Extracellular Volume Fraction by Computed Tomography Predicts Prognosis After Transcatheter Aortic Valve Replacement.

BACKGROUND: Extracellular volume fraction (ECV) on magnetic resonance imaging can predict prognosis after aortic valve replacement in patients with aortic stenosis (AS). However, the usefulness of ECV on computed tomography (CT) for patients who have undergone transcatheter aortic valve replacement (TAVR) is unclear, so we investigated whether ECV analysis on CT is associated with clinical outcomes in TAVR candidates.Methods and Results: We analyzed 127 patients with severe AS who underwent preoperative CT for TAVR. We evaluated the utility of ECV analysis on single-energy CT for predicting patient prognosis after TAVR. The primary outcome was a composite of all-cause death and hospitalization due to heart failure (HF) after TAVR. 15 patients (12%) had composite outcomes: 4 deaths and 11 hospitalizations due to HF. In multivariate survival analysis using the Cox proportional hazard model, atrial fibrillation (AF) (hazard ratio (HR), 7.86; 95% confidence interval (CI), 2.57-24.03; P<0.001), history of congestive HF (HR, 4.91; 95% CI, 1.49-16.2; P=0.009) and ECV ≥32.6% on CT (HR, 6.96; 95% CI, 1.92-25.12; P=0.003) were independent predictors of composite outcomes. On Kaplan-Meier analysis, the higher ECV group (≥32.6%) had a significantly greater number of composite outcomes than the lower ECV group (P<0.001). CONCLUSIONS: ECV on CT is an independent predictor of prognosis after TAVR.

Evaluation of extracellular volume by computed tomography is useful for prediction of prognosis in dilated cardiomyopathy.

Cardiac computed tomography (CT) is useful for the screening of coronary artery stenosis, and extracellular volume fraction (ECV) analysis by CT using new dedicated software is now available. Here, we evaluated the utility of ECV analysis using cardiac CT to predict patient prognosis in cases with dilated cardiomyopathy (DCM). We analyzed 70 cases with DCM and cardiac computed tomography (CT) with available late-phase images. We evaluated the ECV of the left ventricular myocardium (LVM) using commercially available software (Ziostation 2, Ziosoft Inc, Japan). ECV on LVM was 33.96 ± 5.04%. Major adverse cardiac events (MACE) occurred in 21 cases (30%). ECV of the LVM on CT, endo-systolic volume, and rate of significant valvular disease were significantly higher in cases with MACE than in those without (37.16 ± 5.91% vs. 32.59 ± 3.95%, 194 ± 109 vs. 138 ± 78 ml and 57% vs. 20%, all P values < 0.05). LVEF was significantly lower in cases with MACE than in those without (23 ± 8 vs. 31 ± 11%, P = 0.0024). The best cut-off value of ECV on LVM for prediction of MACE was 32.26% based on receiver operating characteristics analysis. Cases with ECV ≥ 32.26% had significantly higher MACE based on Kaplan-Meier analysis (P = 0.0032). Only ECV on LVM was an independent predictor of MACE based on a multivariate Cox proportional hazards model (P = 0.0354). Evaluation of ECV on LVM by CT is useful for predicting MACE in patients with DCM.

Deep learning-based coronary computed tomography analysis to predict functionally significant coronary artery stenosis.

