RESUMEN
AI-powered segmentation of hip and knee bony anatomy has revolutionized orthopedics, transforming pre-operative planning and post-operative assessment. Despite the remarkable advancements in AI algorithms for medical imaging, the potential for biases inherent within these models remains largely unexplored. This study tackles these concerns by thoroughly re-examining AI-driven segmentation for hip and knee bony anatomy. While advanced imaging modalities like CT and MRI offer comprehensive views, plain radiographs (X-rays) predominate the standard initial clinical assessment due to their widespread availability, low cost, and rapid acquisition. Hence, we focused on plain radiographs to ensure the utilization of our contribution in diverse healthcare settings, including those with limited access to advanced imaging technologies. This work provides insights into the underlying causes of biases in AI-based knee and hip image segmentation through an extensive evaluation, presenting targeted mitigation strategies to alleviate biases related to sex, race, and age, using an automatic segmentation that is fair, impartial, and safe in the context of AI. Our contribution can enhance inclusivity, ethical practices, equity, and an unbiased healthcare environment with advanced clinical outcomes, aiding decision-making and osteoarthritis research. Furthermore, we have made all the codes and datasets publicly and freely accessible to promote open scientific research.
Asunto(s)
Inteligencia Artificial , Humanos , Masculino , Femenino , Persona de Mediana Edad , Procesamiento de Imagen Asistido por Computador/métodos , Sesgo , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Adulto , Algoritmos , Articulación de la Cadera/diagnóstico por imagen , Imagen por Resonancia Magnética/métodos , Anciano , Tomografía Computarizada por Rayos X/métodos , OrtopediaRESUMEN
INTRODUCTION/AIM: To evaluate whether patients with haemophilia (PwH) can be enabled to perform ultrasonography (US) of their knees without supervision according to the Haemophilia Early Arthropathy Detection with Ultrasound (HEAD-US) protocol and whether they would be able to recognize pathologies. METHODS: Five PwH (mean age 29.6 years, range 20-48 years) were taught the use of a portable US device and the HEAD-US protocol. Subsequently, the patients performed US unsupervised at home three times a week for a total of 6 weeks with a reteaching after 2 weeks. All images were checked for mapping of the landmarks defined in the HEAD-US protocol by a radiologist. In a final test after the completion of the self-sonography period, participants were asked to identify scanning plane and potential pathology from US images of other PwH. RESULTS: On the images of the self-performed scans, 82.7% of the possible anatomic landmarks could be identified and 67.5% of the requested images were unobjectionable, depicting 100% of the required landmarks. There was a highly significant improvement in image quality following reteaching after 2 weeks (74.80 ± 36.88% vs. 88.31 ± 19.87%, p < .001). In the final test, the participants identified the right scanning plane in 85.0% and they correctly identified pathology in 90.0% of images. CONCLUSION: Appropriately trained PwH can perform the HEAD-US protocol of their knee with high quality and are capable to identify pathologic findings on these standardized images. Asynchronous tele-sonography could enable early therapy adjustment and thereby possibly reduce costs.
Asunto(s)
Estudios de Factibilidad , Hemofilia A , Ultrasonografía , Humanos , Hemofilia A/complicaciones , Hemofilia A/diagnóstico por imagen , Ultrasonografía/métodos , Adulto , Persona de Mediana Edad , Masculino , Adulto Joven , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagenRESUMEN
Biomechanical modeling of the knee during motion is a pivotal component in disease treatment, implant designs, and rehabilitation strategies. Historically, dynamic simulations of the knee have been scant. This study uniquely integrates a dual fluoroscopic imaging system (DFIS) to investigate the in vivo dynamic behavior of the meniscus during functional activities using a finite element (FE) model. The model was subsequently validated through experiments. Motion capture of a single-leg lunge was executed by DFIS. The motion model was reconstructed using 2D-to-3D registration in conjunction with computed tomography (CT) scans. Both CT and magnetic resonance imaging (MRI) data facilitated the development of the knee FE model. In vivo knee displacements and rotations were utilized as driving conditions for the FE model. Moreover, a 3D-printed model, accompanied with digital imaging correlation (DIC), was used to evaluate the accuracy of the FE model. To a better inner view of knees during the DIC analysis, tibia and femur were crafted by transparent resin. The availability of the FE model was guaranteed by the similar strain distribution of the DIC and FE simulation. Subsequent modeling revealed that the compressive stress distribution between the medial and lateral menisci was balanced in the standing posture. As the flexion angle increased, the medial meniscus bore the primary compressive load, with peak stresses occurring between 60 and 80° of flexion. The simulation of a healthy knee provides a critical theoretical foundation for addressing knee pathologies and advancing prosthetic designs.
