The Future of Orthodontics: Deep Learning Technologies.

Deep learning has emerged as a revolutionary technical advancement in modern orthodontics, offering novel methods for diagnosis, treatment planning, and outcome prediction. Over the past 25 years, the field of dentistry has widely adopted information technology (IT), resulting in several benefits, including decreased expenses, increased efficiency, decreased need for human expertise, and reduced errors. The transition from preset rules to learning from real-world examples, particularly machine learning (ML) and artificial intelligence (AI), has greatly benefited the organization, analysis, and storage of medical data. Deep learning, a type of AI, enables robots to mimic human neural networks, allowing them to learn and make decisions independently without the need for explicit programming. Its ability to automate cephalometric analysis and enhance diagnosis through 3D imaging has revolutionized orthodontic operations. Deep learning models have the potential to significantly improve treatment outcomes and reduce human errors by accurately identifying anatomical characteristics on radiographs, thereby expediting analytical processes. Additionally, the use of 3D imaging technologies such as cone-beam computed tomography (CBCT) can facilitate precise treatment planning, allowing for comprehensive examinations of craniofacial architecture, tooth movements, and airway dimensions. In today's era of personalized medicine, deep learning's ability to customize treatments for individual patients has propelled the field of orthodontics forward tremendously. However, it is essential to address issues related to data privacy, model interpretability, and ethical considerations before orthodontic practices can use deep learning in an ethical and responsible manner. Modern orthodontics is evolving, thanks to the ability of deep learning to deliver more accurate, effective, and personalized orthodontic treatments, improving patient care as technology develops.

Simple and low-cost microscopy setup for 3D particle field measurement using incoherent illumination and open-source hardware.

The quantification of 3D particle field is of interest for a vast range of fields. While in-line particle holography (PH) can provide high-resolution measurements of particles, it suffers from speckle noise. Plenoptic imaging (PI) is less susceptible to speckle noises, but it involves a trade-off between spatial and angular resolution, rendering images with low resolution. Here, we report a simple microscopy setup with the goals of getting the strengths of both techniques. It is built with off-the-shelf and cost-effective components including a photographic lens, a diaphragm, and a CCD camera. The cost of the microscopy setup is affordable to small labs and individual researchers. The pupil plane of the proposed setup can be mechanically accessible, allowing us to implement pupil plane modulation and increase the depth of field (DOF) without requiring any additional relay lenses. It also allows us to understand the working principle of pupil plane modulation clearly, benefiting microscopy education. It illuminates the sample (particles) using diffuse white light, and thus avoids the problem of speckle noise. It captures multiple perspective images via pupil plane modulation, without requiring trading off angular and spatial resolution. We validate the setup with 2D and 3D particle samples. RESEARCH HIGHLIGHTS: We report a simple and cost-effective microscopy setup with the goals of getting the strengths of plenoptic imaging and in-line particle holography. It is built with off-the-shelf and cost-effective components. The cost of the microscopy setup is affordable to small labs and individual researchers. The pupil plane of the proposed setup can be mechanically accessible, allowing us to implement pupil plane modulation and increase the DOF without requiring any additional relay lenses. It also allows us to understand the working principle of pupil plane modulation clearly, benefiting microscopy education. It illuminates the sample (particles) using diffuse white light, and thus avoids the problem of speckle noise. It captures multiple perspective images via pupil plane modulation, without requiring trading off angular and spatial resolution. We validate the setup with 2D and 3D particle samples.

Assessing inter and intrafraction uncertainties in adult brain cancer patients using 2D/3D kV and CBCT imaging.

METHOD: 2D/3D kV imaging and CBCT data using 6 degrees of freedom (6DoF) were compared to evaluate inter and intrafraction motion. RESULTS: Results showed that intrafraction errors were low and interfraction levels were within institutional protocols. CONCLUSION: Confidence was given to use low dose 2D/3D kV imaging to confirm daily patient set up errors, and to use pre-treatment CBCT only once weekly for additional imaging information. IMPLICATIONS FOR PRACTICE: Further research is necessary to assess other uncertainties, to enable the calculation of a margin and determining the feasibility of further reduction of this.

From Imaging to Visualization: Seeing the Future of Endometriosis Care.