Fractional flow reserve derived from coronary CT (FFR-CT) is a noninvasive physiological technique that has shown a good correlation with invasive FFR. However, the use of FFR-CT is restricted by strict application standards, and the diagnostic accuracy of FFR-CT analysis may potentially be decreased by severely calcified coronary arteries because of blooming and beam hardening artifacts. The aim of this study was to evaluate the utility of deep learning (DL)-based coronary computed tomography (CT) data analysis in predicting invasive fractional flow reserve (FFR), especially in cases with severely calcified coronary arteries. We analyzed 184 consecutive cases (241 coronary arteries) which underwent coronary CT and invasive coronary angiography, including invasive FFR, within a three-month period. Mean coronary artery calcium scores were 963 ± 1226. We evaluated and compared the vessel-based diagnostic accuracy of our proposed DL model and a visual assessment to evaluate functionally significant coronary artery stenosis (invasive FFR < 0.80). A deep neural network was trained with consecutive short axial images of coronary arteries on coronary CT. Ninety-one coronary arteries of 89 cases (48%) had FFR-positive functionally significant stenosis. On receiver operating characteristics (ROC) analysis to predict FFR-positive stenosis using the trained DL model, average area under the curve (AUC) of the ROC curve was 0.756, which was superior to the AUC of visual assessment of significant (≥ 70%) coronary artery stenosis on CT (0.574, P = 0.011). The sensitivity, specificity, positive and negative predictive value (PPV and NPV), and accuracy of the DL model and visual assessment for detecting FFR-positive stenosis were 82 and 36%, 68 and 78%, 59 and 48%, 87 and 69%, and 73 and 63%, respectively. Sensitivity and NPV for the prediction of FFR-positive stenosis were significantly higher with our DL model than visual assessment (P = 0.0004, and P = 0.024). DL-based coronary CT data analysis has a higher diagnostic accuracy for functionally significant coronary artery stenosis than visual assessment.

[Nationwide Questionnaire Survey on Disaster Power Outage Countermeasures in the Radiology Departments of Hospitals].

PURPOSE: To identify the countermeasures and current status of disaster power outages in the radiology departments of hospitals. METHODS: A web-based questionnaire survey of 600 hospitals nationwide was conducted. The questionnaire survey covered 34 items, including availability of power in the radiology department in the event of a disaster and the impact of power outages on medical equipment in the radiology department. RESULTS: In all, 242 facilities (40.3%) responded to our survey. During power outages, 55.8%-68.2% of facilities were able to use CT, digital radiography, and angiography systems with their private generators. In 28.1%-40.7% of facilities, medical information systems were not available in all laboratories. In addition, power outages caused equipment malfunctions in 81.4% of facilities' radiology departments. CONCLUSION: We have identified the power supplied by private generators to the radiology department's medical equipment and medical information systems. Many medical equipment have malfunctioned due to power outages. Therefore, drills should be conducted to simulate various situations caused by power outages.

Four-dimensional computed tomography is useful for detection of apical sparing of cardiac amyloidosis.

A 73-year-old male was admitted because of recurrent syncope. He was diagnosed with transient bradycardia caused by a 2:1 atrioventricular block, and he underwent cardiac computed tomography (CT) using 320 detector-row CT to screen for coronary artery disease. Significant coronary artery stenosis was not detected, but diffuse late iodinate enhancement was found on the epi-myocardium and endo-myocardium of the interventricular septum, and endo-myocardium of the anterior and lateral left ventricular (LV) myocardium (LVM) on CT. The ejection fraction and global longitudinal strain (LS) of LVM were 53.97% and - 9.87% on CT. Apical sparing was present, meaning the LS of LV apical segments were preserved compared with basal segments on CT. Pathological findings of LVM demonstrated loss of myocardial cells and extra-cellular amyloid deposition on the direct fast scarlet staining. He was finally diagnosed with transthyretin amyloidosis.

Strong Diagnostic Performance of Single Energy 256-row Multidetector Computed Tomography with Deep Learning Image Reconstruction in the Assessment of Myocardial Fibrosis.

