Prediction of an oxygen extraction fraction map by convolutional neural network: validation of input data among MR and PET images.

PURPOSE: Oxygen extraction fraction (OEF) is a biomarker for the viability of brain tissue in ischemic stroke. However, acquisition of the OEF map using positron emission tomography (PET) with oxygen-15 gas is uncomfortable for patients because of the long fixation time, invasive arterial sampling, and radiation exposure. We aimed to predict the OEF map from magnetic resonance (MR) and PET images using a deep convolutional neural network (CNN) and to demonstrate which PET and MR images are optimal as inputs for the prediction of OEF maps. METHODS: Cerebral blood flow at rest (CBF) and during stress (sCBF), cerebral blood volume (CBV) maps acquired from oxygen-15 PET, and routine MR images (T1-, T2-, and T2*-weighted images) for 113 patients with steno-occlusive disease were learned with U-Net. MR and PET images acquired from the other 25 patients were used as test data. We compared the predicted OEF maps and intraclass correlation (ICC) with the real OEF values among combinations of MRI, CBF, CBV, and sCBF. RESULTS: Among the combinations of input images, OEF maps predicted by the model learned with MRI, CBF, CBV, and sCBF maps were the most similar to the real OEF maps (ICC: 0.597 ± 0.082). However, the contrast of predicted OEF maps was lower than that of real OEF maps. CONCLUSION: These results suggest that the deep CNN learned useful features from CBF, sCBF, CBV, and MR images and predict qualitatively realistic OEF maps. These findings suggest that the deep CNN model can shorten the fixation time for 15O PET by skipping 15O2 scans. Further training with a larger data set is required to predict accurate OEF maps quantitatively.

[Evaluating Eye Lens Dose of Neurovascular and Cardiac Electrophysiology Interventional Physician].

PURPOSE: The International Commission on Radiological Protection recommended that interventional radiologies (IRs) have high radiation doses and that staff may also be exposed to high doses. In the present study, we measured the radiation exposure dose [3 mm dose equivalent, Hp (3) ] in the eye using an appropriate dosimeter placed next to the physician' s eye during neurovascular intervention procedure (Neuro-IR) and interventional cardiac electrophysiology procedure (EP-IR). METHOD: Physicians wore a direct eye dosemeter just lateral to the left eye and an additional direct eye dosemeter outside the radiation protective glasses close to their left eye. Additionally, a neck badge [0.07 mm dose equivalent, Hp (0.07) ] was worn outside the protective apron to the left of the neck, to compare the direct eye dosimeter estimated doses. The occupational eye lens dose was evaluated over a period of 6-month. RESULTS: The maximum Hp (3) of the Neuro-IR physician was estimated 5.1 mSv without the radiation protective glasses and 1.6 mSv with the radiation protective glasses. On the other hand, the maximum Hp (3) of the EP-IR physician was estimated 29 mSv without the radiation protective glasses and 15 mSv with the radiation protective glasses. CONCLUSION: Physicians eye lens dose [Hp (3) ] tended to be overestimated by the neck badge measurements [Hp (0.07)]. A correct evaluation of the lens dose [Hp (3) ] using the direct eye dosimeter is recommended. Although we found a positive correlation between Hp (0.07) and Hp (3), the value of R2 in the regression equation is low, we recommended that the eye lens dose estimated carefully from Hp (0.07).

OCCUPATIONAL RADIATION EXPOSURE DOSE OF THE EYE IN DEPARTMENT OF CARDIAC ARRHYTHMIA PHYSICIAN.

Interventional radiology (IR) procedures tend to be complex, which delivers high radiation exposure to patient. In the present study, we measured the radiation exposure dose [Hp(3)] in the eye using a direct eye dosemeter placed next to the physician's eye during procedures. Physicians wore a direct eye dosemeter just lateral to eyes and an additional direct eye dosemeter outside the radiation protective eyeglasses close to their eyes. Additionally, a neck glass badge was worn at the neck. Although we found a positive correlation between the left neck glass badge dose [Hp(0.07)] and the left eye lens dose [Hp(3)], the value of R2 of the regression equation were 0.62 and 0.71 (outside and inside). We thought that the exact eye lens dose might not be estimated from the neck glass badge. In conclusion, a correct evaluation of the lens dose [Hp(3)] using the direct eye dosemeter is recommended for tachyarrhythmia physicians.

