Improvement of image quality in diffusion-weighted imaging with model-based deep learning reconstruction for evaluations of the head and neck.

OBJECTIVES: To investigate the utility of deep learning (DL)-based image reconstruction using a model-based approach in head and neck diffusion-weighted imaging (DWI). MATERIALS AND METHODS: We retrospectively analyzed the cases of 41 patients who underwent head/neck DWI. The DWI in 25 patients demonstrated an untreated lesion. We performed qualitative and quantitative assessments in the DWI analyses with both deep learning (DL)- and conventional parallel imaging (PI)-based reconstructions. For the qualitative assessment, we visually evaluated the overall image quality, soft tissue conspicuity, degree of artifact(s), and lesion conspicuity based on a five-point system. In the quantitative assessment, we measured the signal-to-noise ratio (SNR) of the bilateral parotid glands, submandibular gland, the posterior muscle, and the lesion. We then calculated the contrast-to-noise ratio (CNR) between the lesion and the adjacent muscle. RESULTS: Significant differences were observed in the qualitative analysis between the DWI with PI-based and DL-based reconstructions for all of the evaluation items (p < 0.001). In the quantitative analysis, significant differences in the SNR and CNR between the DWI with PI-based and DL-based reconstructions were observed for all of the evaluation items (p = 0.002 ~ p < 0.001). DISCUSSION: DL-based image reconstruction with the model-based technique effectively provided sufficient image quality in head/neck DWI.

High Resolution TOF-MRA Using Compressed Sensing-based Deep Learning Image Reconstruction for the Visualization of Lenticulostriate Arteries: A Preliminary Study.

PURPOSE: To investigate the visibility of the lenticulostriate arteries (LSAs) in time-of-flight (TOF)-MR angiography (MRA) using compressed sensing (CS)-based deep learning (DL) image reconstruction by comparing its image quality with that obtained by the conventional CS algorithm. METHODS: Five healthy volunteers were included. High-resolution TOF-MRA images with the reduction (R)-factor of 1 were acquired as full-sampling data. Images with R-factors of 2, 4, and 6 were then reconstructed using CS-DL and conventional CS (the combination of CS and sensitivity conceding; CS-SENSE) reconstruction, respectively. In the quantitative assessment, the number of visible LSAs (identified by two radiologists), length of each depicted LSA (evaluated by one radiological technologist), and normalized mean squared error (NMSE) value were assessed. In the qualitative assessment, the overall image quality and the visibility of the peripheral LSA were visually evaluated by two radiologists. RESULTS: In the quantitative assessment of the DL-CS images, the number of visible LSAs was significantly higher than those obtained with CS-SENSE in the R-factors of 4 and 6 (Reader 1) and in the R-factor of 6 (Reader 2). The length of the depicted LSAs in the DL-CS images was significantly longer in the R-factor 6 compared to the CS-SENSE result. The NMSE value in CS-DL was significantly lower than in CS-SENSE for R-factors of 4 and 6. In the qualitative assessment of DL-CS images, the overall image quality was significantly higher than that obtained with CS-SENSE in the R-factors 4 and 6 (Reader 1) and in the R-factor 4 (Reader 2). The visibility of the peripheral LSA was significantly higher than that shown by CS-SENSE in all R-factors (Reader 1) and in the R-factors 2 and 4 (Reader 2). CONCLUSION: CS-DL reconstruction demonstrated preserved image quality for the depiction of LSAs compared to the conventional CS-SENSE when the R-factor is elevated.

Improved image quality in contrast-enhanced 3D-T1 weighted sequence by compressed sensing-based deep-learning reconstruction for the evaluation of head and neck.

PURPOSE: To assess the utility of deep learning (DL)-based image reconstruction with the combination of compressed sensing (CS) denoising cycle by comparing images reconstructed by conventional CS-based method without DL in fat-suppressed (Fs)-contrast enhanced (CE) three-dimensional (3D) T1-weighted images (T1WIs) of the head and neck. MATERIALS AND METHODS: We retrospectively analyzed the cases of 39 patients who had undergone head and neck Fs-CE 3D T1WI applying reconstructions based on conventional CS and CS augmented by DL, respectively. In the qualitative assessment, we evaluated overall image quality, visualization of anatomical structures, degree of artifacts, lesion conspicuity, and lesion edge sharpness based on a five-point system. In the quantitative assessment, we calculated the signal-to-noise ratios (SNRs) of the lesion and the posterior neck muscle and the contrast-to-noise ratio (CNR) between the lesion and the adjacent muscle. RESULTS: For all items of the qualitative analysis, significantly higher scores were awarded to images with DL-based reconstruction (p < 0.001). In the quantitative analysis, DL-based reconstruction resulted in significantly higher values for both the SNR of lesions (p < 0.001) and posterior neck muscles (p < 0.001). Significantly higher CNRs were also observed in images with DL-based reconstruction (p < 0.001). CONCLUSION: DL-based image reconstruction integrating into the CS-based denoising cycle offered superior image quality compared to the conventional CS method. This technique will be useful for the assessment of patients with head and neck disease.

