skip to main content

Title: MAGIC: Multitask Automated Generation of Inter-modal CT Perfusion Maps via Generative Adversarial Network
Introduction: Computed tomography perfusion (CTP) imaging requires injection of an intravenous contrast agent and increased exposure to ionizing radiation. This process can be lengthy, costly, and potentially dangerous to patients, especially in emergency settings. We propose MAGIC, a multitask, generative adversarial network-based deep learning model to synthesize an entire CTP series from only a non-contrasted CT (NCCT) input. Materials and Methods: NCCT and CTP series were retrospectively retrieved from 493 patients at UF Health with IRB approval. The data were deidentified and all images were resized to 256x256 pixels. The collected perfusion data were analyzed using the RapidAI CT Perfusion analysis software (iSchemaView, Inc. CA) to generate each CTP map. For each subject, 10 CTP slices were selected. Each slice was paired with one NCCT slice at the same location and two NCCT slices at a predefined vertical offset, resulting in 4.3K CTP images and 12.9K NCCT images used for training. The incorporation of a spatial offset into the NCCT input allows MAGIC to more accurately synthesize cerebral perfusive structures, increasing the quality of the generated images. The studies included a variety of indications, including healthy tissue, mild infarction, and severe infarction. The proposed MAGIC model incorporates a novel multitask architecture, allowing for the simultaneous synthesis of four CTP modalities: mean transit time (MTT), cerebral blood flow (CBF), cerebral blood volume (CBV), and time to peak (TTP). We propose a novel Physicians-in-the-loop module in the model's architecture, acting as a tunable layer that allows physicians to manually adjust the amount of anatomic detail present in the synthesized CTP series. Additionally, we propose two novel loss terms: multi-modal connectivity loss and extrema loss. The multi-modal connectivity loss leverages the multi-task nature to assert that the mathematical relationship between MTT, CBF, and CBV is satisfied. The extrema loss aids in learning regions of elevated and decreased activity in each modality, allowing for MAGIC to accurately learn the characteristics of diagnostic regions of interest. Corresponding NCCT and CTP slices were paired along the vertical axis. The model was trained for 100 epochs on a NVIDIA TITAN X GPU. Results and Discussion: The MAGIC model’s performance was evaluated on a sample of 40 patients from the UF Health dataset. Across all CTP modalities, MAGIC was able to accurately produce images with high structural agreement between the entire synthesized and clinical perfusion images (SSIMmean=0.801 , UQImean=0.926). MAGIC was able to synthesize CTP images to accurately characterize cerebral circulatory structures and identify regions of infarct tissue, as shown in Figure 1. A blind binary evaluation was conducted to assess the presence of cerebral infarction in both the synthesized and clinical perfusion images, resulting in the synthesized images correctly predicting the presence of cerebral infarction with 87.5% accuracy. Conclusions: We proposed a MAGIC model whose novel deep learning structures and loss terms enable high-quality synthesis of CTP maps and characterization of circulatory structures solely from NCCT images, potentially eliminating the requirement for the injection of an intravenous contrast agent and elevated radiation exposure during perfusion imaging. This makes MAGIC a beneficial tool in a clinical scenario increasing the overall safety, accessibility, and efficiency of cerebral perfusion and facilitating better patient outcomes. Acknowledgements: This work was partially supported by the National Science Foundation, IIS-1908299 III: Small: Modeling Multi-Level Connectivity of Brain Dynamics + REU Supplement, to the University of Florida.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
Biomedical Engineering Society Annual Meeting
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Introduction Multi-series CT (MSCT) scans, including non-contrast CT (NCCT), CT Perfusion (CTP), and CT Angiography (CTA), are widely used in acute stroke imaging. While each scan has its advantage in disease diagnosis, the varying image resolution of different series hinders the ability of the radiologist to discern subtle suspicious findings. Besides, higher image quality requires high radiation doses, leading to increases in health risks such as cataract formation and cancer induction. Thus, it is highly crucial to develop an approach to improve MSCT resolution and to lower radiation exposure. Hypothesis MSCT imaging of the same patient is highly correlated in structural features, the transferring and integration of the shared and complementary information from different series are beneficial for achieving high image quality. Methods We propose TL-GAN, a learning-based method by using Transfer Learning (TL) and Generative Adversarial Network (GAN) to reconstruct high-quality diagnostic images. Our TL-GAN method is evaluated on 4,382 images collected from nine patients’ MSCT scans, including 415 NCCT slices, 3,696 CTP slices, and 271 CTA slices. We randomly split the nine patients into a training set (4 patients), a validation set (2 patients), and a testing set (3 patients). In preprocessing, we remove the background and skull and visualize in brain window. The low-resolution images (1/4 of the original spatial size) are simulated by bicubic down-sampling. For training without TL, we train different series individually, and for with TL, we follow the scanning sequence (NCCT, CTP, and CTA) by finetuning. Results The performance of TL-GAN is evaluated by the peak-signal-to-noise ratio (PSNR) and structural similarity (SSIM) index on 184 NCCT, 882 CTP, and 107 CTA test images. Figure 1 provides both visual (a-c) and quantity (d-f) comparisons. Through TL-GAN, there is a significant improvement with TL than without TL (training from scratch) for NCCT, CTP, and CTA images, respectively. These significances of performance improvement are evaluated by one-tailed paired t-tests (p < 0.05). We enlarge the regions of interest for detail visual comparisons. Further, we evaluate the CTP performance by calculating the perfusion maps, including cerebral blood flow (CBF) and cerebral blood volume (CBV). The visual comparison of the perfusion maps in Figure 2 demonstrate that TL-GAN is beneficial for achieving high diagnostic image quality, which are comparable to the ground truth images for both CBF and CBV maps. Conclusion Utilizing TL-GAN can effectively improve the image resolution for MSCT, provides radiologists more image details for suspicious findings, which is a practical solution for MSCT image quality enhancement. 
    more » « less
  2. Key points