OCCUPATIONAL RADIATION EXPOSURE OF THE EYE IN NEUROVASCULAR INTERVENTIONAL PHYSICIAN.

Neurovascular interventional radiology (neuro-IR) procedures tend to require an extended fluoroscopic exposure time and repeated digital subtraction angiography. To evaluate the actual measurement of eye lens dose using a direct eye dosemeter in neuro-IR physicians is important. Direct dosimetry using the DOSIRIS™ (IRSN, France) [3 mm dose equivalent, Hp(3)] was performed on 86 cases. Additionally, a neck personal dosemeter (glass badge) [0.07 mm dose equivalent, Hp(0.07)] was worn outside the protective apron to the left of the neck. The average doses per case of neuro-IR physicians were 0.04 mSv/case and 0.02 mSv/case, outside and inside the radiation protection glasses, respectively. The protective effect of radiation protection glasses was approximately 60%. The physician eye lens dose tended to be overestimated by the neck glass badge measurements. A correct evaluation of the lens dose [Hp(3)] using an eye dosemeter such as DOSIRIS™ is needed for neuro-IR physicians.

New real-time patient radiation dosimeter for use in radiofrequency catheter ablation.

In a previous study, we reported on a novel (prototype) real-time patient dosimeter with non-toxic phosphor sensors. In this study, we developed new types of sensors that were smaller than in the previous prototype, and clarified the clinical feasibility of our newly proposed dosimeter. Patient dose measurements obtained with the newly proposed real-time dosimeter were compared with measurements obtained using a calibrated radiophotoluminescence glass reference dosimeter (RPLD). The reference dosimeters were set at almost the same positions as the new real-time dosimeter sensors. We found excellent correlations between the reference RPLD measurements and those obtained using our new real-time dosimeter (r2 = 0.967). However, the new type of dosimeter was found to underestimate radiation skin dose measurements when compared with an RPLD. The most probable reason for this was the size reduction in the phosphor sensor of the new type of dosimeter. We believe that, as a result of reducing the phosphor sensor size, the backscattered X-ray irradiation was underestimated. However, the new dosimeter can accurately determine the absorbed dose by correcting the measured value with calibration factors. The calibration factor for the new type dosimeter was determined (by linear regression) to be ~1.15. New real-time patient dosimeter design would be an effective tool for the real-time measurement of patient skin doses during interventional radiology treatments.

Spatial coefficient of variation in pseudo-continuous arterial spin labeling cerebral blood flow images as a hemodynamic measure for cerebrovascular steno-occlusive disease: A comparative 15O positron emission tomography study.

Pseudo-continuous arterial spin labeling (pCASL) is a completely non-invasive method of cerebral perfusion measurement. However, cerebral blood flow (CBF) quantification is hampered by arterial transit artifacts characterized by bright vascular signals surrounded by decreased signals in tissue regions, which commonly appear in patients with reduced cerebral perfusion pressure. The spatial coefficient of variation (CoV) of pCASL CBF images has been proposed as an alternative region-of-interest (ROI)-based hemodynamic measure to predict prolonged arterial transit time (ATT). This retrospective study investigates the utility of spatial CoV by comparison with 15O positron emission tomography (PET). For patients with cerebrovascular steno-occlusive disease ( n = 17), spatial CoV was positively correlated with ATT independently measured by pulsed arterial spin labeling ( r = 0.597, p < 0.001), confirming its role as an ATT-like hemodynamic measure. Comparisons with 15O PET demonstrated that spatial CoV was positively correlated with vascular mean transit time ( r = 0.587, p < 0.001) and negatively correlated with both resting CBF ( r = -0.541, p = 0.001) and CBF response to hypercapnia ( r = -0.373, p = 0.030). ROI-based spatial CoV calculated from single time-point pCASL can potentially detect subtle perfusion abnormalities in clinical settings.

[Novel Perfusion Evaluation Method Using Phase-ratio Image Map in Head 4D-CT].