Objective Although magnetic resonance imaging (MRI) is the gold standard for evaluating abnormal myocardial fibrosis and extracellular volume (ECV) of the left ventricular myocardium (LVM), a similar evaluation has recently become possible using computed tomography (CT). In this study, we investigated the diagnostic accuracy of a new 256-row multidetector CT with a low tube-voltage single energy scan and deep-learning-image reconstruction (DLIR) in detecting abnormal late enhancement (LE) in LVM. Methods We evaluated the diagnostic performance of CT for detecting LE in LVM and compared the results with those of MRI as a reference. We also measured the ECV of the LVM on CT and compared the results with those on MRI. Patients or Materials We analyzed 50 consecutive patients who underwent cardiac CT, including a late-phase scan and MRI, within three months of suspected cardiomyopathy. All patients underwent 256-slice CT (Revolution CT Apex; GE Healthcare) with a low tube-voltage (70 kV) single energy scan and DLIR for a late-phase scan. Results In patient- and segment-based analyses, the sensitivity, specificity, and accuracy of detection of LE on CT were 94% and 85%, 100% and 95%, and 96% and 93%, respectively. The ECV of LVM per patient on CT and MRI was 33.0% ±6.2% and 35.9% ±6.1%, respectively. These findings were extremely strongly correlated, with a correlation coefficient of 0.87 (p <0.0001). The effective radiation dose on late-phase scanning was 2.4±0.9 mSv. Conclusion The diagnostic performance of 256-row multislice CT with a low tube voltage and DLIR for detecting LE and measuring ECV in LVM is credible.

An Increased Diagnostic Accuracy of Significant Coronary Artery Stenosis Using 320-slice Computed Tomography with Model-based Iterative Reconstruction in Cases with Severely Calcified Coronary Arteries.

Objective High-quality images can be obtained with 320-slice computed tomography (CT) with model-based iterative reconstruction (MBIR). We therefore investigated the diagnostic accuracy of 320-slice CT with MBIR for detecting significant coronary artery stenosis. Methods This was a retrospective study of 160 patients who underwent coronary CT and invasive coronary angiography (ICA). The first 100 consecutive patients (Group 1) underwent 320-slice CT without MBIR or small-focus scanning. The next 60 consecutive patients (Group 2) underwent 320-slice CT with both MBIR and small-focus scanning. Patients who underwent coronary artery bypass surgery were excluded. The diagnostic performance of 320-slice CT without MBIR or small-focus scanning and 320-slice CT with both of them, with ICA regarded as a reference standard, was compared to detect significant coronary artery stenosis (≥70% on CT, ≥75% on ICA). Results In a patient-based analysis, the sensitivity, specificity, and overall accuracy of detection of significant stenosis on CT against ICA were 95%, 85%, and 91% in Group 1, and 93%, 83%, and 90% in Group 2, respectively. No significant differences were observed between the two groups in the patient- and segment-based analyses. However, among cases with a severe coronary artery calcium score >400 (31 cases in Group 1 and 28 in Group 2), the specificity and overall accuracy were significantly higher (all p<0.01) in Group 2 than in Group 1 according to the segment-based analysis. Conclusion The diagnostic accuracy of the detection of coronary artery stenosis on CT was improved using 320-slice CT with MBIR.

PHANTOM STUDY OF THE ENTRANCE SURFACE DOSE OF A NEW CT SCOUT ACQUISITION THAT ALSO SERVES AS A TUBE WARM-UP.

We determined the effect of a new scout image acquisition technique ('smart scout'), which also serves as a tube warm-up, on radiation dose and automatic exposure control (AEC) mA settings. The entrance surface dose (ESD) of a chest phantom with and without the smart scout was measured. A conical AEC phantom was scanned in the setting for abdominal CT, and AEC curves were generated. ESD when the smart scout was not used was 0.75 mGy at 120 kV, 50 mA. ESD when using the smart scout was 0.24 mGy for a body habitus setting of 'Less', 0.54 mGy for 'Moderate' and 0.95 mGy for 'More'. When the diameter of the subject was ≥32 cm, the mA setting became lower in 'Less'. The smart scout reduced exposure at the 'Less' and Moderate' settings compared to the conventional scout scan.

Head CT dose reduction with organ-based tube current modulation.