Asunto(s)
Análisis de Elementos Finitos , Rodilla , Fenómenos Biomecánicos , Humanos , Rodilla/fisiología , Rodilla/diagnóstico por imagen , Fenómenos Mecánicos , Tomografía Computarizada por Rayos X , Movimiento , Articulación de la Rodilla/fisiología , Articulación de la Rodilla/diagnóstico por imagenRESUMEN
Neoplastic and non-neoplastic soft tissue masses around the knee are often incidental findings. Most of these lesions are benign with typical imaging characteristics that allow a confident diagnosis. However, some of these incidental neoplastic masses are characterized by morbidity and potential mortality. This review highlights the typical aspects of these lesions, facilitating a correct diagnosis.
Asunto(s)
Neoplasias de los Tejidos Blandos , Humanos , Neoplasias de los Tejidos Blandos/diagnóstico por imagen , Diagnóstico Diferencial , Rodilla/diagnóstico por imagen , Imagen por Resonancia Magnética/métodos , Articulación de la Rodilla/diagnóstico por imagen , Articulación de la Rodilla/patologíaRESUMEN
Numerous anatomical variants are described around the knee, many of which look like bony lesions, so it is important to know them to avoid unnecessary complementary tests and inadequate management. Likewise, several alterations in relation to normal development can also simulate bone lesions.However, numerous pathologic processes frequently affect the knee, including traumatic, inflammatory, infectious, and tumor pathology. Many of these entities show typical radiologic features that facilitate their diagnosis. In other cases, a correct differential diagnosis is necessary for proper clinical management.Despite the availability of increasingly advanced imaging techniques, plain radiography is still the technique of choice in the initial study of many of these pathologies. This article reviews the radiologic characteristics of tumor and nontumor lesions that may appear around the knee to make a correct diagnosis and avoid unnecessary complementary radiologic examinations and inadequate clinical management.
Asunto(s)
Enfermedades Óseas , Neoplasias Óseas , Humanos , Neoplasias Óseas/diagnóstico por imagen , Diagnóstico Diferencial , Enfermedades Óseas/diagnóstico por imagen , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Imagen por Resonancia Magnética/métodosRESUMEN
BACKGROUND: Recent Convolutional Neural Networks (CNNs) perform low-error reconstruction in fast Magnetic Resonance Imaging (MRI). Most of them convolve the image with kernels and successfully explore the local information. Nonetheless, the non-local image information, which is embedded among image patches relatively far from each other, may be lost due to the limitation of the receptive field of the convolution kernel. We aim to incorporate a graph to represent non-local information and improve the reconstructed images by using the Graph Convolutional Enhanced Self-Similarity (GCESS) network. METHODS: First, the image is reconstructed into the graph to extract the non-local self-similarity in the image. Second, GCESS uses spatial convolution and graph convolution to process the information in the image, so that local and non-local information can be effectively utilized. The network strengthens the non-local similarity between similar image patches while reconstructing images, making the reconstruction of structure more reliable. RESULTS: Experimental results on in vivo knee and brain data demonstrate that the proposed method achieves better artifact suppression and detail preservation than state-of-the-art methods, both visually and quantitatively. Under 1D Cartesian sampling with 4 × acceleration (AF = 4), the PSNR of knee data reached 34.19 dB, 1.05 dB higher than that of the compared methods; the SSIM achieved 0.8994, 2% higher than the compared methods. Similar results were obtained for the reconstructed images under other sampling templates as demonstrated in our experiment. CONCLUSIONS: The proposed method successfully constructs a hybrid graph convolution and spatial convolution network to reconstruct images. This method, through its training process, amplifies the non-local self-similarities, significantly benefiting the structural integrity of the reconstructed images. Experiments demonstrate that the proposed method outperforms the state-of-the-art reconstruction method in suppressing artifacts, as well as in preserving image details.
