Honors & Awards
Reviewer of the Year Award, European Radiology Experimental (ER-X) (2019)
Editors’ Recognition Awards for Distinction in Reviewing, Journal of Thoracic Imaging (JTI) (2018)
Invest in the Youth Programme - European Congress of Radiology (ECR), European Society of Radiology (ESR) (2018)
ISMRM Trainee Research Prize, International Society for Magnetic Resonance in Medicine (ISMRM) (2017)
AHA-CVRI Young Investigator Award Winner, North American Society for Cardiovascular Imaging (NASCI) (2017)
Student Travel Award, Radiological Society of North America (RSNA) (2017)
Boards, Advisory Committees, Professional Organizations
Editorial Board Member, European Radiology Experimental (2019 - Present)
Residency, "G. d'Annunzio University" - Chieti, Italy, Radiology (2018)
Doctor of Medicine, Univ Degli Studi G D'Annunzio (2012)
Cost-effectiveness of dual-energy CT versus multiphasic single-energy CT and MRI for characterization of incidental indeterminate renal lesions.
Abdominal radiology (New York)
PURPOSE: To evaluate the cost-effectiveness of DECT versus multiphasic CT and MRI for characterizing small incidentally detected indeterminate renal lesions using a Markov Monte Carlo decision-analytic model.BACKGROUND: Incidental renal lesions are commonly encountered due to the increasing utilization of medical imaging and the increasing prevalence of renal lesions with age. Currently recommended imaging modalities to further characterize incidental indeterminate renal lesions have some inherent drawbacks. Single-phase DECT may overcome these limitations, but its cost-effectiveness remains uncertain.MATERIALS AND METHODS: A decision-analytic (Markov) model was constructed to estimate life expectancy and lifetime costs for otherwise healthy 64-year-old patients with small (≤4cm) incidentally detected, indeterminate renal lesions on routine imaging (e.g., ultrasound or single-phase CT). Three strategies for evaluating renal lesions for enhancement were compared: multiphase SECT (e.g., true unenhanced and nephrographic phase), multiphasic MRI, and single-phase DECT (nephrographic phase in dual-energy mode). The model incorporated modality-specific diagnostic test performance, incidence, and prevalence of incidental renal cell carcinomas (RCCs), effectiveness, costs, and health outcomes. An incremental cost-effectiveness analysis was performed to identify strategy preference at willingness-to-pay (WTP) thresholds of $50,000 and $100,000 per quality-adjusted life-year (QALY) gained. Deterministic and probabilistic sensitivity analysis were performed.RESULTS: In the base case analysis, expected mean costs per patient undergoing characterization of incidental renal lesions were $2567 for single-phase DECT, $3290 for multiphasic CT, and $3751 for multiphasic MRI. Associated quality-adjusted life-years were the highest for single-phase DECT at 0.962, for multiphasic MRI it was 0.940, and was the lowest for multiphasic CT at 0.925. Because of lower associated costs and higher effectiveness, the single-phase DECT strategy dominated the other two strategies.CONCLUSIONS: Single-phase DECT is potentially more cost-effective than multiphasic SECT and MRI for evaluating small incidentally detected indeterminate renal lesions.
View details for DOI 10.1007/s00261-019-02380-x
View details for PubMedID 31894384
- Computed Tomographic Angiography-Based Fractional Flow Reserve Compared With Catheter-Based Dobutamine-Stress Diastolic Fractional Flow Reserve in Symptomatic Patients With a Myocardial Bridge and No Obstructive Coronary Artery Disease. Circulation. Cardiovascular imaging 2020; 13 (2): e009576
AppendiXNet: Deep Learning for Diagnosis of Appendicitis from A Small Dataset of CT Exams Using Video Pretraining.