BACKGROUND: A helical head CT examination uses a pitch factor (PF) of <1.0, resulting in a part of the slice being directly irradiated twice. This raises the possibility of double irradiation, which may increase the amount of radiation to the lens. Organ-based tube current modulation (OBTCM) is an effective method for reducing lens exposure because it reduces the dose to the anterior aspect of the patient. However, it is challenging to visualize the complex dose distribution when factoring in double irradiation. PURPOSE: To visualize twice-irradiated areas in helical head CT in three dimensions and to clarify the exposure reduction effect of OBTCM. MATERIAL AND METHODS: A leuco crystal violet (LCV) dosimeter was placed into an empty polyethylene terephthalate bottle 16.5 cm in diameter. Helical scans were performed without and with OBTCM using the following parameters: tube voltage 120 kV, tube current 600 mA, pitch factor 0.637, rotation time 0.5 s, 80 (detector rows) × 0.5 mm (detector collimation), and ten scans. Exposed areas were visualized using an optical computed tomography (OCT) system designed by our group. The dose reduction rate of OBTCM was defined as the ratio of the average values of the histogram with the dose value on the x-axis and the frequency on the y-axis without and with OBTCM at 90° to the anterior midline. RESULTS: The LCV dosimeter visualized the spiral-shaped twice-irradiated areas. Double irradiation resulted in a dose of 2.19/1.90 and 1.38/1.19 Gy (15.0% and 15.9% increase) without and with OBTCM, respectively. The dose reduction using OBTCM was 29.6% at 90° anterolateral. CONCLUSION: The LCV dosimeter visualized the complex three-dimensional irradiated areas and enabled dose measurement in twice-irradiated areas. Increased exposure from double irradiation was attenuated by OBTCM.

[Measurement of Absorbed Dose in the Air in X-ray CT Examination Rooms Using a Special Protective Shield for CT].

PURPOSE: X-ray CT examinations are required not only in routine medical examinations but also in various situations such as emergency medical care. Although medical staff may be exposed to radiation when assisting patients, the distribution of air-absorbed doses in the CT examination room when using a special protective shield for CT has not been clarified. Here, we measured air-absorbed doses at several points simultaneously to clarify the distribution of these doses and the effect of a special protective shield for CT in reducing them. METHOD: A human phantom was imaged with an X-ray CT system. The absorbed dose in the air dose profile distribution was measured with an OSL dosimeter in the presence and absence of a special protective shield for CT. RESULTS: The highest air absorbed doses of 4.27 mGy were at 0 cm in the horizontal direction, 120 cm in the vertical direction, and 50 cm in the body axis direction. The largest reduction in air absorbed dose following installation of the special protective shield for CT was 91.7%, obtained at 0 cm in the horizontal direction, 150 cm in the vertical direction, and 50 cm in the body axis direction. CONCLUSION: A 91.7% reduction in air-absorbed dose was o directly behind the special protective shield for CT. The reduction in air-absorbed dose was 65.8% at the location of a gap between the special protective shield for CT and gantry.

Celiac Artery Dissection and Retroperitoneal Hemorrhage in Median Arcuate Ligament Syndrome Treated With a Stent and Transcatheter Arterial Embolization: Preprocedural 4-Dimensional Computed Tomography Angiography Assessment.

A 61-year-old man presented with retroperitoneal hemorrhage caused by an aneurysm rupture of the pancreaticoduodenal arcade (PDA), and acute celiac artery dissection distal to celiac axis stenosis. Owing to the gradual growth of the false lumen, we planned to deploy a stent to the celiac artery dissection and embolize the PDA aneurysm. Prior to stent placement, we assessed the acute celiac artery dissection distal to the stenosis using four-dimensional computed tomography (CT) angiography through expiration/inspiration/expiration cycle. We diagnosed median arcuate ligament syndrome considering that the celiac axis showed a hooked narrowing at end-expiration, and the compression decreased at end-inspiration. Additionally, the true lumen distal to the stretched axis dilated in the inspiration phase. Therefore, we could advance a catheter into the true lumen during inspiration and successfully deploy a stent. Subsequently, laparoscopic median arcuate ligament release was performed after the stent deployment. A postoperative CT scan showed good patency in the stent, with disappearance of the blood filling the false lumen and with reduced celiac axis stenosis.

[Calculation of Lens Exposure Reduction Using Organ-effective Modulation in Pediatric Head CT].