OBJECTIVE: To describe how the knowledge from standard imaging practices can be translated into three-dimensional visualization techniques and utilized in the surgical planning and management of endometriosis. SETTING: Tertiary care academic centre. PARTICIPANTS: Two case studies of patients with endometriosis are described. INTERVENTIONS: Transvaginal ultrasound (1), magnetic resonance imaging, three-dimensional printing (2), and three-dimensional virtual reality modeling (3) were utilized during patient workup and preparation. Three-dimensional modeling was performed by a virtual reality technician and verified for accuracy by a fellowship trained radiologist. Surgical management for endometriosis was performed. CONCLUSION: While expert transvaginal ultrasound and/or magnetic resonance imaging suffice for the majority of cases, three-dimensional printing and virtual reality modeling are a novel adjunct to standard imaging modalities. Rendering two-dimensional images into a three-dimensional representation allows users to interact with the anatomy and is particularly useful when distorted by complex pathology. These techniques contributed to improved patient understanding and experience, and helped medical learners better grasp regular imaging techniques and its translation to pelvic anatomy. Last, it augmented surgeon comprehension of the relationship between the pelvic structures, allowing for enhanced surgical planning and intraoperative decision making. Further study is being performed to quantify these effects.

2D vs. 3D Evaluation of Osteocyte Lacunae - Methodological Approaches, Recommended Parameters, and Challenges: A Narrative Review by the European Calcified Tissue Society (ECTS).

PURPOSE OF REVIEW: Quantification of the morphology of osteocyte lacunae has become a powerful tool to investigate bone metabolism, pathologies and aging. This review will provide a brief overview of 2D and 3D imaging methods for the determination of lacunar shape, orientation, density, and volume. Deviations between 2D-based and 3D-based lacunar volume estimations are often not sufficiently addressed and may give rise to contradictory findings. Thus, the systematic error arising from 2D-based estimations of lacunar volume will be discussed, and an alternative calculation proposed. Further, standardized morphological parameters and best practices for sampling and segmentation are suggested. RECENT FINDINGS: We quantified the errors in reported estimation methods of lacunar volume based on 2D cross-sections, which increase with variations in lacunar orientation and histological cutting plane. The estimations of lacunar volume based on common practice in 2D imaging methods resulted in an underestimation of lacunar volume of up to 85% compared to actual lacunar volume in an artificial dataset. For a representative estimation of lacunar size and morphology based on 2D images, at least 400 lacunae should be assessed per sample.

A comparative analysis of sphenoid and frontal sinuses using cone beam computed tomography for sex determination.

Objectives: This study aimed to evaluate linear measurements of the frontal sinus (FS) and sphenoid sinus (SS) for sex identification on cone beam computed tomography (CBCT) images. Methods: A comparative CBCT analysis was conducted on 200 full field of view (FOV) scans taken as part of routine dental investigations. Dimensions of the bilateral frontal and sphenoid sinuses were measured. Intra- and interobserver reliability were calculated. Independent t tests were used to compare the various parameters between sexes. Stepwise discriminant function analysis was used to determine sex. Additionally, the receiver operating characteristic (ROC) curve, area under the curve (AUC), sensitivity, and specificity were also determined. A p value < 0.05 was considered significant. Results: A total of 200 CBCT scans were included in the study. The mean age (±SD) among males was 25.66 (±7.11) and that among females was 24.64 (±5.12). The ROC curve revealed that the right length of the frontal sinus showed the greatest accuracy in sex identification in comparison to other linear measurements of the FS and SS. The results of our study indicated that the equation obtained from stepwise discriminant function analysis can aid in sex determination with an accuracy of 76.5 %. Conclusion: Our findings support the sexual dimorphism of linear measurements of FS and SS. There was an improvement in the accuracy of sex prediction when the linear measurements of FS and SS were considered in combination rather than in isolation. The derived equation can be an adjunctive tool for sex identification for the representative population.

The enigma of fine mobile structures on the aortic surface in a patient undergoing transcatheter aortic valve replacement: a case report.

Background: The surface of the aorta generally does not show motion unless mobile atheroma, thrombi, vegetations, or intimal flaps are present. We previously described unusual mobile filamentous structures in the carotid artery. Here, we describe similar findings in the aorta and their possible cause. Case summary: An 88-year-old female with progressive exertional dyspnoea and severe aortic stenosis had a successful transcatheter aortic valve replacement (TAVR). A filamentous structure was noted on the focused pre-operative 2D transoesophageal echocardiography in the proximal descending aorta and post-TAVR as long strand-like structures attached to the thickened intimal wall with a planar component on 3D imaging. These findings were not associated with symptoms or clinical sequelae on short- and long-term follow-up. Discussion: The mobile structures that we describe are atypical for atheroma, thrombi, vegetations, and dissections in terms of their form and clinical presentation. 2D imaging showed that the filaments had focal thickening and emerged from the aortic surface. These findings suggest a relationship with the intima, perhaps from atherogenesis or injury with disruption or lifting of the intimal surface. No clinical sequelae were detected that may also relate to their position in the descending aorta and not the arch.

