Showing 1–14 of 14 results for author: Boyd, K

ORMIR_XCT: A Python package for high resolution peripheral quantitative computed tomography image processing

Authors: Michael T. Kuczynski, Nathan J. Neeteson, Kathryn S. Stok, Andrew J. Burghardt, Michelle A. Espinosa Hernandez, Jared Vicory, Justin J. Tse, Pholpat Durongbhan, Serena Bonaretti, Andy Kin On Wong, Steven K. Boyd, Sarah L. Manske

Abstract: High resolution peripheral quantitative computed tomography (HR-pQCT) is an imaging technique capable of imaging trabecular bone in-vivo. HR-pQCT has a wide range of applications, primarily focused on bone to improve our understanding of musculoskeletal diseases, assess epidemiological associations, and evaluate the effects of pharmaceutical interventions. Processing HR-pQCT images has largely bee… ▽ More High resolution peripheral quantitative computed tomography (HR-pQCT) is an imaging technique capable of imaging trabecular bone in-vivo. HR-pQCT has a wide range of applications, primarily focused on bone to improve our understanding of musculoskeletal diseases, assess epidemiological associations, and evaluate the effects of pharmaceutical interventions. Processing HR-pQCT images has largely been supported using the scanner manufacturer scripting language (Image Processing Language, IPL, Scanco Medical). However, by expanding image processing workflows outside of the scanner manufacturer software environment, users have the flexibility to apply more advanced mathematical techniques and leverage modern software packages to improve image processing. The ORMIR_XCT Python package was developed to reimplement some existing IPL workflows and provide an open and reproducible package allowing for the development of advanced HR-pQCT data processing workflows. △ Less

Submitted 8 September, 2023; originally announced September 2023.

Journal ref: Journal of Open Source Software, 9(97), 6084 (2024)

Almanac: Retrieval-Augmented Language Models for Clinical Medicine

Authors: Cyril Zakka, Akash Chaurasia, Rohan Shad, Alex R. Dalal, Jennifer L. Kim, Michael Moor, Kevin Alexander, Euan Ashley, Jack Boyd, Kathleen Boyd, Karen Hirsch, Curt Langlotz, Joanna Nelson, William Hiesinger

Abstract: Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In… ▽ More Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In this study, we develop Almanac, a large language model framework augmented with retrieval capabilities for medical guideline and treatment recommendations. Performance on a novel dataset of clinical scenarios (n = 130) evaluated by a panel of 5 board-certified and resident physicians demonstrates significant increases in factuality (mean of 18% at p-value < 0.05) across all specialties, with improvements in completeness and safety. Our results demonstrate the potential for large language models to be effective tools in the clinical decision-making process, while also emphasizing the importance of careful testing and deployment to mitigate their shortcomings. △ Less

Submitted 31 May, 2023; v1 submitted 28 February, 2023; originally announced March 2023.

Bone Adaptation as a Geometric Flow

Authors: Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Abstract: This paper presents bone adaptation as a geometric flow. The proposed method is based on two assumptions: first, that the bone surface is smooth (not fractal) permitting the definition of a tangent plane and, second, that the interface between marrow and bone tissue phases is orientable. This permits the analysis of bone adaptation using the well-developed mathematics of geometric flows and the nu… ▽ More This paper presents bone adaptation as a geometric flow. The proposed method is based on two assumptions: first, that the bone surface is smooth (not fractal) permitting the definition of a tangent plane and, second, that the interface between marrow and bone tissue phases is orientable. This permits the analysis of bone adaptation using the well-developed mathematics of geometric flows and the numerical techniques of the level set method. Most importantly, topological changes such as holes forming in plates and rods disconnecting can be treated formally and simulated naturally. First, the relationship between biological theories of bone adaptation and the mathematical object describing geometric flow is described. This is termed the adaptation function, $F$, and is the multi-scale link described by Frost's Utah paradigm between cellular dynamics and bone structure. Second, a model of age-related bone loss termed curvature-based bone adaptation is presented. Using previous literature, it is shown that curvature-based bone adaptation is the limiting continuous equation of simulated bone atrophy, a discrete model of bone aging. Interestingly, the parameters of the model can be defined in such a way that the flow is volume-preserving. This implies that bone health can in principle change in ways that fundamentally cannot be measured by areal or volumetric bone mineral density, requiring structure-level imaging. Third, a numerical method is described and two in silico experiments are performed demonstrating the non-volume-preserving and volume-preserving cases. Taken together, recognition of bone adaptation as a geometric flow permits the recruitment of mathematical and numerical developments over the last 50 years to understanding and describing the complex surface of bone. △ Less

Submitted 8 November, 2021; originally announced November 2021.