PURPOSE: Using a pediatric head phantom constructed in our department, we examined a method to reduce exposure by using organ-effective modulation (OEM; Toshiba Medical Systems Corporation, Tochigi) to tilt the gantry during pediatric head computed tomography (CT) scanning. METHOD: The radiation reduction and CT image standard deviation (SD) were measured at gantry angles at which the orbit was slightly irradiated, partially irradiated, and completely irradiated. The OEM incident surface dose reduction rate was measured using an automatic exposure control (AEC) phantom with a diameter of 6-18 cm. RESULTS: The lens surface dose reduction rate using OEM was 21.2%. When the gantry was tilted and the orbit was completely out of the scanning range, the rate of reduction was 47.8%. OEM incident surface dose reduction rates were 27.4% for a phantom diameter of 18 cm, 22.0% for that of 16 cm, 17.8% for that of 14 cm, 17.2% for that of 12 cm, 8.4% for that of 10 cm, and 0% for that of 8 cm and 6 cm. OEM effectiveness decreased with decreasing phantom diameter. The use of OEM increased the rate of change of SD by 1.25´ when the gantry inclination was 0°, 1.27´ when the gantry inclination was 10°, and 1.27´ when the gantry inclination was 20°in the 12 o'clock position. CONCLUSION: The degree of reduction in exposure dose to the lens in pediatric head CT imaging was 47.8% by completely removing the lens from the irradiation range using gantry tilt and 21.2% by using OEM. The effect of OEM changed in proportion to tube current. The exposure reduction effect of the OEM decreases with decreasing head size, indicating its reduced effectiveness in head CT scans of smaller infants.

Evaluation of Radiation Protection Methods for Assistant Staff during CT Imaging in High-energy Trauma: Lens Dosimetry with a Phantom Study.

ABSTRACT: Staff are exposed to radiation in the scanning room when assisting with CT scans of patients requiring ventilatory support during procedures. We measured lens doses using a phantom during a high-energy trauma protocol. Dosimetry showed that the unprotected lens received 2.02 mGy on the right and 1.91 mGy on the left, which are not negligible doses. Respective exposures to the right and left lens were 53.6% and 55.1% when wearing 0.07 mm Pb protective glasses with side covers; 53.7% and 64.2% when wearing 0.7 mm Pb glasses without side covers when facing away from the patient couch; and 92.1% and 91.2% using protective shielding in the gantry. Since the face direction may change during assistance with CT imaging, it is desirable that the protective glasses have a shape with a side cover. The protective shielding had a major radiation reduction effect, although it is expensive to acquire, install, and maintain.

Development of attenuation correction methods using deep learning in brain-perfusion single-photon emission computed tomography.

PURPOSE: Computed tomography (CT)-based attenuation correction (CTAC) in single-photon emission computed tomography (SPECT) is highly accurate, but it requires hybrid SPECT/CT instruments and additional radiation exposure. To obtain attenuation correction (AC) without the need for additional CT images, a deep learning method was used to generate pseudo-CT images has previously been reported, but it is limited because of cross-modality transformation, resulting in misalignment and modality-specific artifacts. This study aimed to develop a deep learning-based approach using non-attenuation-corrected (NAC) images and CTAC-based images for training to yield AC images in brain-perfusion SPECT. This study also investigated whether the proposed approach is superior to conventional Chang's AC (ChangAC). METHODS: In total, 236 patients who underwent brain-perfusion SPECT were randomly divided into two groups: the training group (189 patients; 80%) and the test group (47 patients; 20%). Two models were constructed using Autoencoder (AutoencoderAC) and U-Net (U-NetAC), respectively. ChangAC, AutoencoderAC, and U-NetAC approaches were compared with CTAC using qualitative analysis (visual evaluation) and quantitative analysis (normalized mean squared error [NMSE] and the percentage error in each brain region). Statistical analyses were performed using the Wilcoxon signed-rank sum test and Bland-Altman analysis. RESULTS: U-NetAC had the highest visual evaluation score. The NMSE results for the U-NetAC were the lowest, followed by AutoencoderAC and ChangAC (P < 0.001). Bland-Altman analysis showed a fixed bias for ChangAC and AutoencoderAC and a proportional bias for ChangAC. ChangAC underestimated counts by 30-40% in all brain regions. AutoencoderAC and U-NetAC produced mean errors of <1% and maximum errors of 3%, respectively. CONCLUSION: New deep learning-based AC methods for AutoencoderAC and U-NetAC were developed. Their accuracy was higher than that obtained by ChangAC. U-NetAC exhibited higher qualitative and quantitative accuracy than AutoencoderAC. We generated highly accurate AC images directly from NAC images without the need for intermediate pseudo-CT images. To verify our models' generalizability, external validation is required.