Image capturing, segmentation and data analysis of shredded refuse streams.

Refuse sorting is an important cornerstone of the recycling industry, but ever-changing refuse compositions and the desire to increase recycling rates still pose many unsolved challenges. The digitalisation of refuse sorting plants promises to overcome these challenges by optimising and automatically adapting the sorting process. This publication describes a system for image capturing, segmentation-based refuse recognition and data analysis of shredded refuse streams. The image capturing collects multispectral 2D and 3D images of the refuse streams on conveyor belts. The image recognition performs a semantic segmentation of the images to determine the refuse composition from the 2D images, whereas the 3D images approximate the volumes on the conveyor belts. The semantic segmentation is done by a combined convolutional neural network model, consisting of a foreground-background and a refuse class segmentation. Both models rely on synthetic training data to reduce the necessary amount of manually labelled training data, whereas the final segmentation performance reaches an Intersection over Union of up to 75%. The results of the semantic segmentation and volume estimation are combined with data of the shredding machinery by transforming it into a unified representation. This combined dataset is the basis for estimating the processed refuse masses from the semantic segmentation and volume estimation.

Three-Dimensional Printing in Breast Reconstruction: Current and Promising Applications.

Three-dimensional (3D) printing is dramatically improving breast reconstruction by offering customized and precise interventions at various stages of the surgical process. In preoperative planning, 3D imaging techniques, such as computer-aided design, allow the creation of detailed breast models for surgical simulation, optimizing surgical outcomes and reducing complications. During surgery, 3D printing makes it possible to customize implants and precisely shape autologous tissue flaps with customized molds and scaffolds. This not only improves the aesthetic appearance, but also conforms to the patient's natural anatomy. In addition, 3D printed scaffolds facilitate tissue engineering, potentially favoring the development and integration of autologous adipose tissue, thus avoiding implant-related complications. Postoperatively, 3D imaging allows an accurate assessment of breast volume and symmetry, which is crucial in assessing the success of reconstruction. The technology is also a key educational tool, enhancing surgeon training through realistic anatomical models and surgical simulations. As the field evolves, the integration of 3D printing with emerging technologies such as biodegradable materials and advanced imaging promises to further refine breast reconstruction techniques and outcomes. This study aims to explore the various applications of 3D printing in breast reconstruction, addressing current challenges and future opportunities.

Deciphering vesicle-assisted transport mechanisms in cytoplasm to cilium trafficking.

The cilium, a pivotal organelle crucial for cell signaling and proper cell function, relies on meticulous macromolecular transport from the cytoplasm for its formation and maintenance. While the intraflagellar transport (IFT) pathway has traditionally been the focus of extensive study concerning ciliogenesis and ciliary maintenance, recent research highlights a complementary and alternative mechanism-vesicle-assisted transport (VAT) in cytoplasm to cilium trafficking. Despite its potential significance, the VAT pathway remains largely uncharacterized. This review explores recent studies providing evidence for the dynamics of vesicle-related diffusion and transport within the live primary cilium, employing high-speed super-resolution light microscopy. Additionally, we analyze the spatial distribution of vesicles in the cilium, mainly relying on electron microscopy data. By scrutinizing the VAT pathways that facilitate cargo transport into the cilium, with a specific emphasis on recent advancements and imaging data, our objective is to synthesize a comprehensive model of ciliary transport through the integration of IFT-VAT mechanisms.

Evaluation of the Impact of Digital Dentistry on the Precision of Implant Placement and Prosthesis Fabrication: An In-Vitro Study.