PURPOSE: CT perfusion (CTP) is a powerful tool for the assessment of cerebrovascular disease. However, CTP maps are significantly different depending on CTP software and algorithm, even when using identical image data. We developed a phase-ratio image map (PI map), which was a novel perfusion map, without using CTP software. The purpose of this study was to investigate the usefulness of the PI map by comparing it with a positron emission tomography (PET) image. METHODS: Twenty patients (16 men, 4 women; mean age: 61.6 years) with unilateral cervical and intracranial steno-occlusive disease underwent CTP. CTP source images were obtained at 1-s intervals of 23 times and 5 intervals using dynamic multiphase imaging. An early-phase image was generated by computing the average of CT images for 5 s in the vicinity of the peak enhancement curve of a normal hemisphere. A delayed-phase image was generated by computing the average of CT images for 5 s immediately after the early phase. The PI map was created by dividing the delayed-phase image by the early-phase image. We investigated the validity of the PI map compared with PET-cerebral blood flow (CBF). Lesion-to-normal ratios between a PET-CBF and the PI map or two conventional CTP-CBFs were observed and compared, and the relative errors were also compared. RESULT: There was a strong correlation between the PET-CBF and the PI map (R=0.82). Correlations between the PET-CBF and two CTP-CBFs were weak (R=0.30) and middle (R=0.62), respectively. The relative error between the PI map and the PET-CBF was within 10% in most cases. CONCLUSION: The PI map was more similar to the PET-CBF on perfusion evaluation, and did not depend on CTP software. The robustness and simplicity of the PI mapping method would be advantageous compared with conventional CTP mapping methods.

Direct Dose Measurement On Patient During Percutaneous Coronary Intervention Procedures Using Radiophotoluminescence Glass Dosimeters.

The purpose of this research was to measure accurate patient entrance skin dose and maximum skin absorbed dose (MSD) to prevent radiation skin injuries in percutaneous coronary interventions (PCIs). We directly measured the MSD on 50 PCIs by using multiple radiophotoluminescence glass dosimeters and a modified dosimetry gown. Also, we analysed the correlation between the MSD and indirect measurement parameters, such as fluoroscopic time (FT), dose-area product (DAP) and cumulative air kerma (C-AK). There were very strong correlations between MSD and FT, DAP and C-AK, with the correlation between MSD and C-AK being the strongest (r = 0.938). In conclusion, the regression lines using MSD as an outcome value (y) and C-AK as predictor variables (x) were y = 1.12x (R2 = 0.880). From the linear regression equation, MSD is estimated to be ~1.12 times that of C-AK in real time.

[Multicenter Study on Evaluation of the Entrance Skin Dose by a Direct Measurement Method in Cardiac Interventional Procedures].

Deterministic effects have been reported in cardiac interventional procedures. To prevent radiation skin injuries in percutaneous coronary intervention (PCI), it is necessary to measure accurate patient entrance skin dose (ESD) and maximum skin absorbed dose (MSD). We measured the MSD on 62 patients in four facilities by using the Chest-RADIREC(Ⓡ) system. The correlation between MSD and fluoroscopic time, dose area product (DAP), and cumulative air kerma (AK) showed good results, with the correlation between MSD and AK being the strongest. The regression lines using MSD as an outcome value (y) and AK as predictor variables (x) was y=1.18x (R(2)=0.787). From the linear regression equation, MSD is estimated to be about 1.18 times that of AK in real time. The Japan diagnostic reference levels (DRLs) 2015 for IVR was established by the use of dose rates using acrylic plates (20- cm thick) at the interventional reference point. Preliminary reference levels proposed by International Atomic Energy Agency (IAEA) were provided using DAP. In this study, AK showed good correlation most of all. Hence we think that Japanese DRLs for IVR should reconsider by clinical patients' exposure dose such as AK.

Z-score-based semi-quantitative analysis of the volume of the temporal horn of the lateral ventricle on brain CT images.