Multidimensional Deep Learning Reduces False-Positives in the Automated Detection of Cerebral Aneurysms on Time-Of-Flight Magnetic Resonance Angiography: A Multi-Center Study.

Current deep learning-based cerebral aneurysm detection demonstrates high sensitivity, but produces numerous false-positives (FPs), which hampers clinical application of automated detection systems for time-of-flight magnetic resonance angiography. To reduce FPs while maintaining high sensitivity, we developed a multidimensional convolutional neural network (MD-CNN) designed to unite planar and stereoscopic information about aneurysms. This retrospective study enrolled time-of-flight magnetic resonance angiography images of cerebral aneurysms from three institutions from June 2006 to April 2019. In the internal test, 80% of the entire data set was used for model training and 20% for the test, while for the external tests, data from different pairs of the three institutions were used for training and the remaining one for testing. Images containing aneurysms > 15 mm and images without aneurysms were excluded. Three deep learning models [planar information-only (2D-CNN), stereoscopic information-only (3D-CNN), and multidimensional information (MD-CNN)] were trained to classify whether the voxels contained aneurysms, and they were evaluated on each test. The performance of each model was assessed using free-response operating characteristic curves. In total, 732 aneurysms (5.9 ± 2.5 mm) of 559 cases (327, 120, and 112 from institutes A, B, and C; 469 and 263 for 1.5T and 3.0T MRI) were included in this study. In the internal test, the highest sensitivities were 80.4, 87.4, and 82.5%, and the FPs were 6.1, 7.1, and 5.0 FPs/case at a fixed sensitivity of 80% for the 2D-CNN, 3D-CNN, and MD-CNN, respectively. In the external test, the highest sensitivities were 82.1, 86.5, and 89.1%, and 5.9, 7.4, and 4.2 FPs/cases for them, respectively. MD-CNN was a new approach to maintain sensitivity and reduce the FPs simultaneously.

Improved Diagnostic Performance of New-generation 320-slice Computed Tomography with Forward-projected Model-based Iterative Reconstruction SoluTion for the Assessment of Late Enhancement in Left Ventricular Myocardium.

Objective Forward-projected Model-based Iterative Reconstruction SoluTion (FIRST) is a novel reconstruction method. We investigated the improvement in the diagnostic performance for the detection of abnormal late enhancement (LE) in left ventricular myocardium (LVM) using a new-generation 320-slice computed tomography (CT) device with FIRST. Methods This is a retrospective study that included 100 adult patients who underwent cardiac CT including a late phase scan and magnetic resonance imaging (MRI) within 3 months. The first 50 consecutive patients (first-generation group) underwent first-generation 320-slice CT without FIRST, and the next 50 consecutive patients (second-generation group) underwent second-generation 320-slice CT with FIRST. We compared the diagnostic performance of the first- and second-generation 320-slice CT with FIRST with MRI as a reference standard to detect LE in LVM. Results In the patient-based analysis, the sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of detection of LE on CT were 79%, 90%, 92%, 76%, and 84%, respectively, in the first-generation group and 97%, 84%, 91%, 94%, and 92%, respectively, in the second-generation group. The sensitivity was significantly higher in the second-generation group than in the first-generation group (p=0.049). In the segment-based analysis, the sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of detection of LE on CT were 69%, 96%, 83%, 92%, and 90%, respectively, in the first-generation group and 87%, 94%, 84%, 95%, and 92%, respectively, in the second-generation group. The sensitivity and negative predictive value were significantly higher in the second-generation group than in the first-generation group (p<0.001 and p=0.016). The contrast-noise ratio was significantly higher in the second-generation group than in the first-generation group (5.6±1.7 vs. 2.8±1.1, p<0.001), and the radiation dose for the assessment of LE on CT was significantly higher in the first-generation group than in the second-generation group (4.7±2.7 mSv vs. 2.3±0.1 mSv, p<0.001). Conclusion The diagnostic performance for the detection of LE in LVM significantly improved with the use of second-generation 320-slice CT and FIRST.