BACKGROUND: Digital dentistry has revolutionized the field of implant dentistry, offering enhanced accuracy and precision in implant placement and prosthesis fabrication. This study aims to evaluate the effect of digital dentistry on the accuracy of implant placement and prosthesis fit through a comprehensive in-vitro assessment. METHODS: In this in-vitro study, a Digital Dentistry Group and a Conventional Group were compared regarding implant placement accuracy and prosthesis fit. Measurements of coronal deviation, apical deviation, global deviation, angulation deviation, and depth deviation were obtained for implant placement accuracy, while marginal fit and internal fit were assessed for prosthesis fit. Statistical analysis was performed to determine significant differences between the two groups. RESULTS: The Digital Dentistry Group demonstrated significantly lower values of coronal deviation, apical deviation, global deviation, angulation deviation, and depth deviation compared to the Conventional Group (p < 0.001). Similarly, the Digital Dentistry Group exhibited superior marginal fit and internal fit (p < 0.001) when compared to the Conventional Group. CONCLUSION: This in-vitro study provides evidence supporting the superior accuracy of implant placement and improved prosthesis fit achieved through digital dentistry techniques. The use of intraoral scanners, computer-aided design/computer-aided manufacturing (CAD/CAM) systems, and three-dimensional (3D) imaging enables precise digital impressions, virtual planning, and custom-made prostheses with superior fit and esthetics. Incorporating digital dentistry into clinical practice can enhance treatment outcomes and patient satisfaction in implant dentistry.

Three-dimensional image analysis specifies the root distribution for drought avoidance in the early growth stage of rice.

BACKGROUND AND AIMS: Root system architecture (RSA) plays a key role in plant adaptation to drought because deep rooting enables better water uptake than shallow rooting under terminal drought. Understanding RSA during early plant development is essential for improving crop yields, as early drought can affect subsequent shoot growth. Herein, we demonstrate that root distribution in the topsoil significantly impacts shoot growth during the early stages of rice (Oryza sativa) development under drought, as assessed through three-dimensional (3D) image analysis. METHODS: We used 109 F12 recombinant inbred lines (RILs) obtained from a cross between shallow-rooting lowland rice and deep-rooting upland rice, representing a population with diverse RSA. We applied a moderate drought during the early development of rice grown in a plant pot (25 cm height) by stopping irrigation 14 days after sowing (DAS). Time-series RSA at 14, 21, and 28 DAS was visualized by X-ray computed tomography, and subsequently compared between drought and well-watered conditions. Following this analysis, we further investigated drought-avoidant RSA by testing 20 randomly selected RILs under drought conditions. KEY RESULTS: We inferred the root location that most influences shoot growth using a hierarchical Bayes approach: the root segment depth, which positively impacted shoot growth, ranged between 1.7-3.4 cm under drought conditions and between 0.0-1.7 cm under well-watered conditions. Drought-avoidant RILs had a higher root density in the lower layers of the topsoil compared to the others. CONCLUSIONS: Fine classification of soil layers using 3D image analysis revealed that increasing root density in the lower layers of the topsoil, rather than in the subsoil, is advantageous for drought avoidance during the early growth stage of rice.

Structured-Light 3D Imaging Based on Vector Iterative Fourier Transform Algorithm.

Quasi-continuous-phase metasurfaces overcome the side effects imposed by high-order diffraction on imaging and can impart optical parameters such as amplitude, phase, polarization, and frequency to incident light at sub-wavelength scales with high efficiency. Structured-light three-dimensional (3D) imaging is a hot topic in the field of 3D imaging because of its advantages of low computation cost, high imaging accuracy, fast imaging speed, and cost-effectiveness. Structured-light 3D imaging requires uniform diffractive optical elements (DOEs), which could be realized by quasi-continuous-phase metasurfaces. In this paper, we design a quasi-continuous-phase metasurface beam splitter through a vector iterative Fourier transform algorithm and utilize this device to realize structured-light 3D imaging of a target object with subsequent target reconstruction. A structured-light 3D imaging system is then experimentally implemented by combining the fabricated quasi-continuous-phase metasurface illuminated by the vertical-cavity surface-emitting laser and a binocular recognition system, which eventually provides a new technological path for the 3D imaging field.

Volumetric Ultrasound Imaging for the Whole Soft Tissue: Toward Enhanced Thyroid Disease Examination.