2020; 10 (1): 3958
The development of deep learning algorithms for complex tasks in digital medicine has relied on the availability of large labeled training datasets, usually containing hundreds of thousands of examples. The purpose of this study was to develop a 3D deep learning model, AppendiXNet, to detect appendicitis, one of the most common life-threatening abdominal emergencies, using a small training dataset of less than 500 training CT exams. We explored whether pretraining the model on a large collection of natural videos would improve the performance of the model over training the model from scratch. AppendiXNet was pretrained on a large collection of YouTube videos called Kinetics, consisting of approximately 500,000 video clips and annotated for one of 600 human action classes, and then fine-tuned on a small dataset of 438 CT scans annotated for appendicitis. We found that pretraining the 3D model on natural videos significantly improved the performance of the model from an AUC of 0.724 (95% CI 0.625, 0.823) to 0.810 (95% CI 0.725, 0.895). The application of deep learning to detect abnormalities on CT examinations using video pretraining could generalize effectively to other challenging cross-sectional medical imaging tasks when training data is limited.
View details for DOI 10.1038/s41598-020-61055-6
View details for PubMedID 32127625
- Artificial intelligence machine learning-based coronary CT fractional flow reserve (CT-FFRML): Impact of iterative and filtered back projection reconstruction techniques JOURNAL OF CARDIOVASCULAR COMPUTED TOMOGRAPHY 2019; 13 (6): 331–35
Prognostic value of CT myocardial perfusion imaging and CT-derived fractional flow reserve for major adverse cardiac events in patients with coronary artery disease.
Journal of cardiovascular computed tomography
The purpose of this study was to analyze the prognostic value of dynamic CT perfusion imaging (CTP) and CT derived fractional flow reserve (CT-FFR) for major adverse cardiac events (MACE).81 patients from 4 institutions underwent coronary computed tomography angiography (CCTA) with dynamic CTP imaging and CT-FFR analysis. Patients were followed-up at 6, 12, and 18 months after imaging. MACE were defined as cardiac death, nonfatal myocardial infarction, unstable angina requiring hospitalization, or revascularization. CT-FFR was computed for each major coronary artery using an artificial intelligence-based application. CTP studies were analyzed per vessel territory using an index myocardial blood flow, the ratio between territory and global MBF. The prognostic value of CCTA, CT-FFR, and CTP was investigated with a univariate and multivariate Cox proportional hazards regression model.243 vessels in 81 patients were interrogated by CCTA with CT-FFR and 243 vessel territories (1296 segments) were evaluated with dynamic CTP imaging. Of the 81 patients, 25 (31%) experienced MACE during follow-up. In univariate analysis, a positive index-MBF resulted in the largest risk for MACE (HR 11.4) compared to CCTA (HR 2.6) and CT-FFR (HR 4.6). In multivariate analysis, including clinical factors, CCTA, CT-FFR, and index-MBF, only index-MBF significantly contributed to the risk of MACE (HR 10.1), unlike CCTA (HR 1.2) and CT-FFR (HR 2.2).Our study provides initial evidence that dynamic CTP alone has the highest prognostic value for MACE compared to CCTA and CT-FFR individually or a combination of the three, independent of clinical risk factors.
View details for DOI 10.1016/j.jcct.2019.02.005
View details for PubMedID 30796003
Tumor detectability and conspicuity comparison of standard b1000 and ultrahigh b2000 diffusion-weighted imaging in rectal cancer.
Abdominal radiology (New York)
To compare tumor detectability and conspicuity of standard b = 1000 s/mm2 (b1000) versus ultrahigh b = 2000 s/mm2 (b2000) diffusion-weighted imaging (DWI) in rectal cancer.Fifty-five patients for a total of 81 3T DWI-MR scans were retrospectively evaluated by two differently experienced readers. A comparison between b1000 and b2000 for tumor detectability and conspicuity was performed. The conspicuity was qualitatively and quantitatively assessed by using three-point scale and whole tumor volume manual delineation, respectively. Receiver-operating characteristic curve (ROC) with area under the curve (AUC) analysis provided diagnostic accuracy in tumor detectability of restaging MR scans. Qualitative scores and quantitative features including mean signal intensity, variance, 10th percentile and 90th percentile, were compared using the Wilcoxon test. Interobserver agreement (IOA) for qualitative and quantitative data was calculated using Cohen's Kappa and intraclass correlation coefficient (ICC) respectively.Diagnostic accuracy was comparable between b1000 and b2000 for both readers (p > 0.05). Overall quality scores were significantly better for b2000 than b1000 (2.29 vs 1.65 Reader 1, p = 0.01; 2.18 vs 1.69 Reader 2, p = 0.04). IOA was equally good for both b values (k = 0.86 b1000, k = 0.86 b2000). Quantitative analysis revealed more uniform signal (measured in variance) of b2000 in both healthy surrounding tissue (p < 0.05) and tumor (p < 0.05), with less outliers (measured using 10th and 90th percentile). Additionally, b2000 offered lower mean signal intensity in tissue sorrounding the tumor (p < 0.05). Finally, ICC improved from 0.92 (b1000) to 0.97 (b2000).Ultrahigh b value (b2000) may improve rectal cancer conspicuity and introbserver agreement maintaining comparable diagnostic accuracy to standard b1000.
