FWFelix Werner Wehrli

Felix Werner Wehrli

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Felix Werner Wehrli has a total of 160 co-authors. Here are the top co-authors by number of shared publications:

  • Hyunyeol Lee

    Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA; School of Electronic and Electrical Engineering, Kyungpook National University, Republic of Korea

    12 shared publications
  • Michael C. Langham

    Department of Radiology, Perelman School of Medicine University of Pennsylvania Philadelphia Pennsylvania USA

    10 shared publications
  • Hyunyeol Lee

    School of Electronic and Electrical Engineering, Kyungpook National University, Daegu, Korea

    9 shared publications
  • Hee Kwon Song

    Department of Radiology, Perelman School of Medicine University of Pennsylvania Philadelphia Pennsylvania USA

    8 shared publications
  • Michael C Langham

    Laboratory for Structural, Physiologic and Functional Imaging (LSPFI), Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA

    6 shared publications
  • Scott P. Bartlett

    Division of Plastic, Reconstructive and Oral Surgery Children's Hospital of Philadelphia Philadelphia Pennsylvania USA

    6 shared publications

Latest Publications

  • Open Access

    Self‐Correction of B 0 Drift and Trajectory Error in Alternating Unbalanced SSFP ‐ FID and SSFP ‐ ECHO ( AUSFIDE )‐Based 3D Transverse Relaxometry of the Human Brain

    by Eunseo Bae, Sung Suk Oh, Dongyeob Han, Kathryn M. Jaroszynski, Felix W. Wehrli, Hyunyeol Lee

    Purpose: The alternating unbalanced SSFP (AUSFIDE) pulse sequence, which consists of multi‐echo trains of SSFP‐FID and SSFP‐ECHO modules, enables rapid estimation of R 2 and R 2 ′, but is prone to errors from temporal drift of the main magnetic field ( B 0 ) and k ‐space trajectory mismatch. Here, we propose a self‐correcting AUSFIDE method for joint correction of these two sources of error without requiring additional scans. Methods: AUSFIDE signals were modeled in the presence of B 0 drift and k ‐space shifts. To estimate and correct for B 0 drift over the entire scan, phase navigators without spatial encoding were inserted before and after each multi‐echo train. In addition, the intrinsic phase symmetry between SSFP‐FID and SSFP‐ECHO signals was exploited, enabling self‐calibration of the k ‐space sampling trajectory. Numerical simulations and phantom experiments were performed for proof‐of‐concept validation, followed by in vivo brain imaging of 10 healthy subjects on two 3 T scanners. Measured parameters were analyzed in six brain regions, and test–retest reproducibility was evaluated on the two MRI platforms. Results: Simulations and phantom imaging consistently demonstrated that B 0 drift and trajectory errors resulted in image distortions and heavily overestimated R 2 ′. Experiments further revealed that these artifacts and quantification errors were substantially reduced by the proposed self‐correction method. Inter‐scan agreement of the quantified parameters was also improved after correction on both scanners. Conclusion: The self‐corrected AUSFIDE method is able to compensate for errors from B 0 drift and k ‐space trajectory mismatch, and thus is a potentially practical and reliable means for 3D transverse relaxometry.

  • Open Access

    Acute Effects of Smoking and Vaping on Markers of Vascular Function via Quantitative MRI

    by Marianne Nabbout, Mahdie Hosseini, Harris Quach, Michael C. Langham, Alessandra S. Caporale, Wensheng Guo, Kathryn Jaroszynski, Frank T. Leone, Andrew A. Strasser, Felix W. Wehrli