    Extreme aviation is accompanied by ever‐present risks of hypobaric hypoxia and decompression sickness. Neuroprotection against those hazards is conferred through fractional inspired oxygen () concentrations of 60–100% (hyperoxia).

    Hyperoxia reduces global cerebral perfusion (gCBF), increases reactive oxygen species within the brain and leads to cell death within the hippocampus. However, an understanding of hyperoxia's effect on cortical activity and concomitant levels of cognitive performance is lacking. This limits our understanding of whether hyperoxia could lower the brain's threshold of tolerance to physiological stressors inherent to extreme aviation, such as high gravitational forces.

    This study aimed to quantify the impact of hyperoxia upon global cerebral perfusion (gCBF), cognitive performance and cortical electroencephalography (EEG).

    Hyperoxia evoked a rapid reduction in gCBF, yet cognitive performance and vigilance were enhanced. EEG measurements revealed enhanced alpha power, suggesting less desynchrony, within the cortical temporal regions.

    Collectively, this work suggests hyperoxia‐induced brain hypoperfusion is accompanied by enhanced cognitive processing and cortical arousal.


    Extreme aviators continually inspire hyperoxic gas to mitigate risk of hypoxia and decompression injury. This neuroprotection carries a physiological cost: reduced cerebral perfusion (CBF). As reduced CBF may increase vulnerability to ever‐present physiological challenges during extreme aviation, we defined the magnitude and duration of hyperoxia‐induced changes in CBF, cortical electrical activity and cognition in 30 healthy males and females. Magnetic resonance imaging with pulsed arterial spin labelling provided serial measurements of global CBF (gCBF), first during exposure to 21% inspired oxygen () followed by a 30‐min exposure to 100% . High‐density EEG facilitated characterization of cortical activity during assessment of cognitive performance, also measured during exposure to 21% and 100% . Acid‐base physiology was measured with arterial blood gases. We found that exposure to 100% reduced gCBF to 63% of baseline values across all participants. Cognitive performance testing at 21% was accompanied by increased theta and beta power with decreased alpha power across multiple cortical areas. During cognitive testing at 100% , alpha activity was less desynchronized within the temporal regions than at 21% . The collective hyperoxia‐induced changes in gCBF, cognitive performance and EEG were similar across observed partial pressures of arterial oxygen (), which ranged between 276–548 mmHg, and partial pressures of arterial carbon dioxide (), which ranged between 34–50 mmHg. Sex did not influence gCBF response to 100% . Our findings suggest hyperoxia‐induced reductions in gCBF evoke enhanced levels of cortical arousal and cognitive processing, similar to those occurring during a perceived threat.