Asunto(s)
Encéfalo , Procesamiento de Imagen Asistido por Computador , Imagen por Resonancia Magnética , Redes Neurales de la Computación , Imagen por Resonancia Magnética/métodos , Humanos , Encéfalo/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Rodilla/diagnóstico por imagen , Algoritmos , ArtefactosRESUMEN
OBJECTIVES: This study addresses the critical need for accurate summarization in radiology by comparing various Large Language Model (LLM)-based approaches for automatic summary generation. With the increasing volume of patient information, accurately and concisely conveying radiological findings becomes crucial for effective clinical decision-making. Minor inaccuracies in summaries can lead to significant consequences, highlighting the need for reliable automated summarization tools. METHODS: We employed two language models - Text-to-Text Transfer Transformer (T5) and Bidirectional and Auto-Regressive Transformers (BART) - in both fine-tuned and zero-shot learning scenarios and compared them with a Recurrent Neural Network (RNN). Additionally, we conducted a comparative analysis of 100 MRI report summaries, using expert human judgment and criteria such as coherence, relevance, fluency, and consistency, to evaluate the models against the original radiologist summaries. To facilitate this, we compiled a dataset of 15,508 retrospective knee Magnetic Resonance Imaging (MRI) reports from our Radiology Information System (RIS), focusing on the findings section to predict the radiologist's summary. RESULTS: The fine-tuned models outperform the neural network and show superior performance in the zero-shot variant. Specifically, the T5 model achieved a Rouge-L score of 0.638. Based on the radiologist readers' study, the summaries produced by this model were found to be very similar to those produced by a radiologist, with about 70% similarity in fluency and consistency between the T5-generated summaries and the original ones. CONCLUSIONS: Technological advances, especially in NLP and LLM, hold great promise for improving and streamlining the summarization of radiological findings, thus providing valuable assistance to radiologists in their work.
Asunto(s)
Estudios de Factibilidad , Imagen por Resonancia Magnética , Procesamiento de Lenguaje Natural , Redes Neurales de la Computación , Humanos , Sistemas de Información Radiológica , Rodilla/diagnóstico por imagen , Estudios RetrospectivosRESUMEN
Magnetic resonance imaging produces detailed anatomical and physiological images of the human body that can be used in the clinical diagnosis and treatment of diseases. However, MRI suffers its comparatively longer acquisition time than other imaging methods and is thus vulnerable to motion artifacts, which ultimately lead to likely failed or even wrong diagnosis. In order to perform faster reconstruction, deep learning-based methods along with traditional strategies such as parallel imaging and compressed sensing come into play in recent years in this field. Meanwhile, in order to better analyze the diseases, it is also often necessary to acquire images in the same region of interest under different modalities, which yield images with different contrast levels. However, most of these aforementioned methods tend to use single-modal images for reconstruction, neglecting the correlation and redundancy information embedded in MR images acquired with different modalities. While there are works on multi-modal reconstruction, the information is yet to be efficiently explored. In this paper, we propose an end-to-end neural network called MLMFNet, which helps the reconstruction of the target modality by using information from the auxiliary modality across feature channels and layers. Specifically, this is highlighted by three components: (I) An encoder based on UNet with a single-stream strategy that fuses auxiliary and target modalities; (II) a decoder that tends to multi-level features from all layers of the encoder, and (III) a channel attention module. Quantitative and qualitative analyses are performed on a public brain dataset and knee brain dataset, which show that the proposed method achieves satisfying results in MRI reconstruction within the multi-modal context, and also demonstrate its effectiveness and potential to be used in clinical practice.