The volume of the temporal horn of the lateral ventricle (THLV) on brain computed tomography (CT) images is important for neurologic diagnosis. Our purpose in this study was to develop a z-score-based semi-quantitative analysis for estimation of the THLV volume by using voxel-based morphometry. The THLV volume was estimated by use of a z-score mapping method that consisted of four main steps: anatomic standardization, construction of a normal reference database, calculation of the z score, and calculation of the mean z score in a volume of interest (VOI). A mean z score of the CT value obtained from a VOI around the THLV was used as an index for the THLV volume. CT scans from 50 subjects were evaluated. For evaluation of the accuracy of this method for estimating the THLV volume, the THLV volume was determined manually by neuroradiologists (serving as the reference volume). A mean z score was calculated from the VOI for each THLV of the 50 subjects by use of the proposed method. The accuracy of this method was evaluated by use of the relationship between the mean z score and the reference volume. The quadratic polynomial regression equation demonstrated a statistically significant correlation between the mean z score and the reference volume of the THLV (R (2) = 0.94; P < 0.0001). In 92 of 100 THLVs (92 %), the 95 % prediction interval of the regional mean z score captured the reference volume of the THLV. The z-score-based semi-quantitative analysis has the potential quantitatively to estimate the THLV volume on CT images.

[Examination of Visual Effect in Low-dose Cerebral CT Perfusion Phantom Image Using Iterative Reconstruction].

CT perfusion (CTP) is obtained cerebrovascular circulation image for assessment of stroke patients; however, at the expense of increased radiation dose by dynamic scan. Iterative reconstruction (IR) method is possible to decrease image noise, it has the potential to reduce radiation dose. The purpose of this study is to assess the visual effect of IR method by using a digital perfusion phantom. The digital perfusion phantom was created by reconstructed filtered back projection (FBP) method and IR method CT images that had five exposure doses. Various exposure dose cerebral blood flow (CBF) images were derived from deconvolution algorithm. Contrast-to-noise ratio (CNR) and visual assessment were compared among the various exposure dose and each reconstructions. Result of low exposure dose with IR method showed, compared with FBP method, high CNR in severe ischemic area, and visual assessment was significantly improvement. IR method is useful for improving image quality of low-dose CTP.

[Reduction of radiation exposure during computed tomography perfusion using an iterative reconstruction method].

The purpose of this study was to evaluate the image noise reduction effect of iterative reconstruction (IR) when used to reduce radiation exposure during computed tomography (CT) perfusion. We scanned a contrast phantom using various radiation doses. Image reconstruction was via filtered back projection (FBP) and IR (adaptive iterative dose reduction 3D: AIDR3D). AIDR3D provided four levels of noise reduction (weak, mild, standard, and strong). We examined the accuracy of CTP map (cerebral blood volume: CBV, mean transist time: MTT, cerebral blood flow: CBF) low-dose IR images to create a digital perfusion phantom that simulates the dynamic curve of ischemic cases using reconstructed images. The optimal filter type of IR was evaluated in the low-frequency area of the NPS at low doses. We were able to obtain the optimal filter type of IR in the low-frequency area of the NPS that was equivalent to that of the reference (150 mA, FBP). The CTP map created using the optimal filter type of IR allowed dose reduction to 80 mA, much lower than the reference. We conclude that it is possible to reduce the dose to 46% of the reference level by using the NPS for dose reduction and IR. IR thus has the potential to contribute to reduction of radiation exposure during CT perfusion.

Whole-brain perfusion measurement using 320-detector row computed tomography in patients with cerebrovascular steno-occlusive disease: comparison with 15O-positron emission tomography.

OBJECTIVE: The 320-detector row computed tomography (CT) can provide whole-brain CT perfusion (CTP) maps with continuous angiographic images by performing a single dynamic scan. We investigated the reliability of CTP cerebral blood flow (CTP-CBF) with 320-detector row CT by comparing findings with O-positron emission tomography (PET-CBF). METHODS: Whole-brain CTP and PET were performed in 10 patients with chronic unilateral steno-occlusive disease. We compared absolute and relative CBF values of bilateral middle cerebral artery territories between CTP and PET. RESULTS: Although mean CTP-CBF values were approximately 30% lower than mean PET-CBF values, the mean ischemic-to-nonischemic CBF ratios of CTP and PET were almost identical (P = 0.804). Regression analysis showed a significant correlation between CTP-CBF and PET-CBF values for each patient (r = 0.52-0.85, P < 0.001). CONCLUSIONS: Whole-brain CTP using 320-detector row CT is useful for evaluating the degree of ischemia for the entire brain with chronic cerebrovascular disease.

[Estimation of T1 and T2 using general-purpose spreadsheet software].