    more » « less
  3. Abstract Background The brain extracellular environment is involved in many critical processes associated with neurodevelopment, neural function, and repair following injury. Organization of the extracellular matrix and properties of the extracellular space vary throughout development and across different brain regions, motivating the need for platforms that provide access to multiple brain regions at different stages of development. We demonstrate the utility of organotypic whole hemisphere brain slices as a platform to probe regional and developmental changes in the brain extracellular environment. We also leverage whole hemisphere brain slices to characterize the impact of cerebral ischemia on different regions of brain tissue. Results Whole hemisphere brain slices taken from postnatal (P) day 10 and P17 rats retained viable, metabolically active cells through 14 days in vitro (DIV). Oxygen-glucose-deprivation (OGD), used to model a cerebral ischemic event in vivo, resulted in reduced slice metabolic activity and elevated cell death, regardless of slice age. Slices from P10 and P17 brains showed an oligodendrocyte and microglia-driven proliferative response after OGD exposure, higher than the proliferative response seen in DIV-matched normal control slices. Multiple particle tracking in oxygen-glucose-deprived brain slices revealed that oxygen-glucose-deprivation impacts the extracellular environment of brain tissue differently depending on brain age and brain region. In most instances, the extracellular space was most difficult to navigate immediately following insult, then gradually provided less hindrance to extracellular nanoparticle diffusion as time progressed. However, changes in diffusion were not universal across all brain regions and ages. Conclusions We demonstrate whole hemisphere brain slices from P10 and P17 rats can be cultured up to two weeks in vitro. These brain slices provide a viable platform for studying both normal physiological processes and injury associated mechanisms with control over brain age and region. Ex vivo OGD impacted cortical and striatal brain tissue differently, aligning with preexisting data generated in in vivo models. These data motivate the need to account for both brain region and age when investigating mechanisms of injury and designing potential therapies for cerebral ischemia. 
    more » « less
  4. null (Ed.)
    Fast and accurate midline shift (MLS) estimation has a significant impact on diagnosis and treatment of patients with Traumatic Brain Injury (TBI). In this paper, we propose an automated method to calculate the amount of shift in the midline structure of TBI patients. The MLS values were annotated by a neuroradiologist. We first select a number of slices among all the slices in a CT scan based on metadata as well as information extracted from the images. After the slice selection, we propose an efficient segmentation technique to detect the ventricles. We use the ventricular geometric patterns to calculate the actual midline and also anatomical information to detect the ideal midline. The distance between these two lines is used as an estimate of MLS. The proposed methods are applied on a TBI dataset where they show a significant improvement of the the proposed method upon existing approach. 
    more » « less
  5. Abstract

    Infancy is marked by rapid neural and emotional development. The relation between brain function and emotion in infancy, however, is not well understood. Methods for measuring brain function predominantly rely on the BOLD signal; however, interpretation of the BOLD signal in infancy is challenging because the neuronal‐hemodynamic relation is immature. Regional cerebral blood flow (rCBF) provides a context for the infant BOLD signal and can yield insight into the developmental maturity of brain regions that may support affective behaviors. This study aims to elucidate the relations among rCBF, age, and emotion in infancy. One hundred and seven mothers reported their infants' (infant ageM ± SD = 6.14 ± 0.51 months) temperament. A subsample of infants completed MRI scans, 38 of whom produced usable perfusion MRI during natural sleep to quantify rCBF. Mother‐infant dyads completed the repeated Still‐Face Paradigm, from which infant affect reactivity and recovery to stress were quantified. We tested associations of infant age at scan, temperament factor scores, and observed affect reactivity and recovery with voxel‐wise rCBF. Infant age was positively associated with CBF in nearly all voxels, with peaks located in sensory cortices and the ventral prefrontal cortex, supporting the formulation that rCBF is an indicator of tissue maturity. Temperamental Negative Affect and recovery of positive affect following a stressor were positively associated with rCBF in several cortical and subcortical limbic regions, including the orbitofrontal cortex and inferior frontal gyrus. This finding yields insight into the nature of affective neurodevelopment during infancy. Specifically, infants with relatively increased prefrontal cortex maturity may evidence a disposition toward greater negative affect and negative reactivity in their daily lives yet show better recovery of positive affect following a social stressor.

    more » « less