Asunto(s)
Algoritmos , Encéfalo , Aprendizaje Profundo , Procesamiento de Imagen Asistido por Computador , Imagen por Resonancia Magnética , Humanos , Imagen por Resonancia Magnética/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Encéfalo/diagnóstico por imagen , Redes Neurales de la Computación , Imagen Multimodal/métodos , Rodilla/diagnóstico por imagenRESUMEN
Kinematic tracking of native anatomy from stereo-radiography provides a quantitative basis for evaluating human movement. Conventional tracking procedures require significant manual effort and call for acquisition and annotation of subject-specific volumetric medical images. The current work introduces a framework for fully automatic tracking of native knee anatomy from dynamic stereo-radiography which forgoes reliance on volumetric scans. The method consists of three computational steps. First, captured radiographs are annotated with segmentation maps and anatomic landmarks using a convolutional neural network. Next, a non-convex polynomial optimization problem formulated from annotated landmarks is solved to acquire preliminary anatomy and pose estimates. Finally, a global optimization routine is performed for concurrent refinement of anatomy and pose. An objective function is maximized which quantifies similarities between masked radiographs and digitally reconstructed radiographs produced from statistical shape and intensity models. The proposed framework was evaluated against manually tracked trials comprising dynamic activities, and additional frames capturing a static knee phantom. Experiments revealed anatomic surface errors routinely below 1.0 mm in both evaluation cohorts. Median absolute errors of individual bone pose estimates were below 1.0 ∘ or mm for 15 out of 18 degrees of freedom in both evaluation cohorts. Results indicate that accurate pose estimation of native anatomy from stereo-radiography may be performed with significantly reduced manual effort, and without reliance on volumetric scans.
Asunto(s)
Rodilla , Humanos , Rodilla/diagnóstico por imagen , Rodilla/anatomía & histología , Rodilla/fisiología , Articulación de la Rodilla/diagnóstico por imagen , Articulación de la Rodilla/anatomía & histología , Articulación de la Rodilla/fisiología , Fantasmas de Imagen , Radiografía , Modelos EstadísticosRESUMEN
Tendon properties impact human locomotion, influencing sports performance, and injury prevention. Hamstrings play a crucial role in sprinting, particularly the biceps femoris long head (BFlh), which is prone to frequent injuries. It remains uncertain if BFlh exhibits distinct mechanical properties compared to other hamstring muscles. This study utilized free-hand three-dimensional ultrasound to assess morphological and mechanical properties of distal hamstrings tendons in 15 men. Scans were taken in prone position, with hip and knee extended, at rest and during 20%, 40%, 60%, and 80% of maximal voluntary isometric contraction of the knee flexors. Tendon length, volume, cross-sectional area (CSA), and anteroposterior (AP) and mediolateral (ML) widths were quantified at three locations. Longitudinal and transverse deformations, stiffness, strain, and stress were estimated. The ST had the greatest tendon strain and the lowest stiffness as well as the highest CSA and AP and ML width strain compared to other tendons. Biceps femoris short head (BFsh) exhibited the least strain, AP and ML deformation. Further, BFlh displayed the highest stiffness and stress, and BFsh had the lowest stress. Additionally, deformation varied by region, with the proximal site showing generally the lowest CSA strain. Distal tendon mechanical properties differed among the hamstring muscles during isometric knee flexions. In contrast to other bi-articular hamstrings, the BFlh high stiffness and stress may result in greater energy absorption by its muscle fascicles, rather than the distal tendon, during late swing in sprinting. This could partly account for the increased incidence of hamstring injuries in this muscle.
Asunto(s)
Músculos Isquiosurales , Músculo Esquelético , Masculino , Humanos , Músculo Esquelético/fisiología , Tendones/diagnóstico por imagen , Tendones/fisiología , Músculos Isquiosurales/fisiología , Rodilla/diagnóstico por imagen , Rodilla/fisiología , Contracción Isométrica/fisiología , UltrasonografíaRESUMEN
This paper introduces a method for measuring 3D tibiofemoral kinematics using a multi-channel A-mode ultrasound system under dynamic conditions. The proposed system consists of a multi-channel A-mode ultrasound system integrated with a conventional motion capture system (i.e., optical tracking system). This approach allows for the non-invasive and non-radiative quantification of the tibiofemoral joint's six degrees of freedom (DOF). We demonstrated the feasibility and accuracy of this method in the cadaveric experiment. The knee joint's motions were mimicked by manually manipulating the leg through multiple motion cycles from flexion to extension. To measure it, six custom ultrasound holders, equipped with a total of 30 A-mode ultrasound transducers and 18 optical markers, were mounted on various anatomical regions of the lower extremity of the specimen. During experiments, 3D-tracked intra-cortical bone pins were inserted into the femur and tibia to measure the ground truth of tibiofemoral kinematics. The results were compared with the tibiofemoral kinematics derived from the proposed ultrasound system. The results showed an average rotational error of 1.51 ± 1.13° and a translational error of 3.14 ± 1.72 mm for the ultrasound-derived kinematics, compared to the ground truth. In conclusion, this multi-channel A-mode ultrasound system demonstrated a great potential of effectively measuring tibiofemoral kinematics during dynamic motions. Its improved accuracy, nature of non-invasiveness, and lack of radiation exposure make this method a promising alternative to incorporate into gait analysis and prosthetic kinematic measurements later.