OBJECTIVES: Ultrasound imaging (USI) is the gold standard in the clinical diagnosis of thyroid diseases. Compared with two-dimensional (2D) USI, three-dimensional (3D) USI could provide more structural information. However, the unstable pressure generated by the hand-hold ultrasound probe scanning can cause tissue deformation, especially in soft tissues such as the thyroid. The deformation is manifested as tissue structure being compressed in 2D USI, which results in structural discontinuity in 3D USI. Furthermore, multiple scans apply pressure in different directions to the tissue, which will cause relative displacement between the 3D images obtained from multiple thyroid scans. METHODS: In this work, we proposed a framework to minimize the influence of the variation of pressure in thyroid 3D USI. To correct pressure artifacts in a single scanning sequence, an adaptive method to smooth the position of the 2D ultrasound (US) image sequence is adopted before performing volumetric reconstruction. To build a whole 3D US image including both sides of the thyroid gland, an iterative closest point (ICP) based registration pipeline is adopted to eliminate the relative displacement caused by different pressure directions. RESULTS: Our proposed method was validated by in vivo experiments, including healthy volunteers and volunteers with thyroid nodules at different grading levels. CONCLUSIONS: The thyroid gland and nodule are rendered intelligently in the whole scanning region to facilitate the observation of 3D USI results by the doctor. This work might make a positive contribution to the clinical diagnosis of diseases of the thyroid or other soft tissues.

A multispectral 3D live organoid imaging platform to screen probes for fluorescence guided surgery.

Achieving complete tumor resection is challenging and can be improved by real-time fluorescence-guided surgery with molecular-targeted probes. However, pre-clinical identification and validation of probes presents a lengthy process that is traditionally performed in animal models and further hampered by inter- and intra-tumoral heterogeneity in target expression. To screen multiple probes at patient scale, we developed a multispectral real-time 3D imaging platform that implements organoid technology to effectively model patient tumor heterogeneity and, importantly, healthy human tissue binding.

Novel AI-based automated virtual implant placement: Artificial versus human intelligence.

OBJECTIVES: To assess quality, clinical acceptance, time-efficiency, and consistency of a novel artificial intelligence (AI)-driven tool for automated presurgical implant planning for single tooth replacement, compared to a human intelligence (HI)-based approach. MATERIALS AND METHODS: To validate a novel AI-driven implant placement tool, a dataset of 10 time-matching cone beam computed tomography (CBCT) scans and intra-oral scans (IOS) previously acquired for single mandibular molar/premolar implant placement was included. An AI pre-trained model for implant planning was compared to human expert-based planning, followed by the export, evaluation and comparison of two generic implants-AI-generated and human-generated-for each case. The quality of both approaches was assessed by 12 calibrated dentists through blinded observations using a visual analogue scale (VAS), while clinical acceptance was evaluated through an AI versus HI battle (Turing test). Subsequently, time efficiency and consistency were evaluated and compared between both planning methods. RESULTS: Overall, 360 observations were gathered, with 240 dedicated to VAS, of which 95 % (AI) and 96 % (HI) required no major, clinically relevant corrections. In the AI versus HI Turing test (120 observations), 4 cases had matching judgments for AI and HI, with AI favoured in 3 and HI in 3. Additionally, AI completed planning more than twice as fast as HI, taking only 198 ± 33 s compared to 435 ± 92 s (p < 0.05). Furthermore, AI demonstrated higher consistency with zero-degree median surface deviation (MSD) compared to HI (MSD=0.3 ± 0.17 mm). CONCLUSION: AI demonstrated expert-quality and clinically acceptable single-implant planning, proving to be more time-efficient and consistent than the HI-based approach. CLINICAL SIGNIFICANCE: Presurgical implant planning often requires multidisciplinary collaboration between highly experienced specialists, which can be complex, cumbersome and time-consuming. However, AI-driven implant planning has the potential to allow clinically acceptable planning, significantly more time-efficient and consistent than the human expert.

Understanding the Interplay Between Skin, Fascia, and Muscles of the Midface in Facial Aging.

