyser Software

Purpose: The CDRAD phantom is frequently used to subjectively evaluate the contrast‐detail response of diagnostic imagingsystems. We investigated the use of this phantom in conjunction with the automated CDRAD Analyser program as a quantitative tool for the evaluation of digital radiographicsystems.Method and Materials: Three DR systems (Trixell Pixium 4600, Canon CXDI‐50G, and DirectRay detectors), and one CRsystem (35‐cm × 43‐cm FUJI ST‐VI plates) were evaluated under two experimental conditions. For the first, the CDRAD phantom was positioned directly on the detector housing, the anti‐scatter grid was removed, and the detector was exposed with two different beam qualities. Quality A: 0.5‐mm Cu filtration, ∼75 kV, 7.1‐mm Al HVL, and Quality B: 0.5‐mm Cu filtration, ∼125 kV, 10.2‐mm Al HVL. The detectors were exposed to ∼0.4, ∼0.8, ∼1.5, ∼2.5, and ∼5.5 mR, for both beam qualities. For the second condition the phantom was sandwiched between two 10‐cm slabs of Lucite, and placed on the patient table. The exposure conditions were the defaults for the Abdomen examination for each system, and the “skin” entrance exposures employed were ∼180, ∼250 and ∼350 mR. Results were compared using the CDRAD Analyser contrast‐detail figure of merit IQFinv. Results: For the Trixell detector the IQFinv increased with exposure from 6.31 to 7.67 for beam quality A, from 5.45 to 7.43 for beam quality B, and from 2.91 to 3.91 for the Abdomen exam. Corresponding values for the Canon detector were 2.88 to 4.77, 2.50 to 4.61, and 1.56 to 2.14. Those for the DirectRay detector were 2.49 to 5.00, 1.86 to 4.93, and 1.34 to 2.15, and those for the CRsystem were 2.89 to 4.16, 2.82 to 4.00, and 1.89 to 2.32. Conclusion: Our results indicate the potential of the CDRAD phantom/CDRAD Analyser as a quantitative image quality analysis tool.