Asunto(s)
Imagenología Tridimensional , Articulación de la Rodilla , Ultrasonografía , Humanos , Fenómenos Biomecánicos , Articulación de la Rodilla/fisiología , Articulación de la Rodilla/diagnóstico por imagen , Ultrasonografía/métodos , Imagenología Tridimensional/métodos , Tibia/diagnóstico por imagen , Tibia/fisiología , Rango del Movimiento Articular/fisiología , Fémur/fisiología , Fémur/diagnóstico por imagen , Rodilla/fisiología , Rodilla/diagnóstico por imagenRESUMEN
Precise measurement of joint-level motion from stereo-radiography facilitates understanding of human movement. Conventional procedures for kinematic tracking require significant manual effort and are time intensive. The current work introduces a method for fully automatic tracking of native knee kinematics from stereo-radiography sequences. The framework consists of three computational steps. First, biplanar radiograph frames are annotated with segmentation maps and key points using a convolutional neural network. Next, initial bone pose estimates are acquired by solving a polynomial optimization problem constructed from annotated key points and anatomic landmarks from digitized models. A semidefinite relaxation is formulated to realize the global minimum of the non-convex problem. Pose estimates are then refined by registering computed tomography-based digitally reconstructed radiographs to masked radiographs. A novel rendering method is also introduced which enables generating digitally reconstructed radiographs from computed tomography scans with inconsistent slice widths. The automatic tracking framework was evaluated with stereo-radiography trials manually tracked with model-image registration, and with frames which capture a synthetic leg phantom. The tracking method produced pose estimates which were consistently similar to manually tracked values; and demonstrated pose errors below 1.0 degree or millimeter for all femur and tibia degrees of freedom in phantom trials. Results indicate the described framework may benefit orthopaedics and biomechanics applications through acceleration of kinematic tracking.
Asunto(s)
Articulación de la Rodilla , Rodilla , Humanos , Fenómenos Biomecánicos , Radiografía , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Tomografía Computarizada por Rayos X/métodos , Imagenología Tridimensional/métodosRESUMEN
PURPOSE: To develop a novel deep learning-based method inheriting the advantages of data distribution prior and end-to-end training for accelerating MRI. METHODS: Langevin dynamics is used to formulate image reconstruction with data distribution before facilitate image reconstruction. The data distribution prior is learned implicitly through the end-to-end adversarial training to mitigate the hyper-parameter selection and shorten the testing time compared to traditional probabilistic reconstruction. By seamlessly integrating the deep equilibrium model, the iteration of Langevin dynamics culminates in convergence to a fix-point, ensuring the stability of the learned distribution. RESULTS: The feasibility of the proposed method is evaluated on the brain and knee datasets. Retrospective results with uniform and random masks show that the proposed method demonstrates superior performance both quantitatively and qualitatively than the state-of-the-art. CONCLUSION: The proposed method incorporating Langevin dynamics with end-to-end adversarial training facilitates efficient and robust reconstruction for MRI. Empirical evaluations conducted on brain and knee datasets compellingly demonstrate the superior performance of the proposed method in terms of artifact removing and detail preserving.