    Background: E‐cigarettes contribute to endothelial dysfunction, a known cause of systemic atherosclerotic disease that cannot be adequately characterized by a single physiological parameter measured at a single vascular bed. Purpose/Hypothesis: Quantify and compare acute effects of smoking or vaping with tobacco cigarettes (t‐cigs), nicotinized e‐cigarettes (e‐cigs + ), non‐nicotinized e‐cigarettes (e‐cigs − ) on the vascular system via quantitative MRI. Study Type: Prospective. Subjects: Forty one smokers/vapers (mean age 29.8 ± 7.6 years), 17 non‐smokers/non‐vapers (mean age 26.5 ± 3.9 years). Field Strength/Sequence: 3 T/oximetry, blood flow velocimetry. Assessment: Smokers/vapers were scanned pre‐ and post‐ three challenges (ecig + , ecig, tcig). Each challenge was on a separate day. Blood flow velocity waveform in the femoral artery was measured at rest and in response to cuff occlusion, yielding parameters characterizing hyperemia (arterial peak velocity, time to peak, time of forward flow (T FF , duration of monophasic forward flow), hyperemic index, peripheral flow reserve, (PFR, peak over baseline velocity)). The dynamics of venous oxygen saturation (SvO 2 ) in the superficial femoral vein were quantified in terms of washout time of oxygen‐depleted blood, upslope (rate of resaturation), and overshoot (peak SvO 2 ). Aortic pulse wave velocity was evaluated. Cerebrovascular reactivity and cerebral metabolic rate of oxygen consumption were assessed at the superior sagittal sinus. Neurovascular compliance was assessed at the common carotid arteries. Statistical Tests: Paired t ‐tests, Wilcoxon rank sum tests, linear mixed‐effects models, Bonferonni corrections. p < 0.05 indicated statistical significance. Results: At the femoral vein and artery, across users, post‐challenge, the following decreased significantly: baseline SvO 2 (e‐cig + : −12%; e‐cig − : −14%; t‐cig: −9%); baseline velocity (e‐cig + : −23%; e‐cig − : −22%; t‐cig: −18%); T FF (e‐cig + : −17%; e‐cig − : −12%). The following significantly increased: upslope (e‐cig + : +22%; e‐cig − : +23%; t‐cig: +19%); overshoot (e‐cig + : +19%; e‐cig − : +23%); PFR (e‐cig − : +32%; t‐cig: +31%). Accounting for user type, the mixed model showed T FF to decrease significantly post‐challenge for all interventions: −15, −17, −23% for smokers, vapers, and dual users. Data Conclusion: e‐cig + , e‐cig − , t‐cig exposure resulted in strong, significant acute effects on vascular function markers. Evidence Level: 1. Stage of Technical Efficacy: Stage 1.

  • Open Access

    Age dependence of brain oxygen metabolism in adults assessed by 3D constrained quantitative BOLD MRI

    by Kathryn M. Jaroszynski, Sina Dindarian, Hyunyeol Lee, John A. Detre, Felix W. Wehrli


  • A reference-based PET/MRI method for quantifying activation-induced changes in cerebral oxygen metabolism

    by Graham Deller, Linshan Liu, Justin W Hicks, Lucas Narciso, Felix W Wehrli, Kathryn Jaroszynski, Michael C Langham, Keith St Lawrence

    Hybrid PET/MRI can overcome the complexity of PET imaging of the cerebral metabolic rate of oxygen (CMRO 2 ), while retaining the ability to directly measure oxygen uptake in the brain. One technique, PMROx, incorporates complementary MRI methods acquired simultaneously with [ 15 O]O 2 -PET. Specifically, the MRI-based method arterial spin labelling (ASL) to image cerebral blood flow (CBF) and MR-susceptometry to measure whole-brain CMRO 2 . PMROx is non-invasive with imaging times around 5 min, making it feasible to image CMRO 2 under different conditions in one session. This study presents the first application of PMROx to humans with the aims of evaluating its reliability and sensitivity to increased CMRO 2 during functional activation (right-handed sequential finger tapping). In addition, blood-oxygen level dependent (BOLD) images were acquired to compare BOLD contrast to underlying changes in CBF and CMRO 2 . Across 14 participants, mean CMRO 2 was 3.2 ± 0.5 mLO 2 /100 g/min with excellent within-session repeatability. Significant increases in CBF, CMRO 2 and BOLD contrast were detected in the primary sensorimotor cortex, supplementary motor area and secondary somatosensory cortex. This study demonstrated the ability of PMROx to image CMRO 2 non-invasively, its sensitivity to increased regional CMRO 2 and how PET/MRI provides the opportunity to compare BOLD contrast to underlying changes in perfusion and oxygen metabolism.

