RESUMEN
BACKGROUND: Patient head motion is a common source of image artifacts in computed tomography (CT) of the head, leading to degraded image quality and potentially incorrect diagnoses. The partial angle reconstruction (PAR) means dividing the CT projection into several consecutive angular segments and reconstructing each segment individually. Although motion estimation and compensation using PAR has been developed and investigated in cardiac CT scans, its potential for reducing motion artifacts in head CT scans remains unexplored. PURPOSE: To develop a deep learning (DL) model capable of directly estimating head motion from PAR images of head CT scans and to integrate the estimated motion into an iterative reconstruction process to compensate for the motion. METHODS: Head motion is considered as a rigid transformation described by six time-variant variables, including the three variables for translation and three variables for rotation. Each motion variable is modeled using a B-spline defined by five control points (CP) along time. We split the full projections from 360° into 25 consecutive PARs and subsequently input them into a convolutional neural network (CNN) that outputs the estimated CPs for each motion variable. The estimated CPs are used to calculate the object motion in each projection, which are incorporated into the forward and backprojection of an iterative reconstruction algorithm to reconstruct the motion-compensated image. The performance of our DL model is evaluated through both simulation and phantom studies. RESULTS: The DL model achieved high accuracy in estimating head motion, as demonstrated in both the simulation study (mean absolute error (MAE) ranging from 0.28 to 0.45 mm or degree across different motion variables) and the phantom study (MAE ranging from 0.40 to 0.48 mm or degree). The resulting motion-corrected image, I D L , P A R ${I}_{DL,\ PAR}$ , exhibited a significant reduction in motion artifacts when compared to the traditional filtered back-projection reconstructions, which is evidenced both in the simulation study (image MAE drops from 178 ± $ \pm $ 33HU to 37 ± $ \pm $ 9HU, structural similarity index (SSIM) increases from 0.60 ± $ \pm $ 0.06 to 0.98 ± $ \pm $ 0.01) and the phantom study (image MAE drops from 117 ± $ \pm $ 17HU to 42 ± $ \pm $ 19HU, SSIM increases from 0.83 ± $ \pm $ 0.04 to 0.98 ± $ \pm $ 0.02). CONCLUSIONS: We demonstrate that using PAR and our proposed deep learning model enables accurate estimation of patient head motion and effectively reduces motion artifacts in the resulting head CT images.
Asunto(s)
Artefactos , Aprendizaje Profundo , Procesamiento de Imagen Asistido por Computador , Tomografía Computarizada por Rayos X , Humanos , Tomografía Computarizada por Rayos X/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Cabeza/diagnóstico por imagen , Movimientos de la Cabeza , Fantasmas de ImagenRESUMEN
Objective.The mean squared error (MSE), also known asL2loss, has been widely used as a loss function to optimize image denoising models due to its strong performance as a mean estimator of the Gaussian noise model. Recently, various low-dose computed tomography (LDCT) image denoising methods using deep learning combined with the MSE loss have been developed; however, this approach has been observed to suffer from the regression-to-the-mean problem, leading to over-smoothed edges and degradation of texture in the image.Approach.To overcome this issue, we propose a stochastic function in the loss function to improve the texture of the denoised CT images, rather than relying on complicated networks or feature space losses. The proposed loss function includes the MSE loss to learn the mean distribution and the Pearson divergence loss to learn feature textures. Specifically, the Pearson divergence loss is computed in an image space to measure the distance between two intensity measures of denoised low-dose and normal-dose CT images. The evaluation of the proposed model employs a novel approach of multi-metric quantitative analysis utilizing relative texture feature distance.Results.Our experimental results show that the proposed Pearson divergence loss leads to a significant improvement in texture compared to the conventional MSE loss and generative adversarial network (GAN), both qualitatively and quantitatively.Significance.Achieving consistent texture preservation in LDCT is a challenge in conventional GAN-type methods due to adversarial aspects aimed at minimizing noise while preserving texture. By incorporating the Pearson regularizer in the loss function, we can easily achieve a balance between two conflicting properties. Consistent high-quality CT images can significantly help clinicians in diagnoses and supporting researchers in the development of AI-diagnostic models.
Asunto(s)
Procesamiento de Imagen Asistido por Computador , Dosis de Radiación , Relación Señal-Ruido , Tomografía Computarizada por Rayos X , Tomografía Computarizada por Rayos X/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Humanos , Aprendizaje ProfundoRESUMEN
Image-guided tumor ablative therapies are mainstay cancer treatment options but often require intra-procedural protective tissue displacement to reduce the risk of collateral damage to neighboring organs. Standard of care strategies, such as hydrodissection (fluidic injection), are limited by rapid diffusion of fluid and poor retention time, risking injury to adjacent organs, increasing cancer recurrence rates from incomplete tumor ablations, and limiting patient qualification. Herein, a "gel-dissection" technique is developed, leveraging injectable hydrogels for longer-lasting, shapeable, and transient tissue separation to empower clinicans with improved ablation operation windows and greater control. A rheological model is designed to understand and tune gel-dissection parameters. In swine models, gel-dissection achieves 24 times longer-lasting tissue separation dynamics compared to saline, with 40% less injected volume. Gel-dissection achieves anti-dependent dissection between free-floating organs in the peritoneal cavity and clinically significant thermal protection, with the potential to expand minimally invasive therapeutic techniques, especially across locoregional therapies including radiation, cryoablation, endoscopy, and surgery.
RESUMEN
Background: Recent reports suggested that dual-energy CT (DECT) may help discriminate between different types of calcium phosphate crystals in vivo, which would have important implications for the characterization of crystal deposition occurring in osteoarthritis. Purpose: Our aim was to test the hypothesis that DECT can effectively differentiate basic calcium phosphate (BCP) from calcium pyrophosphate (CPP) deposition diseases. Methods: Discarded tissue after total knee replacement specimens in a 71 year-old patient with knee osteoarthritis and chondrocalcinosis was scanned using DECT at standard clinical parameters. Specimens were then examined on light microscopy which revealed CPP deposition in 4 specimens (medial femoral condyle, lateral tibial plateau and both menisci) without BCP deposition. Regions of interest were placed on post-processed CT images using Rho/Z maps (Syngo.via, Siemens Healthineers, VB10B) in different areas of CPP deposition, trabecular bone BCP (T-BCP) and subchondral bone plate BCP (C-BCP). Results: Dual Energy Index (DEI) of CPP was 0.12 (SD â= â0.02) for reader 1 and 0.09 (SD â= â0.03) for reader 2, The effective atomic number (Zeff) of CPP was 10.83 (SD â= â0.44) for reader 1 and 10.11 (SD â= â0.66) for reader 2. Nearly all DECT parameters of CPP were higher than those of T-BCP, lower than those of C-BCP, and largely overlapping with Aggregate-BCP (aggregate of T-BCP and C-BCP). Conclusion: Differentiation of different types of calcium crystals using DECT is not feasible in a clinical setting.