Asunto(s)
Algoritmos , Encéfalo , Procesamiento de Imagen Asistido por Computador , Rodilla , Imagen por Resonancia Magnética , Humanos , Imagen por Resonancia Magnética/métodos , Encéfalo/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Rodilla/diagnóstico por imagen , Aprendizaje Profundo , Estudios Retrospectivos , ArtefactosRESUMEN
Objective.Task-adapted image reconstruction methods using end-to-end trainable neural networks (NNs) have been proposed to optimize reconstruction for subsequent processing tasks, such as segmentation. However, their training typically requires considerable hardware resources and thus, only relatively simple building blocks, e.g. U-Nets, are typically used, which, albeit powerful, do not integrate model-specific knowledge.Approach.In this work, we extend an end-to-end trainable task-adapted image reconstruction method for a clinically realistic reconstruction and segmentation problem of bone and cartilage in 3D knee MRI by incorporating statistical shape models (SSMs). The SSMs model the prior information and help to regularize the segmentation maps as a final post-processing step. We compare the proposed method to a simultaneous multitask learning approach for image reconstruction and segmentation (MTL) and to a complex SSMs-informed segmentation pipeline (SIS).Main results.Our experiments show that the combination of joint end-to-end training and SSMs to further regularize the segmentation maps obtained by MTL highly improves the results, especially in terms of mean and maximal surface errors. In particular, we achieve the segmentation quality of SIS and, at the same time, a substantial model reduction that yields a five-fold decimation in model parameters and a computational speedup of an order of magnitude.Significance.Remarkably, even for undersampling factors of up toR= 8, the obtained segmentation maps are of comparable quality to those obtained by SIS from ground-truth images.
Asunto(s)
Aprendizaje Profundo , Imagenología Tridimensional , Imagen por Resonancia Magnética , Imagen por Resonancia Magnética/métodos , Humanos , Imagenología Tridimensional/métodos , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagenRESUMEN
PURPOSE: This paper proposes a novel self-supervised learning framework that uses model reinforcement, REference-free LAtent map eXtraction with MOdel REinforcement (RELAX-MORE), for accelerated quantitative MRI (qMRI) reconstruction. The proposed method uses an optimization algorithm to unroll an iterative model-based qMRI reconstruction into a deep learning framework, enabling accelerated MR parameter maps that are highly accurate and robust. METHODS: Unlike conventional deep learning methods which require large amounts of training data, RELAX-MORE is a subject-specific method that can be trained on single-subject data through self-supervised learning, making it accessible and practically applicable to many qMRI studies. Using quantitative T 1 $$ {\mathrm{T}}_1 $$ mapping as an example, the proposed method was applied to the brain, knee and phantom data. RESULTS: The proposed method generates high-quality MR parameter maps that correct for image artifacts, removes noise, and recovers image features in regions of imperfect image conditions. Compared with other state-of-the-art conventional and deep learning methods, RELAX-MORE significantly improves efficiency, accuracy, robustness, and generalizability for rapid MR parameter mapping. CONCLUSION: This work demonstrates the feasibility of a new self-supervised learning method for rapid MR parameter mapping, that is readily adaptable to the clinical translation of qMRI.
Asunto(s)
Algoritmos , Encéfalo , Aprendizaje Profundo , Procesamiento de Imagen Asistido por Computador , Imagen por Resonancia Magnética , Fantasmas de Imagen , Imagen por Resonancia Magnética/métodos , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Encéfalo/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Artefactos , Aprendizaje Automático SupervisadoRESUMEN
The deep learning method is an efficient solution for improving the quality of undersampled magnetic resonance (MR) image reconstruction while reducing lengthy data acquisition. Most deep learning methods neglect the mutual constraints between the real and imaginary components of complex-valued k-space data. In this paper, a new complex-valued convolutional neural network, namely, Dense-U-Dense Net (DUD-Net), is proposed to interpolate the undersampled k-space data and reconstruct MR images. The proposed network comprises dense layers, U-Net, and other dense layers in sequence. The dense layers are used to simulate the mutual constraints between real and imaginary components, and U-Net performs feature sparsity and interpolation estimation for k-space data. Two MRI datasets were used to evaluate the proposed method: brain magnitude-only MR images and knee complex-valued k-space data. Several operations were conducted for data preprocessing. First, the complex-valued MR images were synthesized by phase modulation on magnitude-only images. Second, a radial trajectory based on the golden angle was used for k-space undersampling, whereby a reversible normalization method was proposed to balance the distribution of positive and negative values in k-space data. The optimal performance of DUD-Net was demonstrated based on a quantitative evaluation of inter-method and intra-method comparisons. When compared with other methods, significant improvements were achieved, PSNRs were increased by 10.78 and 5.74dB, whereas RMSEs were decreased by 71.53% and 30.31% for magnitude and phase image, respectively. It is concluded that DUD-Net significantly improves the performance of MR image reconstruction.