Comments: 14 pages, 6 figures

Constructing High-Order Signed Distance Maps from Computed Tomography Data with Application to Bone Morphometry

Authors: Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Abstract: An algorithm is presented for constructing high-order signed distance fields for two phase materials imaged with computed tomography. The signed distance field is high-order in that it is free of the quantization artifact associated with the distance transform of sampled signals. The narrowband is solved using a closest point algorithm extended for implicit embeddings that are not a signed distanc… ▽ More An algorithm is presented for constructing high-order signed distance fields for two phase materials imaged with computed tomography. The signed distance field is high-order in that it is free of the quantization artifact associated with the distance transform of sampled signals. The narrowband is solved using a closest point algorithm extended for implicit embeddings that are not a signed distance field. The high-order fast swee** algorithm is used to extend the narrowband to the remainder of the domain. The order of accuracy of the narrowband and extension methods are verified on ideal implicit surfaces. The method is applied to ten excised cubes of bovine trabecular bone. Localization of the surface, estimation of phase densities, and local morphometry is validated with these subjects. Since the embedding is high-order, gradients and thus curvatures can be accurately estimated locally in the image data. △ Less

Submitted 1 November, 2021; originally announced November 2021.

Comments: 14 pages, 14 figures

High-Order Signed Distance Transform of Sampled Signals

Authors: Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Abstract: Signed distance transforms of sampled signals can be constructed better than the traditional exact signed distance transform. Such a transform is termed the high-order signed distance transform and is defined as satisfying three conditions: the Eikonal equation, recovery by a Heaviside function, and has an order of accuracy greater than unity away from the medial axis. Such a transform is an impro… ▽ More Signed distance transforms of sampled signals can be constructed better than the traditional exact signed distance transform. Such a transform is termed the high-order signed distance transform and is defined as satisfying three conditions: the Eikonal equation, recovery by a Heaviside function, and has an order of accuracy greater than unity away from the medial axis. Such a transform is an improvement to the classic notion of an exact signed distance transform because it does not exhibit artifacts of quantization. A large constant, linear time complexity high-order signed distance transform for arbitrary dimensionality sampled signals is developed based on the high order fast swee** method. The transform is initialized with an exact signed distance transform and quantization corrected through an upwind solver for the boundary value Eikonal equation. The proposed method cannot attain arbitrary order of accuracy and is limited by the initialization method and non-uniqueness of the problem. However, meshed surfaces are visually smoother and do not exhibit artifacts of quantization in local mean and Gaussian curvature. △ Less

Submitted 25 October, 2021; originally announced October 2021.

Comments: 11 pages, 11 figures

Local Morphometry of Closed, Implicit Surfaces

Authors: Bryce A Besler, Tannis D. Kemp, Andrew S. Michalski, Nils D. Forkert, Steven K. Boyd

Abstract: Anatomical structures such as the hippocampus, liver, and bones can be analyzed as orientable, closed surfaces. This permits the computation of volume, surface area, mean curvature, Gaussian curvature, and the Euler-Poincaré characteristic as well as comparison of these morphometrics between structures of different topology. The structures are commonly represented implicitly in curve evolution pro… ▽ More Anatomical structures such as the hippocampus, liver, and bones can be analyzed as orientable, closed surfaces. This permits the computation of volume, surface area, mean curvature, Gaussian curvature, and the Euler-Poincaré characteristic as well as comparison of these morphometrics between structures of different topology. The structures are commonly represented implicitly in curve evolution problems as the zero level set of an embedding. Practically, binary images of anatomical structures are embedded using a signed distance transform. However, quantization prevents the accurate computation of curvatures, leading to considerable errors in morphometry. This paper presents a fast, simple embedding procedure for accurate local morphometry as the zero crossing of the Gaussian blurred binary image. The proposed method was validated based on the femur and fourth lumbar vertebrae of 50 clinical computed tomography datasets. The results show that the signed distance transform leads to large quantization errors in the computed local curvature. Global validation of morphometry using regression and Bland-Altman analysis revealed that the coefficient of determination for the average mean curvature is improved from 93.8% with the signed distance transform to 100% with the proposed method. For the surface area, the proportional bias is improved from -5.0% for the signed distance transform to +0.6% for the proposed method. The Euler-Poincaré characteristic is improved from unusable in the signed distance transform to 98% accuracy for the proposed method. The proposed method enables an improved local and global evaluation of curvature for purposes of morphometry on closed, implicit surfaces. △ Less

Submitted 29 July, 2021; originally announced August 2021.