View details for DOI 10.1007/s00261-019-02177-y
View details for PubMedID 31444557
Prenatal planning of placenta previa: diagnostic accuracy of a novel MRI-based prediction model for placenta accreta spectrum (PAS) and clinical outcome.
Abdominal radiology (New York)
PURPOSE: To investigate the diagnostic accuracy of MRI for placenta accreta spectrum (PAS) and clinical outcome prediction in women with placenta previa, using a novel MRI-based predictive model.METHODS: Thirty-eight placental MRI exams performed on a 1.5T scanner were retrospectively reviewed by two radiologists in consensus. The presence of T2 dark bands, myometrial thinning, abnormal vascularity, uterine bulging, placental heterogeneity, placental protrusion sign, placental recess, and percretism signs was scored using a 5-point scale. Pathology and clinical intrapartum findings were the standard of reference for PAS, while intrapartum/peripartum bleeding and emergency hysterectomy defined the clinical outcome. Receiver-operating characteristic (ROC) analysis and discriminant function analysis were performed to test the predictive power of MRI findings for both PAS and clinical outcome prediction.RESULTS: Abnormal vascularity and percretism signs were the two most predictive MRI features of PAS. The area under the curve (AUC) of the predictive function was 0.833 (cutoff 0.39, 67% sensitivity, 100% specificity, p=0.001). Percretism signs and myometrial thinning were the two most predictive MRI features of poor outcome. AUC of the predictive function was 0.971 (cutoff -0.55, 100% sensitivity, 77% specificity, p<0.001).CONCLUSION: The diagnostic accuracy of MRI, especially considering the combination of the most predictive MRI findings, is higher when the target of the prediction is the clinical outcome rather than the PAS.
View details for PubMedID 30600374
Dual-Energy CT of the Pancreas.
Seminars in ultrasound, CT, and MR
2019; 40 (6): 509–14
This article explores the technical background of dual-energy CT (DECT) imaging along with its basic principles, before turning to a review of the various DECT applications specific to pancreatic imaging. In light of the most recent literature, we will review the constellation of DECT applications available for pancreatic imaging in both oncologic and non-oncologic applications. We emphasize the increased lesion conspicuity and the improved tissue characterization available with DECT post-processing tools. Finally, future clinical applications and opportunities for research will be overviewed.