Asunto(s)
Encéfalo , Aprendizaje Profundo , Procesamiento de Imagen Asistido por Computador , Rodilla , Imagen por Resonancia Magnética , Humanos , Imagen por Resonancia Magnética/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Encéfalo/diagnóstico por imagen , Rodilla/diagnóstico por imagen , AlgoritmosRESUMEN
BACKGROUND: T2 mapping of short-T2 tissues in the knee (meniscus, tendon, and ligament) is needed to aid the clinical MRI knee diagnosis, which is hard to realize using traditional clinical methods. PURPOSE: To accelerate the acquisition of T2 values for short-T2 tissues in the knee by analyzing the signal equation of balanced steady-state free precession (bSSFP) sequence in MRI. METHODS: Effect of half-radial acquisition on pixel bandwidth was analyzed mathematically. A modified 3D radial dual-echo bSSFP sequence was proposed for 0.53 mm isotropic resolution knee imaging with 2 different TEs at 3 T, which alleviated the problem of off-resonance artifacts caused by traditional half-radial acquisition scheme. A novel pixel-based optimization method was proposed for efficient T2 mapping of short-T2 tissues in the knee given off-resonance values. Simulation was conducted to evaluate the sensitivity of the proposed method to other parameters. Phantom results were compared with 2D spin-echo (SE), and in vivo results were compared with SE and previously studies. RESULTS: Simulation showed that the proposed method is insensitive to T1 and B1 variations (estimation error < 1% for T1/B1 error of ±90%), avoiding the need for separated T1 and B1 scans. High isotropic resolution knee imaging was achieved using the modified dual-echo bSSFP. The total scan time was within 3.5 min, including a separate off-resonance scan for T2 measurement. Measured mean T2 values for phantoms correlated well with SE (R2 = 0.99), and no significant difference was observed (P = 0.45). In vivo meniscus T2 measurements and ligament T2 measurements agreed with the literature, while tendon T2 measurements were much lower (31.7% lower for patellar tendon, and 13.5% lower for quadriceps tendon), which might result in its bi-component property. CONCLUSIONS: The proposed method provides an efficient way for fast, robust, high-resolution imaging and T2 mapping of short-T2 tissues in the knee.
Asunto(s)
Imagenología Tridimensional , Ligamento Rotuliano , Humanos , Imagenología Tridimensional/métodos , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Imagen por Resonancia Magnética/métodos , Fantasmas de ImagenRESUMEN
BACKGROUND: Magnetic resonance imaging (MRI) is the preferred imaging modality for diagnosing knee disease. Segmentation of the knee MRI images is essential for subsequent quantification of clinical parameters and treatment planning for knee prosthesis replacement. However, the segmentation remains difficult due to individual differences in anatomy, the difficulty of obtaining accurate edges at lower resolutions, and the presence of speckle noise and artifacts in the images. In addition, radiologists must manually measure the knee's parameters which is a laborious and time-consuming process. PURPOSE: Automatic quantification of femoral morphological parameters can be of fundamental help in the design of prosthetic implants for the repair of the knee and the femur. Knowledge of knee femoral parameters can provide a basis for femoral repair of the knee, the design of fixation materials for femoral prostheses, and the replacement of prostheses. METHODS: This paper proposes a new deep network architecture to comprehensively address these challenges. A dual output model structure is proposed, with a high and low layer fusion extraction feature module designed to extract rich features through the cross-fusion mechanism. A multi-scale edge information extraction spatial feature module is also developed to address the boundary-blurring problem. RESULTS: Based on the precise automated segmentation results, 10 key clinical parameters were automatically measured for a knee femoral prosthesis replacement program. The correlation coefficients of the quantitative results of these parameters compared to manual results all achieved at least 0.92. The proposed method was extensively evaluated with MRIs of 78 patients' knees, and it consistently outperformed other methods used for segmentation. CONCLUSIONS: The automated quantization process produced comparable measurements to those manually obtained by radiologists. This paper demonstrates the viability of automatic knee MRI image segmentation and quantitative analysis with the proposed method. This provides data to support the accuracy of assessing the progression and biomechanical changes of osteoarthritis of the knee using an automated process, thus saving valuable time for the radiologists and surgeons.