Sphere Under Advection and Mean Curvature Flow

Authors: Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Abstract: Advection and mean curvature flow is used as a model of bone microarchitecture adaptation. It is an equivalent geometric flow to prescribed mean curvature flow with an additional rate term. In order to validate numerical methods for simulating this flow and develo** an inverse solver, a closed-form solution for advection and mean curvature flow of a sphere is derived. Advection and mean curvature flow is used as a model of bone microarchitecture adaptation. It is an equivalent geometric flow to prescribed mean curvature flow with an additional rate term. In order to validate numerical methods for simulating this flow and develo** an inverse solver, a closed-form solution for advection and mean curvature flow of a sphere is derived. △ Less

Submitted 18 December, 2020; originally announced December 2020.

Comments: 5 pages, 5 figures

Artifacts of Quantization in Distance Transforms

Authors: Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Abstract: Distance transforms are a central tool in shape analysis, morphometry, and curve evolution problems. This work describes and investigates an artifact present in distance maps computed from sampled signals. Namely, sampling reflects through the distance transform causing quantization in the resulting distance map. Gradients of the quantized distance map show banding, affecting the quality of subseq… ▽ More Distance transforms are a central tool in shape analysis, morphometry, and curve evolution problems. This work describes and investigates an artifact present in distance maps computed from sampled signals. Namely, sampling reflects through the distance transform causing quantization in the resulting distance map. Gradients of the quantized distance map show banding, affecting the quality of subsequence processing. Furthermore, this error is independent of the sampling period of the signal and cannot be removed by modifying the number of samples across an objects width. Where needed, distance maps should be computed from representations other than binary images. In the case where exact representations are needed, a dithering and noise removal algorithm is proposed. △ Less

Submitted 17 November, 2020; originally announced November 2020.

Comments: 13 pages, 16 figures, This work has been submitted to the IEEE for possible publication

Robust Self-Supervised Learning of Deterministic Errors in Single-Plane (Monoplanar) and Dual-Plane (Biplanar) X-ray Fluoroscopy

Authors: Jacky C. K. Chow, Steven K. Boyd, Derek D. Lichti, Janet L. Ronsky

Abstract: Fluoroscopic imaging that captures X-ray images at video framerates is advantageous for guiding catheter insertions by vascular surgeons and interventional radiologists. Visualizing the dynamical movements non-invasively allows complex surgical procedures to be performed with less trauma to the patient. To improve surgical precision, endovascular procedures can benefit from more accurate fluorosco… ▽ More Fluoroscopic imaging that captures X-ray images at video framerates is advantageous for guiding catheter insertions by vascular surgeons and interventional radiologists. Visualizing the dynamical movements non-invasively allows complex surgical procedures to be performed with less trauma to the patient. To improve surgical precision, endovascular procedures can benefit from more accurate fluoroscopy data via calibration. This paper presents a robust self-calibration algorithm suitable for single-plane and dual-plane fluoroscopy. A three-dimensional (3D) target field was imaged by the fluoroscope in a strong geometric network configuration. The unknown 3D positions of targets and the fluoroscope pose were estimated simultaneously by maximizing the likelihood of the Student-t probability distribution function. A smoothed k-nearest neighbour (kNN) regression is then used to model the deterministic component of the image reprojection error of the robust bundle adjustment. The Maximum Likelihood Estimation step and the kNN regression step are then repeated iteratively until convergence. Four different error modeling schemes were compared while varying the quantity of training images. It was found that using a smoothed kNN regression can automatically model the systematic errors in fluoroscopy with similar accuracy as a human expert using a small training dataset. When all training images were used, the 3D map** error was reduced from 0.61-0.83 mm to 0.04 mm post-calibration (94.2-95.7% improvement), and the 2D reprojection error was reduced from 1.17-1.31 to 0.20-0.21 pixels (83.2-83.8% improvement). When using biplanar fluoroscopy, the 3D measurement accuracy of the system improved from 0.60 mm to 0.32 mm (47.2% improvement). △ Less

Submitted 2 January, 2020; originally announced January 2020.