Comparison of Echo-Planar Imaging and Compressed Sensing in the Estimation of Flow Metrics from Aortic 4D Flow MR Imaging: A Healthy Volunteer Study.

PURPOSE: Prolonged scanning of time-resolved 3D phase-contrast MRI (4D flow MRI) limits its routine use in clinical practice. An echo-planar imaging (EPI)-based sequence and compressed sensing can reduce the scan duration. We aimed to determine the impact of EPI for 4D flow MRI on the scan duration, image quality, and quantitative flow metrics. METHODS: This was a prospective study of 15 healthy volunteers (all male, mean age 33 ± 5 years). Conventional sensitivity encoding (SENSE), EPI with SENSE (EPI), and compressed SENSE (CS) (reduction factors: 6 and 12, respectively) were scanned.Scan duration, qualitative indexes of image quality, and quantitative flow parameters of net flow volume, maximum flow velocity, wall shear stress (WSS), and energy loss (EL) in the ascending aorta were assessed. Two-dimensional phase-contrast cine MRI (2D-PC) was considered the gold standard of net flow volume and maximum flow velocity. RESULTS: Compared to SENSE, EPI and CS12 shortened scan durations by 71% and 73% (EPI, 4 min 39 sec; CS6, 7 min 29 sec; CS12, 4 min 14 sec; and SENSE, 15 min 51 sec). Visual image quality was significantly better for EPI than for SENSE and CS (P < 0.001). The net flow volumes obtained with SENSE, EPI, and CS12 and those obtained with 2D-PC were correlated well (r = 0.950, 0.871, and 0.850, respectively). However, the maximum velocity obtained with EPI was significantly underestimated (P < 0.010). The average WSS was significantly higher with EPI than with SENSE, CS6, and CS12 (P < 0.001, P = 0.040, and P = 0.012, respectively). The EL was significantly lower with EPI than with CS6 and CS12 (P = 0.002 and P = 0.007, respectively). CONCLUSION: EPI reduced the scan duration, improved visual image quality, and was associated with more accurate net flow volume than CS. However, the flow velocity, WSS, and EL values obtained with EPI and other sequences may not be directly comparable.

Enhancing the image quality of prostate diffusion-weighted imaging in patients with prostate cancer through model-based deep learning reconstruction.

Purpose: To evaluate the utility of model-based deep learning reconstruction in prostate diffusion-weighted imaging (DWI). Methods: This retrospective study evaluated two prostate diffusion-weighted imaging (DWI) methods: deep learning reconstruction (DL-DWI) and traditional parallel imaging (PI-DWI). We examined 32 patients with radiologically diagnosed and histologically confirmed prostate cancer (PCa) lesions ≥10 mm. Image quality was evaluated both qualitatively (for overall quality, prostate conspicuity, and lesion conspicuity) and quantitatively, using the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and apparent diffusion coefficient (ADC) for prostate tissue. Results: In the qualitative evaluation, DL-DWI scored significantly higher than PI-DWI for all three parameters (p<0.0001). In the quantitative analysis, DL-DWI showed significantly higher SNR and CNR values compared to PI-DWI (p<0.0001). Both the prostate tissue and the lesions exhibited significantly higher ADC values in DL-DWI compared to PI-DWI (p<0.0001, p=0.0014, respectively). Conclusion: Model-based DL reconstruction enhanced both qualitative and quantitative aspects of image quality in prostate DWI. However, this study did not include comparisons with other DL-based methods, which is a limitation that warrants future research.

Technical Advancements in Abdominal Diffusion-weighted Imaging.

Since its first observation in the 18th century, the diffusion phenomenon has been actively studied by many researchers. Diffusion-weighted imaging (DWI) is a technique to probe the diffusion of water molecules and create a MR image with contrast based on the local diffusion properties. The DWI pixel intensity is modulated by the hindrance the diffusing water molecules experience. This hindrance is caused by structures in the tissue and reflects the state of the tissue. This characteristic makes DWI a unique and effective tool to gain more insight into the tissue's pathophysiological condition. In the past decades, DWI has made dramatic technical progress, leading to greater acceptance in clinical practice. In the abdominal region, however, acquiring DWI with good quality is challenging because of several reasons, such as large imaging volume, respiratory and other types of motion, and difficulty in achieving homogeneous fat suppression. In this review, we discuss technical advancements from the past decades that help mitigate these problems common in abdominal imaging. We describe the use of scan acceleration techniques such as parallel imaging and compressed sensing to reduce image distortion in echo planar imaging. Then we compare techniques developed to mitigate issues due to respiratory motion, such as free-breathing, respiratory-triggering, and navigator-based approaches. Commonly used fat suppression techniques are also introduced, and their effectiveness is discussed. Additionally, the influence of the abovementioned techniques on image quality is demonstrated. Finally, we discuss the current and future clinical applications of abdominal DWI, such as whole-body DWI, simultaneous multiple-slice excitation, intravoxel incoherent motion, and the use of artificial intelligence. Abdominal DWI has the potential to develop further in the future, thanks to scan acceleration and image quality improvement driven by technological advancements. The accumulation of clinical proof will further drive clinical acceptance.