View details for DOI 10.1053/j.sult.2019.05.002
View details for PubMedID 31806149
- Multiple liver pseudotumors due to hepatic steatosis and fatty sparing: A non-invasive imaging approach EUROPEAN JOURNAL OF RADIOLOGY OPEN 2019; 6: 56–59
Prevalence and Clinical Relevance of Extracardiac Findings in Cardiovascular Magnetic Resonance Imaging
JOURNAL OF THORACIC IMAGING
2019; 34 (1): 48–55
To assess the prevalence of extracardiac findings (ECF) during cardiovascular magnetic resonance (CMR) examinations and their downstream effect on clinical management.We retrospectively identified 500 consecutive patients. Trans-axial balanced steady-state free precession nongated images acquired from the upper thorax to the upper abdomen were evaluated independently by 2 radiologists. ECF were classified as nonsignificant (benign, with no need for further investigation), significant (mandatory to be reported/monitored), and major (clinically remarkable pathology, mandatory to be reported/investigated/treated). Fifteen-month clinical follow-up information was collected through hospital records.Of 500 patients, 108 (21.6%) showed a total of 153 ECF: 59 (11.8% of the entire study population; 38.5% of all ECF) nonsignificant, 76 (15.2%; 49.7%) significant, and 18 (3.6%; 11.8%) major ECF. The most frequent ECF were pleural effusion, hepatic cyst, renal cyst, and ascending aorta dilatation. Of 94 significant and major ECF, 46 were previously unknown and more common in older patients. Newly diagnosed major ECF (n=11, 2.2% of the entire study population, and 7.2% of all ECF)-including 5 tumors (1% of study population)-were confirmed by downstream evaluations and required specific treatment. Patients with major ECF were significantly older than patients without with major ECF. Newly diagnosed clinically significant and major ECF prompted downstream diagnostic tests in 44% and 100% of cases, respectively.The detection of significant and major ECF is common during CMR reporting. The knowledge and the correct identification of most frequent ECF enable earlier diagnoses and faster treatment initiation of unknown extracardiac pathologies in patients referred to CMR imaging.
View details for DOI 10.1097/RTI.0000000000000360
View details for Web of Science ID 000458277000010
View details for PubMedID 30142138
- The Multi-modality Cardiac Imaging Approach to Cardiac Sarcoidosis CURRENT MEDICAL IMAGING REVIEWS 2019; 15 (1): 10–20
- Cardiac implantable electronic devices and chemotherapy: A risky combination COR ET VASA 2018; 60 (5): E469–E471
Quantitative inversion time prescription for myocardial late gadolinium enhancement using T1-mapping-based synthetic inversion recovery imaging: reducing subjectivity in the estimation of inversion time
INTERNATIONAL JOURNAL OF CARDIOVASCULAR IMAGING
2018; 34 (6): 921–29
To develop a quantitative T1-mapping-based synthetic inversion recovery (IRsynth) approach to calculate the optimal inversion time (TI0) for late gadolinium enhancement (LGE) imaging. Prospectively enrolled patients (n = 130, 58 ± 16 years) underwent cardiac MRI on a 1.5T system including Look-Locker TI-scout (LL), modified LL IR (MOLLI)-based T1-mapping, and LGE acquisitions. Patients were randomized into two groups: LL group (TI-scout followed T1-mapping) or MOLLI group (T1-mapping followed TI-scout). In both groups, the second acquisition was used to determine the TI0 for LGE. IRsynth images were generated from T1-maps between TI = 200-400 ms in 5 ms increments. Image quality was rated on a 3-point scale and the remote/background signal intensity ratio (SIR) was calculated. In the LL group (n = 53), the TI-scout-based TI0 was significantly shorter compared to IRsynth [230 ms (219-242) vs. 280 ms (263-297), P < 0.0001]. The TI0 used for LGE was set 30-40 ms longer [261 ms (247-276), P < 0.0001] than the TI-scout-based TI0, resulting in a TI0 ~ 20 ms shorter than what was obtained by IRsynth (P = 0.0156). In the MOLLI group (n = 63), IRsynth-based TI0 was significantly longer than the TI-scout-based TI0 [298 ms (262-334) vs. 242 ms (217-267), P = 0.0313]. The quality of myocardial nulling was rated higher [2.4 (2.2-2.5) vs. 2.0 (1.8-2.1), P = 0.0042] and the remote/background SIR was found to be more optimal (1.6 [1.1-2.1] vs. 2.6 [1.8-3.3], P = 0.0256) in the MOLLI group. T1-based IRsynth selects TI0 for LGE more accurately than conventional TI-scout imaging. IRsynth improves TI0 selection by providing excellent visualization of the representative image contrast for LGE images, reducing operator dependence in LGE acquisition.
View details for DOI 10.1007/s10554-017-1294-9
View details for Web of Science ID 000437753200010
View details for PubMedID 29305739
Computer-assisted detection of acute pulmonary embolism at CT pulmonary angiography in children and young adults: a diagnostic performance analysis.