BACKGROUND: Surgical, minimally-invasive, and non-invasive aesthetic procedures try to ameliorate the signs of facial aging, but also focus on enhancing various individual features of beauty in each patient. Herein, the midface plays a central role due to its location but also its importance for the aesthetic perception and facial expression. OBJECTIVE: To date, no study has investigated the interplay between facial muscles and its connecting subdermal architecture during facial aging to provide a more comprehensive understanding of the middle face. MATERIALS AND METHODS: A total of 76 subjects, consisting of 30 males (39.5%) and 46 females (60.5%) with a mean age of 42.2 (18.7) years [range 19-80] and a mean BMI of 24.6 (3.7) kg/m2 [range 18-35], were enrolled in this investigation. Cutometry (skin aging), 3D skin displacement analyses (subdermal connective tissue aging), and sEMG (muscle aging) analyses were utilized. RESULTS: The results revealed that overall skin firmness increased, and skin elasticity decreased (p < 0.001), sEMG signal of the investigated muscles decreased (p < 0.001), whereas midfacial mobility remained unaltered (p = 0.722). CONCLUSION: The results of this study indicate that midfacial aging is a measurable effect when utilizing individual measurement modalities for assessing skin, subdermal fascia, and midfacial muscles. The function of midfacial muscles revealed a potential threshold effect, which is not reached during midfacial aging due to the unchanged soft tissue mobility at older age. However, to understand its clinical presentation all midfacial soft tissues need to be factored in and a holistic picture needs to be created. NO LEVEL ASSIGNED: This journal requires that authors assign a level of evidence to each submission to which Evidence-Based Medicine rankings are applicable. This excludes review articles, book reviews, and manuscripts that concern Basic Science, Animal Studies, Cadaver Studies, and Experimental Studies. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to authors www.springer.com/00266 .

Three-dimensional video recordings: Accuracy, reliability, clinical and research guidelines - Reliability assessment of a 4D camera.

OBJECTIVES: In addition to studying facial anatomy, stereophotogrammetry is an efficient diagnostic tool for assessing facial expressions through 3D video recordings. Current technology produces high-quality recordings but also generates extremely excessive data. Here, we compare various recording speeds for three standardized movements using the 3dMDface camera system, to assess its accuracy and reliability. MATERIALS AND METHODS: A linear and two circular movements were performed using a 3D-printed cube mounted on a robotic arm. All movements were recorded initially at 60 fps (frames/second) and then at 30 and 15 fps. Recording accuracy was tested with best-fit superimpositions of consecutive frames of the 3D cube and calculation of the Mean Absolute Distance (MAD). The reliability of the recordings were tested with evaluation of the inter- and intra-examiner error. RESULTS: The accuracy of movement recordings was excellent at all speeds (60, 30 and 15 fps), with variability in MAD values consistently being less than 1 mm. The reliability of the camera recordings was excellent at all recording speeds. CONCLUSIONS: This study demonstrated that 3D recordings of facial expressions can be performed at 30 or even at 15 fps without significant loss of information. This considerably reduces the amount of produced data facilitating further processing and analyses.

Utility of White Light Scanning as an Alternative to Computed Tomography to Evaluate Severity of Pectus Excavatum Deformity.

BACKGROUND: Pectus excavatum (PE) severity and surgical candidacy are determined by computed tomography (CT)-delineated Haller Index (HI) and Correction Index (CI). White light scanning (WLS) has been proposed as a non-ionizing alternative. The purpose of this retrospective study is to create models to determine PE severity using WLS as a non-ionizing alternative to CT. METHODS: Between November 2015 and February 2023, CT and WLS were performed for children ≤18 years undergoing evaluation at a high-volume, chest-wall deformity clinic. Separate quadratic discriminate analysis models were developed to predict CT HI ≥ 3.25 and CT CI ≥ 28% indicating surgical candidacy. Two bootstrap forest models were trained on WLS measurements and patient demographics to predict CT HI and CT CI values then compared to actual index values by intraclass correlation coefficient (ICC). RESULTS: In total, 242 patients were enrolled (86.4% male, mean [SD] age 15.2 [1.3] years). Quadratic discriminate analysis models predicted CT HI ≥ 3.25 with specificity = 91.7%, PPV = 97.7% (AUC = 0.91), and CT CI ≥ 28% with specificity = 92.3%, PPV = 93.5% (AUC = 0.84). Bootstrap forest model predicted CT HI with training dataset ICC (95% CI) = 0.91 (0.88-0.93, R2 = 0.85) and test dataset ICC (95% CI) = 0.86 (0.71-0.94, R2 = 0.77). For CT CI, training dataset ICC (95% CI) = 0.91 (0.81-0.93, R2 = 0.86) and test dataset ICC (95% CI) = 0.75 (0.50-0.88, R2 = 0.63). CONCLUSIONS: Using noninvasive and nonionizing WLS imaging, we can predict PE severity at surgical threshold with high specificity obviating the need for CT. Furthermore, we can predict actual CT HI and CI with moderate-excellent reliability. We anticipate this point-of-care tool to obviate the need for most cross-sectional imaging during surgical evaluation of PE. LEVEL OF EVIDENCE: Level III. STUDY TYPE: Study of Diagnostic Test.