Ultra-broadband photodetectors based on epitaxial graphene quantum dots

Authors: Abdel El Fatimy, Anindya Nath, Byoung Don Kong, Anthony K. Boyd, Rachael L. Myers-Ward, Kevin Daniels, M. Mehdi Jadidi, Thomas E. Murphy, D. Kurt Gaskill, Paola Barbara

Abstract: Graphene is an ideal material for hot-electron bolometers, due to its low heat capacity and weak electron-phonon coupling. Nanostructuring graphene with quantum dot constrictions yields detectors with extraordinarily high intrinsic responsivity, higher than 1x10^9 V/W at 3K. The sensing mechanism is bolometric in nature: the quantum confinement gap causes a strong dependence of the electrical resi… ▽ More Graphene is an ideal material for hot-electron bolometers, due to its low heat capacity and weak electron-phonon coupling. Nanostructuring graphene with quantum dot constrictions yields detectors with extraordinarily high intrinsic responsivity, higher than 1x10^9 V/W at 3K. The sensing mechanism is bolometric in nature: the quantum confinement gap causes a strong dependence of the electrical resistance on the electron temperature. Here we show that this quantum confinement gap does not impose a limitation on the photon energy for light detection and these quantum dot bolometers work in a very broad spectral range, from terahertz, through telecom to ultraviolet radiation, with responsivity independent of wavelength. We also measure the power dependence of the response. Although the responsivity decreases with increasing power, it stays higher than 1x10^8 V/W in a wide range of absorbed power, from 1 pW to 0.4 nW. △ Less

Submitted 13 September, 2017; originally announced September 2017.

Journal ref: Nanophotonics 20170100 (2018)

Epitaxial graphene quantum dots for high-performance THz bolometers

Authors: A. El Fatimy, R. L. Myers-Ward, A. K. Boyd, K. M. Daniels, D. K. Gaskill, P. Barbara

Abstract: Light absorption in graphene causes a large change in electron temperature, due to low electronic heat capacity and weak electron phonon coupling [1-3], making it very attractive as a hot-electron bolometer material. Unfortunately, the weak variation of electrical resistance with temperature has substantially limited the responsivity of graphene bolometers. Here we show that quantum dots of epitax… ▽ More Light absorption in graphene causes a large change in electron temperature, due to low electronic heat capacity and weak electron phonon coupling [1-3], making it very attractive as a hot-electron bolometer material. Unfortunately, the weak variation of electrical resistance with temperature has substantially limited the responsivity of graphene bolometers. Here we show that quantum dots of epitaxial graphene on SiC exhibit an extraordinarily high variation of resistance with temperature due to quantum confinement, higher than 430 Mohm/K at 2.5 K, leading to responsivities for absorbed THz power above 10^10 V/W. This is five orders of magnitude higher than other types of graphene hot electron bolometers. The high responsivity combined with an extremely low noise-equivalent power, about 0.2 fW/Hz^0.5 at 2.5K, place the performance of graphene quantum dot bolometers well above commercial cooled bolometers. Additionally, these quantum dot bolometers have the potential for superior performance for operation above 77K. △ Less

Submitted 15 September, 2015; originally announced September 2015.

Journal ref: Nature Nanotechnology (2016)

Hybrid Metal-Graphene Plasmons for Tunable Terahertz Technology

Authors: Mohammad M. Jadidi, Andrei B. Sushkov, Rachael L. Myers-Ward, Anthony K. Boyd, Kevin M. Daniels, D. Kurt Gaskill, Michael S. Fuhrer, H. Dennis Drew, Thomas E. Murphy