Utility of computed tomography in cases of aortic valve stenosis before and after transcatheter aortic valve implantation.

Trans-catheter aortic valve implantation (TAVI) has been recognized as a useful treatment for patients with severe aortic valve stenosis, particularly those with moderate to high risks of open heart surgery. A thorough evaluation of the aortic valve complex, including the size or presence of calcifications of the leaflets and annulus, is important for the selection of appropriate candidates, artificial valve types and approach. Echocardiography is useful for the precise evaluation of aortic valve stenosis severity and aortic valve complex morphology, but it is not useful to evaluate three-dimensional aortic valve anatomy and pathway for the catheter of aortic valve implantation. Electrocardiography (ECG)-gating computed tomography (CT) has recently been recognized as a useful modality for evaluating significant coronary artery stenosis because of its higher spatial and temporal resolution and diagnostic accuracy based on recent studies. ECG-gating CT is also useful for evaluating aortic valve complex morphology, including calcifications and whole aorta and iliac arteries, as the access route of catheter in TAVI. TAVI candidates, who are at high risk of open surgery, tend to be old and require anti-platelet after TAVI; therefore CT, is also useful for screening for non-cardiac diseases including malignant tumors just before TAVI. Therefore, here we introduce the utility of cardiac and whole body CT in cases of severe aortic valve stenosis before and after TAVI.

A Deep Convolutional Neural Network With Performance Comparable to Radiologists for Differentiating Between Spinal Schwannoma and Meningioma.

STUDY DESIGN: Retrospective analysis of magnetic resonance imaging (MRI). OBJECTIVE: The aim of this study was to evaluate the performance of our convolutional neural network (CNN) in differentiating between spinal schwannoma and meningioma on MRI. We compared the performance of the CNN and that of two expert radiologists. SUMMARY OF BACKGROUND DATA: Preoperative discrimination between spinal schwannomas and meningiomas is crucial because different surgical procedures are required for their treatment. A deep-learning approach based on CNNs is gaining interest in the medical imaging field. METHODS: We retrospectively reviewed data from patients with spinal schwannoma and meningioma who had undergone MRI and tumor resection. There were 50 patients with schwannoma and 34 patients with meningioma. Sagittal T2-weighted magnetic resonance imaging (T2WI) and sagittal contrast-enhanced T1-weighted magnetic resonance imaging (T1WI) were used for the CNN training and validation. The deep learning framework Tensorflow was used to construct the CNN architecture. To evaluate the performance of the CNN, we plotted the receiver-operating characteristic (ROC) curve and calculated the area under the curve (AUC). We calculated and compared the sensitivity, specificity, and accuracy of the diagnosis by the CNN and two board-certified radiologists. RESULTS: . The AUC of ROC curves of the CNN based on T2WI and contrast-enhanced T1WI were 0.876 and 0.870, respectively. The sensitivity of the CNN based on T2WI was 78%; 100% for radiologist 1; and 95% for radiologist 2. The specificity was 82%, 26%, and 42%, respectively. The accuracy was 80%, 69%, and 73%, respectively. By contrast, the sensitivity of the CNN based on contrast-enhanced T1WI was 85%; 100% for radiologist 1; and 96% for radiologist 2. The specificity was 75%, 56, and 58%, respectively. The accuracy was 81%, 82%, and 81%, respectively. CONCLUSION: We have successfully differentiated spinal schwannomas and meningiomas using the CNN with high diagnostic accuracy comparable to that of experienced radiologists. LEVEL OF EVIDENCE: 4.