In vivo hypoxia characterization using blood oxygen level dependent magnetic resonance imaging in a preclinical glioblastoma mouse model.

PURPOSE: Hypoxia measurements can provide crucial information regarding tumor aggressiveness, however current preclinical approaches are limited. Blood oxygen level dependent (BOLD) Magnetic Resonance Imaging (MRI) has the potential to continuously monitor tumor pathophysiology (including hypoxia). The aim of this preliminary work was to develop and evaluate BOLD MRI followed by post-image analysis to identify regions of hypoxia in a murine glioblastoma (GBM) model. METHODS: A murine orthotopic GBM model (GL261-luc2) was used and independent images were generated from multiple slices in four different mice. Image slices were randomized and split into training and validation cohorts. A 7 T MRI was used to acquire anatomical images using a fast-spin-echo (FSE) T2-weighted sequence. BOLD images were taken with a T2*-weighted gradient echo (GRE) and an oxygen challenge. Thirteen images were evaluated in a training cohort to develop the MRI sequence and optimize post-image analysis. An in-house MATLAB code was used to evaluate MR images and generate hypoxia maps for a range of thresholding and ΔT2* values, which were compared against respective pimonidazole sections to optimize image processing parameters. The remaining (n = 6) images were used as a validation group. Following imaging, mice were injected with pimonidazole and collected for immunohistochemistry (IHC). A test of correlation (Pearson's coefficient) and agreement (Bland-Altman plot) were conducted to evaluate the respective MRI slices and pimonidazole IHC sections. RESULTS: For the training cohort, the optimized parameters of "thresholding" (20 ≤ T2* ≤ 35 ms) and ΔT2* (±4 ms) yielded a Pearson's correlation of 0.697. These parameters were applied to the validation cohort confirming a strong Pearson's correlation (0.749) when comparing the respective analyzed MR and pimonidazole images. CONCLUSION: Our preliminary study supports the hypothesis that BOLD MRI is correlated with pimonidazole measurements of hypoxia in an orthotopic GBM mouse model. This technique has further potential to monitor hypoxia during tumor development and therapy.

Use of 3-D Contrast-Enhanced Ultrasound to Evaluate Tumor Microvasculature After Nanoparticle-Mediated Modulation.

A cost-effective method for serial in vivo imaging of tumor microvasculature has been developed. We evaluated acoustic angiography (AA) for visualizing and assessing non-small cell lung tumor (A549) microvasculature in mice before and after tumor vascular disruption by vascular-targeted gold nanoparticles and radiotherapy. Standard B-mode and microbubble-enhanced AA images were acquired at pre- and post-treatment time points. Using these modes, a new metric, 50% vessel penetration depth, was developed to characterize the 3-D spatial heterogeneity of microvascular networks. We observed an increase in tumor perfusion after radiation-induced vascular disruption, relative to control animals. This was also visualized in vessel morphology mode, which revealed a loss in vessel integrity. We found that tumors with poorly perfused vasculature at day 0 exhibited a reduced growth rate over time. This suggested a new method to reduce in-group treatment response variability using pre-treatment microvessel maps to objectively identify animals for study removal.

Noninvasive imaging of tumor hypoxia after nanoparticle-mediated tumor vascular disruption.

We have previously demonstrated that endothelial targeting of gold nanoparticles followed by external beam irradiation can cause specific tumor vascular disruption in mouse models of cancer. The induced vascular damage may lead to changes in tumor physiology, including tumor hypoxia, thereby compromising future therapeutic interventions. In this study, we investigate the dynamic changes in tumor hypoxia mediated by targeted gold nanoparticles and clinical radiation therapy (RT). By using noninvasive whole-body fluorescence imaging, tumor hypoxia was measured at baseline, on day 2 and day 13, post-tumor vascular disruption. A 2.5-fold increase (P<0.05) in tumor hypoxia was measured two days after combined therapy, resolving by day 13. In addition, the combination of vascular-targeted gold nanoparticles and radiation therapy resulted in a significant (P<0.05) suppression of tumor growth. This is the first study to demonstrate the tumor hypoxic physiological response and recovery after delivery of vascular-targeted gold nanoparticles followed by clinical radiation therapy in a human non-small cell lung cancer athymic Foxn1nu mouse model.