Abstract: Among its many outstanding properties, graphene supports terahertz surface plasma waves -- sub-wavelength charge density oscillations connected with electromagnetic fields that are tightly localized near the surface[1,2]. When these waves are confined to finite-sized graphene, plasmon resonances emerge that are characterized by alternating charge accumulation at the opposing edges of the graphene.… ▽ More Among its many outstanding properties, graphene supports terahertz surface plasma waves -- sub-wavelength charge density oscillations connected with electromagnetic fields that are tightly localized near the surface[1,2]. When these waves are confined to finite-sized graphene, plasmon resonances emerge that are characterized by alternating charge accumulation at the opposing edges of the graphene. The resonant frequency of such a structure depends on both the size and the surface charge density, and can be electrically tuned throughout the terahertz range by applying a gate voltage[3,4]. The promise of tunable graphene THz plasmonics has yet to be fulfilled, however, because most proposed optoelectronic devices including detectors, filters, and modulators[5-10] desire near total modulation of the absorption or transmission, and require electrical contacts to the graphene -- constraints that are difficult to meet using existing plasmonic structures. We report here a new class of plasmon resonance that occurs in a hybrid graphene-metal structure. The sub-wavelength metal contacts form a capacitive grid for accumulating charge, while the narrow interleaved graphene channels, to first order, serves as a tunable inductive medium, thereby forming a structure that is resonantly-matched to an incident terahertz wave. We experimentally demonstrate resonant absorption near the theoretical maximum in readily-available, large-area graphene, ideal for THz detectors and tunable absorbers. We further predict that the use of high mobility graphene will allow resonant THz transmission near 100%, realizing a tunable THz filter or modulator. The structure is strongly coupled to incident THz radiation, and solves a fundamental problem of how to incorporate a tunable plasmonic channel into a device with electrical contacts. △ Less

Submitted 10 July, 2015; v1 submitted 18 June, 2015; originally announced June 2015.

Journal ref: Nano Lett., 2015, 15 (10), pp 7099-7104

Electron-hole transport and photovoltaic effect in gated MoS2 Schottky junctions

Authors: Marcio Fontana, Tristan Deppe, Anthony K. Boyd, Mohamed Rinzan, Amy Y. Liu, Makarand Paranjape, Paola Barbara

Abstract: Atomically thin MoS2 has recently emerged as a very attractive material for nanoscale optoelectronic devices. While n-type transport in MoS2 devices has been demonstrated, hole conduction has been more challenging. Here we show work-function engineering to be an effective approach for controlling the polarity of MoS2 devices. Gated multi-layer MoS2 transistors with Au source/drain contacts exhibit… ▽ More Atomically thin MoS2 has recently emerged as a very attractive material for nanoscale optoelectronic devices. While n-type transport in MoS2 devices has been demonstrated, hole conduction has been more challenging. Here we show work-function engineering to be an effective approach for controlling the polarity of MoS2 devices. Gated multi-layer MoS2 transistors with Au source/drain contacts exhibit n-type operation, while those with Pd contacts are shown to have p-type behavior. Devices with one Au and one Pd contact exhibit asymmetric ambipolar behavior and diode characteristics over a wide range of gate voltage, as well as a sizable photovoltaic effect. We argue that the photovoltaic effect arises from the built-in potential of the space charge accumulated at the source and drain contacts. △ Less

Submitted 9 April, 2013; v1 submitted 26 June, 2012; originally announced June 2012.

Comments: 11 pages, 5 figures

Journal ref: Sci. Rep. 3, 1634 (2013)

Unachievable Region in Precision-Recall Space and Its Effect on Empirical Evaluation

Authors: Kendrick Boyd, Vitor Santos Costa, Jesse Davis, David Page

Abstract: Precision-recall (PR) curves and the areas under them are widely used to summarize machine learning results, especially for data sets exhibiting class skew. They are often used analogously to ROC curves and the area under ROC curves. It is known that PR curves vary as class skew changes. What was not recognized before this paper is that there is a region of PR space that is completely unachievable… ▽ More Precision-recall (PR) curves and the areas under them are widely used to summarize machine learning results, especially for data sets exhibiting class skew. They are often used analogously to ROC curves and the area under ROC curves. It is known that PR curves vary as class skew changes. What was not recognized before this paper is that there is a region of PR space that is completely unachievable, and the size of this region depends only on the skew. This paper precisely characterizes the size of that region and discusses its implications for empirical evaluation methodology in machine learning. △ Less

Submitted 18 July, 2012; v1 submitted 18 June, 2012; originally announced June 2012.

Comments: ICML2012, fixed citations to use correct tech report number