  • Open Access

    Neurometabolic Response to Apneic Stimuli Tracks Global Grey Matter Volume Deficits in Patients With Obstructive Sleep Apnea

    by Jeffrey B. Dennison, Pei‐Hsin Wu, Michael C. Langham, Richard J. Schwab, John A. Detre, Felix W. Wehrli

    Obstructive sleep apnea (OSA) has been linked to cortical atrophy and increased risk of neurodegeneration, a process potentially mediated by neurometabolic dysregulation. Recent work using high temporal‐resolution metabolic MRI found greater changes in the cerebral metabolic rate of oxygen consumption (CMRO 2 ) during cued breath‐hold stimuli simulating spontaneous apneas in OSA patients than in healthy people. Here, we conducted a secondary analysis on this work to assess whether the observed neurometabolic responses were associated with grey matter volume fraction. We examined previously collected data from OSA patients matched to non‐OSA participants by age and sex to evaluate associations between MRI‐based measures of neurometabolism and grey matter volume. Grey matter volume fraction (GMVF) was computed by segmenting T1‐weighted structural images and regressing out age effects in OSA patients relative to the age‐matched non‐OSA participants. We then examined the relationship between GMVF, apnea‐hypopnea index (AHI), and neurometabolic responses of cerebral blood flow (CBF), venous oxygen saturation (SvO 2 ), and CMRO 2 in response to breath‐hold stimuli. OSA severity, expressed in terms of AHI, was negatively associated with GMVF in both OSA and non‐OSA reference subjects after controlling for age. Incorporating the quantitative measures of brain oxygen metabolism showed breath‐hold responses in CBF and SvO 2 to be positively associated with GMVF. However, changes in CMRO 2 were not significantly associated with age‐normalized GMVF. The findings suggest that noninvasive global measures of brain oxygen metabolism may serve as quantitative markers of neural health. Further research is warranted to explore these measures and neurodegenerative risk in OSA.


  • Open Access

    Renal Metabolic Rate of Oxygen in Response to Hypoxia Challenges by Means of Quantitative MRI in Humans

    by Nada Kamona, Mahdie Hosseini, Michael C. Langham, Felix W. Wehrli

    In early kidney disease, tissue hypoxia occurs due to an imbalance between ATP supply and demand. Whole‐organ renal metabolic rate of oxygen (rMRO 2 ) is therefore a potential biomarker for assessing renal function. This study evaluated the sensitivity of a quantitative MRI method to detect within‐subject changes in metabolic parameters during hypoxic gas challenges. Ten healthy adults were imaged at 3 T (5 female, ages 23–53 years) while undergoing mild and moderate hypoxia (P ET O 2 62 and 52 mmHg, respectively). The utilized MRI sequence simultaneously quantified blood flow rate (BFR) and venous oxygen saturation (SvO 2 ) at the left renal vein, yielding, together with arterial oxygen saturation (SaO 2 ) obtained by pulse oximetry, whole‐organ rMRO 2 by invoking Fick's Principle. Repeated‐measures ANOVA was used to test differences in metabolic parameters between baseline and hypoxic conditions. SaO 2 at baseline was 99% ± 1%, while renal SvO 2 was 92% ± 3%. During progressive hypoxemia, the drop in SvO 2 (mild 83% ± 4%, moderate 76% ± 5%, p < 0.01) paralleled the drop in SaO 2 (mild 90% ± 1%, moderate 84% ± 2%), such that the arteriovenous difference in oxygenation (AVDO 2 ) was constant when compared to baseline ( p = 1). Renal BFR did not vary significantly between baseline (410 ± 65 mL/min) and hypoxemic conditions (mild, moderate of 430 ± 56 and 440 ± 48 mL/min, p > 0.34). Thus, rMRO 2 did not significantly change during hypoxemia (baseline, mild, and moderate of 140 ± 50, 180 ± 80, and 170 ± 90 (μmol O 2 /min)/100 g, respectively, p = 1). In conclusion, the results demonstrate the method's sensitivity in detecting within‐subject changes in metabolic parameters in response to graded hypoxia. Quantitative MRI oximetry may be a feasible tool to assess and longitudinally monitor early metabolic changes in kidney disease.


  • Open Access

    DREAMER : Rapid and Simultaneous Multiple Contrast Magnetic Resonance Imaging of Solid and Soft Tissue

    by Brian‐Tinh Duc Vu, Nada Kamona, Felix W. Wehrli, Rajiv S. Deshpande, Hee Kwon Song, Allison Hu, Emily Baccaglini, Scott P. Bartlett, Chamith S. Rajapakse