We performed an estimation of longitudinal (T1) and transverse relaxation (T2) time using the general-purpose spreadsheet software Microsoft Excel. The Excel tool "solver" is useful for the simultaneous estimation of both T1 and steady-state magnetization from the non-linear least square method. The estimation time is quick enough for the purpose. T1 and T2 estimated from handwritten semi-log plots were compared with the results from spreadsheet software from the viewpoint of accuracy using the phantom data. Although the data from handwritten plots include an estimation error of several percent among the subjects, the mean values are almost the same as compared with the data from spreadsheet software.

Cerebral vascular mean transit time in healthy humans: a comparative study with PET and dynamic susceptibility contrast-enhanced MRI.

Cerebral vascular mean transit time (MTT), defined as the ratio of cerebral blood volume to cerebral blood flow (CBV/CBF), is a valuable indicator of the cerebral circulation. Positron emission tomography (PET) and dynamic susceptibility contrast-enhanced magnetic resonance imaging (DSC-MRI) are useful for the quantitative determination of MTT in the clinical setting. The aim of this study was to establish a normal value set of MTT as determined by PET and by DSC-MRI and to identify differences between these methods. Seven healthy volunteers were studied with (15)O-PET (H(2)(15)O and C(15)O) and gradient-echo echo-planar DSC-MRI at 1.5 T. In the DSC-MRI study with bolus injection of contrast agent, deconvolution analysis was performed. Comparison of gray-to-white matter ratios showed fairly good agreement between PET and DSC-MRI for all parameters (relative CBV, relative CBF, and relative MTT), confirming the validity of relative measurements with DSC-MRI. However, quantitative MTT measured by DSC-MRI was significantly shorter than that measured by PET in cerebral cortical regions (2.8 to 3.0 secs for DSC-MRI versus 3.9 to 4.3 secs for PET) and the centrum semiovale (3.5 secs for DSC-MRI versus 4.8 secs for PET). These discrepancies may be because of the differences in the intrinsic sensitivity of each imaging modality to vascular components; whereas PET measurement of CBV is equally sensitive to all vascular components, measurement with DSC-MRI originates from the microvasculature in the vicinity of the brain parenchyma. This underlying difference may influence interpretation of MTT determined by PET or by DSC-MRI for patients with cerebrovascular disease.

18F-FDG accumulation in atherosclerosis: use of CT and MR co-registration of thoracic and carotid arteries.

PURPOSE: The purpose of this study was to depict( 18)F-fluoro-2-deoxy-D: -glucose (FDG) accumulation in atherosclerotic lesions of the thoracic and carotid arteries on CT and MR images by means of automatic co-registration software. METHODS: Fifteen hospitalised men suffering cerebral infarction or severe carotid stenosis requiring surgical treatment participated in this study. Automatic co-registration of neck MR images and FDG-PET images and of contrast-enhanced CT images and FDG-PET images was achieved with co-registration software. We calculated the count ratio, which was standardised to the blood pool count of the superior vena cava, for three arteries that branch from the aorta, i.e. the brachial artery, the left common carotid artery and the subclavian artery (n=15), for atherosclerotic plaques in the thoracic aorta (n=10) and for internal carotid arteries with and without plaque (n=13). RESULTS: FDG accumulated to a significantly higher level in the brachial artery, left common carotid artery and left subclavian artery at their sites of origin than in the superior vena cava (p=0.000, p=0.000 and p=0.002, respectively). Chest CT showed no atherosclerotic plaque at these sites. Furthermore, the average count ratio of thoracic aortic atherosclerotic plaques was not higher than that of the superior vena cava. The maximum count ratio of carotid atherosclerotic plaques was significantly higher than that of the superior vena cava but was not significantly different from that of the carotid artery without plaque. CONCLUSION: The results of our study suggest that not all atherosclerotic plaques show high FDG accumulation. FDG-PET studies of plaques with the use of fused images can potentially provide detailed information about atherosclerosis.

Effect of regional tracer delay on CBF in healthy subjects measured with dynamic susceptibility contrast-enhanced MRI: comparison with 15O-PET.