Asunto(s)
Procesamiento de Imagen Asistido por Computador , Articulación de la Rodilla , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Articulación de la Rodilla/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Imagen por Resonancia Magnética/métodos , Fémur/diagnóstico por imagenRESUMEN
OBJECTIVE: The Dixon method is frequently employed in clinical and scientific research for fat suppression, because it has lower sensitivity to static magnetic field inhomogeneity compared to chemical shift selective saturation or its variants and maintains image signal-to-noise ratio (SNR). Recently, research on very-low-field (VLF < 100 mT) magnetic resonance imaging (MRI) has regained popularity. However, there is limited literature on water-fat separation in VLF MRI. Here, we present a modified two-point Dixon method specifically designed for VLF MRI. METHODS: Most experiments were performed on a homemade 50 mT portable MRI scanner. The receiving coil adopted a homemade quadrature receiving coil. The data were acquired using spin-echo and gradient-echo sequences. We considered the T2* effect, and added priori information to existing two-point Dixon method. Then, the method used regional iterative phasor extraction (RIPE) to extract the error phasor. Finally, least squares solutions for water and fat were obtained and fat signal fraction was calculated. RESULTS: For phantom evaluation, water-only and fat-only images were obtained and the local fat signal fractions were calculated, with two samples being 0.94 and 0.93, respectively. For knee imaging, cartilage, muscle and fat could be clearly distinguished. The water-only images were able to highlight areas such as cartilage that could not be easily distinguished without separation. CONCLUSION: This work has demonstrated the feasibility of using a 50 mT MRI scanner for water-fat separation. SIGNIFICANCE: To the best of our knowledge, this is the first reported result of water-fat separation at a 50 mT portable MRI scanner.
Asunto(s)
Tejido Adiposo , Imagen por Resonancia Magnética , Fantasmas de Imagen , Imagen por Resonancia Magnética/métodos , Imagen por Resonancia Magnética/instrumentación , Humanos , Tejido Adiposo/diagnóstico por imagen , Agua Corporal/diagnóstico por imagen , Rodilla/diagnóstico por imagen , Articulación de la Rodilla/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Diseño de EquipoRESUMEN
BACKGROUND: Maintaining normal patellar alignment is important for knee health. Altered activation of individual quadriceps muscles have been found related to patellar alignment. However, the relationships between strength and passive stiffness of the quadriceps and patellar alignment remains unexplored. METHODS: Participants aged between 60 and 80 years with activity-induced knee pain were recruited. Knee pain was quantified using an 11-point numeric rating scale. Quadriceps strength was assessed using a Cybex dynamometer and passive stiffness of rectus femoris, vastus lateralis, and vastus medialis were measured by shear-wave ultrasound elastography. Patellar alignments were assessed using MR imaging. Linear regression was used to examine relationships between quadriceps properties and patellar alignments with and without controlling for potential covariates. FINDINGS: Ninety-two eligible participants were assessed (71.7% females, age: 65.6 ± 3.8 years; pain scale: 4.6 ± 2.0), most of whom had knee pain during stair climbing (85.9%). We found that 17% of patellar lateral tilt angle could be explained by lower quadriceps strength (adjusted R2 = 0.117; P < 0.001), especially in females (R2 = 0.281; P < 0.001; adjusted R2 = 0.211; P < 0.001). In addition, a higher stiffness ratio of vastus lateralis/medialis accounted for 12% of patellar lateral displacement (adjusted R2 = 0.112; P = 0.008). INTERPRETATION: Quadriceps strength and relative stiffness of lateral to medial heads are associated with patellar alignment in older adults with knee pain. It suggests that quadriceps weakness and relatively stiffer lateral quadriceps may be risk factors related to patellar malalignments in the elderly.