Acta radiologica (Stockholm, Sweden : 1987)
To diagnose pulmonary embolism (PE) in children and adults since evaluating tiny pulmonary vasculature beyond segmental level is a challenging and demanding task with thousands of images.To evaluate the effect of computer-assisted detection (CAD) on acute PE on CTPA in children and young adults by readers with varying experience levels.Six radiologists were retrospectively divided into three groups according to experience levels and assessed the CTPA studies on a per-emboli basis. All readers identified independently the PE presence, and ranked diagnostic confidence on a 5-point scale with and without CAD. Reading time, sensitivities, specificities, accuracies, positive predictive values (PPVs), and negative predictive values (NPVs) were calculated for each reading.The sensitivities and NPVs differed significantly in most readers ( P = 0.004, 0.001, 0.010, 0.010, and 0.012 for sensitivities and P = 0.011, 0.003, 0.016, 0.017, and 0.019 for NPVs) except for reader 6 ( P = 0.148 and 0.165, respectively), and the accuracies of all readers differed significantly (all P < 0.05) in peripheral PE (beyond segmental level) detection readings with CAD versus without CAD between two reading methods. The overall time using CAD was longer than those without CAD (76.6 ± 54.4 s vs. 49.4 ± 17.7 s, P = 0.000) for all readers. Significant differences were found for confidence scores in inter-group measurements with CAD ( P = 0.045) and without CAD ( P < 0.001).At the expense of longer reading time, the use of the CAD algorithms improves sensitivities, NPVs, and the accuracies of readers in peripheral PE detection, especially for readers with a poor level of interpretation experience.
View details for DOI 10.1177/0284185118808547
View details for PubMedID 30376717
Nonbinary quantification technique accounting for myocardial infarct heterogeneity: Feasibility of applying percent infarct mapping in patients.
Journal of magnetic resonance imaging : JMRI
Binary threshold-based quantification techniques ignore myocardial infarct (MI) heterogeneity, yielding substantial misquantification of MI.To assess the technical feasibility of MI quantification using percent infarct mapping (PIM), a prototype nonbinary algorithm, in patients with suspected MI.Prospective cohort POPULATION: Patients (n = 171) with suspected MI referred for cardiac MRI.Inversion recovery balanced steady-state free-precession for late gadolinium enhancement (LGE) and modified Look-Locker inversion recovery (MOLLI) T1 -mapping on a 1.5T system.Infarct volume (IV) and infarct fraction (IF) were quantified by two observers based on manual delineation, binary approaches (2-5 standard deviations [SD] and full-width at half-maximum [FWHM] thresholds) in LGE images, and by applying the PIM algorithm in T1 and LGE images (PIMT1 ; PIMLGE ).IV and IF were analyzed using repeated measures analysis of variance (ANOVA). Agreement between the approaches was determined with Bland-Altman analysis. Interobserver agreement was assessed by intraclass correlation coefficient (ICC) analysis.MI was observed in 89 (54.9%) patients, and 185 (38%) short-axis slices. IF with 2, 3, 4, 5SDs and FWHM techniques were 15.7 ± 6.6, 13.4 ± 5.6, 11.6 ± 5.0, 10.8 ± 5.2, and 10.0 ± 5.2%, respectively. The 5SD and FWHM techniques had the best agreement with manual IF (9.9 ± 4.8%) determination (bias 1.0 and 0.2%; P = 0.1426 and P = 0.8094, respectively). The 2SD and 3SD algorithms significantly overestimated manual IF (9.9 ± 4.8%; both P < 0.0001). PIMLGE measured significantly lower IF (7.8 ± 3.7%) compared to manual values (P < 0.0001). PIMLGE , however, showed the best agreement with the PIMT1 reference (7.6 ± 3.6%, P = 0.3156). Interobserver agreement was rated good to excellent for IV (ICCs between 0.727-0.820) and fair to good for IF (0.589-0.736).The application of the PIMLGE technique for MI quantification in patients is feasible. PIMLGE , with its ability to account for voxelwise MI content, provides significantly smaller IF than any thresholding technique and shows excellent agreement with the T1 -based reference.2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018.