PURPOSE: Deconvolution based on truncated singular value decomposition (SVD deconvolution) is a promising method for measuring cerebral blood flow (CBF) with dynamic susceptibility contrast-enhanced magnetic resonance imaging (DSC-MRI), but it has proved extremely sensitive to tracer delay. The purpose of this study was to investigate the effect of regional tracer delay on CBF determined by SVD deconvolution (SVD-CBF). SVD-CBFs with and without correction for the delay were compared with CBF measured by positron emission tomography (PET-CBF), which is regarded as the gold standard for quantification of CBF. METHODS: Perfusion MRI and PET were performed on seven healthy men. In the PET study, the CBF image was obtained with bolus injection of H2(15)O and continuous arterial sampling. In the DSC-MRI study with bolus injection of Gd-based contrast agent, dynamic perfusion data were obtained with a 1.5T scanner at 1-s intervals by means of gradient-echo echo-planar imaging. CBF was determined by the SVD deconvolution method with and without correction for the tracer delay. Region-of-interest measurements were obtained in the gray matter (cerebral cortex in the middle cerebral artery territory) and white matter (centrum semiovale). RESULTS: Tracer delay was significantly longer in white matter than in gray matter (1.45+/-0.61 s vs. 0.59+/-0.35 s, P<0.01). Correction for the delay increased SVD-CBF in the white matter and consequently reduced the gray-to-white SVD-CBF ratio. The uncorrected gray-to-white SVD-CBF ratio was significantly larger than that of PET-CBF (3.33+/-0.66 vs. 2.54+/-0.49, P<0.01). However, the gray-to-white delay-corrected SVD-CBF ratio did not differ significantly from that of PET-CBF (2.83+/-0.31 vs. 2.54+/-0.49, P=0.10). CONCLUSION: The tracer delay in DSC-MRI causes errors in CBF estimates, even in healthy persons, and therefore should be corrected for when delay-sensitive deconvolution, such as SVD deconvolution, is used.

Metabolic penumbra of acute brain infarction: a correlation with infarct growth.

Volume expansion associated with brain infarction occurs in perfusion-diffusion mismatch of magnetic resonance imaging. We aimed at elucidating the metabolic impairment of this phenomenon with (15)O positron emission tomography and perfusion and diffusion magnetic resonance imaging. Eleven patients with acute unilateral embolic occlusion of the internal carotid or middle cerebral artery were studied within 6 hours of onset. Regional cerebral blood flow and cerebral metabolic rate of oxygen (CMRO(2)) were compared with those in the contralateral cerebral hemisphere. The relative apparent diffusion coefficient of water was estimated as a marker of cytotoxic edema. Relative cerebral blood flow and relative CMRO(2) in an evolving infarct (normal diffusion initially, but abnormal on day 3) were significantly (p < 0.05) less than those in the periinfarct area (normal diffusion initially and on day 3). The relative apparent diffusion coefficient between the evolving infarct and periinfarct showed no significant difference. These findings indicated that the initial 3-day volume expansion of an embolic brain infarction was associated with disturbed CMRD(2) but not with cytotoxic edema as early as 6 hours after onset. The "metabolic penumbra" defined as normal water diffusion with depressed CMRO(2) is a target to reduce the volume expansion of brain infarction.

Tracer delay correction of cerebral blood flow with dynamic susceptibility contrast-enhanced MRI.

Cerebral blood flow (CBF) and vascular mean transit time (MTT) can be determined by dynamic susceptibility contrast-enhanced magnetic resonance imaging and deconvolution with an arterial input function. However, deconvolution by a singular value decomposition (SVD) method is sensitive to the tracer delay that often occurs in patients with cerebrovascular disease. We investigated the effect of tracer delay on CBF determined by SVD deconvolution. Simulation study showed that underestimation of CBF due to tracer delay was larger for shorter MTTs. We developed a delay correction method that determines tracer delay by means of least-squares fitting pixel-by-pixel. The corrected CBF was determined by SVD deconvolution after time-shifting of the measured concentration curve. The simulations showed that the corrected CBF was insensitive to tracer delay irrespective of the vascular model, although CBF fluctuation increased slightly. We applied the delay correction to the CBF and MTT images acquired for nine patients with hyperacute stroke and unilateral occlusion of the middle cerebral artery. We found in some patients that the delay correction modulated the contrast of CBF and MTT images. For hyperacute stroke patients, tracer delay correction is essential to obtain reliable perfusion image when SVD deconvolution is used.