Physiological mechanism underlying the improvement in visuospatial performance due to 30% oxygen inhalation.

This study investigated the effect of 30% oxygen inhalation on visuospatial cognitive performance, blood oxygen saturation, and heart rate. Six male (25.8(mean)+/-1.0(SD) years) and six female (23.8+/-1.9 years) college students participated in this experiment. Two psychological tests were developed to measure the performance level of visuospatial cognition. The experiment consisted of two runs: one was a visuospatial cognition task under normal air (21% oxygen) condition and the other under hyperoxic air (30% oxygen) condition. The experimental sequence in each run consisted of four phases, that were Rest1 (1min), Control (1min), Task (4min), and Rest2 (4min). Blood oxygen saturation and heart rate were measured throughout the course of four phases. The analysis of behavioral performance with 30% oxygen administration when compared to 21% oxygen revealed that the mean performance was improved. When supplied 30% oxygen in the air, the blood oxygen saturation was increased while the heart rate was decreased compared to those under 21% oxygen condition. We conclude that 30% oxygen inhalation enhanced visuospatial performance by the increased the oxygen saturation in the blood.

A comparison of the mean signal change method and the voxel count method to evaluate the sensitivity of individual variability in visuospatial performance.

This study compared the mean signal change method and the voxel count method in evaluating the sensitivity of individual variability in visuospatial performance using functional Magnetic Resonance Imaging (fMRI). Sixteen right-handed male college students (mean age 23.2 years) participated in this study as subjects. Functional brain images were scanned with a 3T MRI single-shot EPI method during a visuospatial task. No correlation was found between visuospatial performance and the number of activated voxels in the activated brain areas. Significant positive correlations, however, were found between visuospatial performance and the mean signal changes of activated voxels in the parietal, frontal and other areas. In conclusion, the mean signal change is more sensitive to individual variability in visuospatial performance than the number of activated voxels.

Effects of high concentration oxygen administration on n-back task performance and physiological signals.

This study investigated the effect of 40% oxygen administration on n-back task performance, blood oxygen saturation and heart rate. Five male (25.8 +/- 1.3 years) and five female (23.0 +/- 1.0 years) college students were selected as the subjects for this study. The experiment consisted of two runs: one was an n-back task with normal air (21% oxygen) administered and the other was with hyperoxic air (40% oxygen) administered. The experimental sequence in each run consisted of Rest1 (1 min), 0-back task (1 min), 2-back task (2 min) and Rest2 (4 min). Blood oxygen saturation and heart rate were measured throughout the four phases. The results of the n-back behavioural analysis reveal that accuracy rates were enhanced with 40% oxygen administration compared to 21% oxygen. When 40% oxygen was supplied, blood oxygen saturation was increased and heart rate was decreased compared to that with 21% oxygen administration. It is suggested that 40% oxygen can stimulate brain activation by increasing actual blood oxygen concentration in the process of cognitive performance, and hyperoxia makes heart rate decrease. This result supports the hypothesis that 40% oxygen administration would lead to increases in n-back task performance.

Development of a magnetic-resonance-compatible photoplethysmograph amplifier for behavioral and emotional studies.

The purpose of the present study was to develop a magnetic resonance (MR) compatible photoplethysmograph (PPG) system that can measure the raw PPG signal during MR image acquisition. The system consists of an optic sensor that measures the optic signal, an optic cable that transmits a near-infrared optic signal, a signal amplifier, and a filter for noise removal. To minimize interactive noise, only the optic cable and the optic sensor module are located inside the MR room; the signal amplifier and filter are located outside the MR room. An experiment verified that a reliable PPG signal can be obtained without causing a deterioration in the MR image.

Effect of 30% oxygen administration on verbal cognitive performance, blood oxygen saturation and heart rate.

This study investigated the effect of 30% oxygen administration on verbal cognitive performance, blood oxygen saturation, and heart rate. Five male (24.6(+/-0.9) years) and five female (22.2(+/-1.9) years) college students were selected as the subjects for this study. Two psychological tests were developed to measure the performance level of verbal cognition. The experiment consisted of two runs: one was a verbal cognition task, with normal air (21% oxygen) administered and the other was with hyperoxic air (30% oxygen) administered. The experimental sequence in each run consisted of Rest 1 (1 min), Control (1 min), Task (4 min), and Rest 2 (4 min). Blood oxygen saturation and heart rate were measured throughout the four phases. The results of the verbal behavioural analysis reveal that accuracy rates were enhanced with 30% oxygen administration compared to 21% oxygen. When 30% oxygen was supplied, blood oxygen saturation was increased significantly compared to that with 21% oxygen administration, whereas heart rate showed no significant difference. Significant positive correlations were found between changes in oxygen saturation and cognitive performance. This result supports the hypothesis that 30% oxygen administration would lead to increases in verbal cognitive performance.