View details for DOI 10.1002/jmri.25973
View details for PubMedID 29446527
- Uncommon Isolated Unilocular Myocardial Cyst in a Dog-Friendly Young Female Patient - Multimodality Imaging - CIRCULATION JOURNAL 2017; 81 (7): 1056-+
Results of Late Gadolinium Enhancement in Children Affected by Dilated Cardiomyopathy
FRONTIERS IN PEDIATRICS
2017; 5: 13
Little is known about the clinical value of late gadolinium enhancement (LGE), in children affected by dilated cardiomyopathy (DCM).We retrospectively evaluated 15 patients (8 ± 6 years, 6 males) with diagnosis of DCM who underwent cardiac magnetic resonance since 2014. All scans were performed with a 1.5 T system (Aera, Siemens). Study protocol included cine steady-state free precession sequences, followed by administration of 0.2 mmol/kg of gadolinium-based contrast agent. Inversion recovery Turbo Flash sequences, in the same position of cine images, were acquired 10-15 min after the injection of contrast agent, in order to assess the presence of LGE. The latter was considered positive with a signal intensity >6 SD from normal myocardial tissue. Indexed end-diastolic volume (EDVi) and end-systolic volume (ESVi), and left ventricle (LV) ejection fraction (EF) were calculated by using dedicated software on off-line workstation. Global longitudinal strain and diastolic function were evaluated by echocardiography. Clinical follow-up, including death, transplant, and listing for heart transplant [major adverse cardiac events (MACE)], were evaluated. Patients were divided into two different subgroups: negative (Group A) and positive (Group B) for presence of LGE. Statistical analysis was performed by using Mann-Whitney U test (p < 0.05 considered as statistically significant).Seven patients (47%) showed LGE. A global diffuse subendocardial pattern was evident in all patients presenting LGE (7/7, 100%). The following main LV indexes were observed in the two subgroups. Group A: EDVi = 96 ± 33 ml, ESVi = 56 ± 29 ml, LV EF = 45 ± 10%, global longitudinal strain = -16 ± 5%, E/e' ratio = 10 ± 3, MACE = 1. Group B: EDVi = 130 ± 60 ml, ESVi = 89 ± 43 ml, LV EF = 31 ± 6%, global longitudinal strain = -13 ± 4%, E/e' ratio = 9 ± 3, MACE = 3. There was no statistically significant difference between the two groups, in terms of EDVi (p: 0.2), ESVi (p: 0.2), and E/e' ratio (0.9), whereas a significant difference of LV EF, presence of significative mitral regurgitation, and global longitudinal strain were observed (respectively, p: 0.03, p: 0.009, and p: 0.03).In our population of children with DCM, LGE shows a global diffuse subendocardial pattern. Presence of LGE seems to play a role in these patients determining a worst global systolic function.
View details for DOI 10.3389/fped.2017.00013
View details for Web of Science ID 000393308600002
View details for PubMedID 28220144
View details for PubMedCentralID PMC5292614
New Imaging Techniques for Atherosclerotic Plaque Characterization
Current Radiology Reports
View details for DOI 10.1007/s40134-017-0252-6
Unexplained Cardiac Arrest After Near Drowning in a Young Experienced Swimmer: Insight from Cardiovascular Magnetic Resonance Imaging
IRANIAN JOURNAL OF RADIOLOGY
2016; 13 (4): e36779
Cardiac magnetic resonance imaging (cMRI) is a well-established noninvasive imaging modality in clinical cardiology. Its ability to provide tissue characterization make it well suited for the study of patients with cardiac diseases. We describe a multi-modality imaging evaluation of a 45-year-old man who experienced a near drowning event during swimming. We underline the unique capability of tissue characterization provided by cMRI, which allowed detection of subtle, clinically unrecognizable myocardial damage for understanding the causes of sudden cardiac arrest and also showed the small damages caused by cardiopulmonary resuscitation.
View details for DOI 10.5812/iranjradiol.36779
View details for Web of Science ID 000395456300007
View details for PubMedID 27895877
View details for PubMedCentralID PMC5116988