    Purpose: Pediatric craniofacial imaging may involve examination of both the skull and brain tissues via CT and MRI, respectively. DREAMER (Dual Repetition and Echo Acquisition with Multi‐contrast Encoding and Reconstruction) simultaneously acquires solid‐ and soft‐tissue images, potentially providing a rapid, high‐resolution, and radiation‐free protocol whenever bone‐selective, T 1 w, and T 2 w images are required. Methods: The DREAMER sequence combines a solid‐state MRI method with phase‐based T 2 encoding to produce a multi‐contrast signal model that enables retrospective customization of image contrast weighting. DREAMER is paired with an iterative image reconstruction algorithm for accelerated and high‐resolution structural imaging of solid‐ and soft‐tissue compartments. Two healthy adult volunteers and two pediatric patients were scanned at 3 T to qualitatively compare soft‐tissue DREAMER image contrasts with their corresponding clinical standards, T 1 w MPRAGE and T 2 w fast spin‐echo (FSE). Two patients also underwent clinical CT to compare the bone‐selective images and skull renderings. Results: DREAMER images are self‐registered, high‐resolution, and spatially isotropic. The bone‐selective, T 1 w, and T 2 w images approximate the image contrasts and structural imaging capabilities of their corresponding clinical standards (CT, T 1 w MPRAGE, and T 2 w FSE). Unlike the standard techniques, DREAMER imaging occurs at a single scanner using a single pulse sequence. Conclusion: DREAMER combines mechanisms for solid‐ and multiple‐contrast soft‐tissue imaging into a single scan. For craniofacial imaging, DREAMER may consolidate CT and MRI demand, reduce exposure to ionizing radiation, decrease patient examination and wait times, and simplify the radiological workflow.


  • Open Access

    Comparison of Brain Oxygen Metabolic Parameters Between Constrained qBOLD and Whole‐Brain Oximetric Methods at Baseline and in Response to a Physiologic Stimulus

    by Kathryn M. Jaroszynski, Hyunyeol Lee, Michael C. Langham, Felix W. Wehrli

    The measurement of cerebral oxygen metabolism is important to understand and treat many disorders. Constrained quantitative BOLD (qBOLD) MRI is a calibration‐free method for 3D voxel‐wise whole‐brain mapping of brain oxygen metabolism. This study aimed to evaluate the agreement between constrained qBOLD and global oximetry methods both at baseline and in response to a caffeine stimulus. Healthy volunteers (N = 10, age 30 ± 8 years) were imaged with constrained qBOLD, MOTIVE (metabolism of oxygen via T2 and interleaved velocity encoding), dual‐slice (DS), and single‐slice (SS) OxFlow. Subjects were then given a 200 mg caffeine pill and imaged at 2‐s temporal resolution immediately thereafter for 30 min by SS‐OxFlow. After 30 min, the baseline protocol was repeated. Constrained qBOLD uses prior constraints to the QSM + qBOLD model to solve for voxel‐wise oxygen extraction fraction (OEF). Quantification of cerebral blood flow (CBF) was accomplished for qBOLD from a separate measurement via pseudo‐continuous arterial spin labeling (pCASL) to yield CMRO2. Constrained qBOLD measured OEF (31 ± 5% gray matter [GM], 31 ± 6% white matter [WM] at baseline; 36 ± 7 GM, 35 ± 8 WM post‐caffeine) was in good agreement with global oximetry methods DS‐OxFlow (30 ± 4, 37 ± 5), SS‐OxFlow (31 ± 4, 37 ± 4), and MOTIVE (32 ± 5, 39 ± 5). Temporal data showed a gradual increase in OEF with a commensurate reduction in CBF while the caffeine was taking effect. No significant change in CMRO2 was noted with any of the techniques. Regional analysis of the basal ganglia, hippocampus, and thalamus found there was a significant increase in OEF post caffeine. The results indicate constrained qBOLD to yield OEF with negligible bias to global oximetry methods, both at baseline and post caffeine. The results also suggest that constrained qBOLD has the sensitivity to detect changes in oxygen metabolism due to a vasoconstrictive stimulus.


  • Open Access

    Concurrent evaluation of cerebral oxygen metabolism and upper airway architecture via temporally resolved MRI

    by Jeffrey B Dennison, Michael C Langham, Andrew S Wiemken, Jing Xu, Richard J Schwab, John A Detre, Felix W Wehrli

    Obstructive sleep apnea (OSA) disrupts the oxygen supply during apneic and hypopneic events. To evaluate the feasibility of concurrently monitoring cerebral metabolic rate of oxygen (CMRO 2 ) and airway anatomy, a magnetic resonance imaging (MRI) pulse sequence was developed that interleaves measurements of CMRO 2 with anatomic imaging of the upper airway at a temporal resolution of 5 seconds. The sequence was first tested in healthy subjects during wakefulness to detect the effect of volitional breath-hold and swallowing apneas on neuro-metabolic parameters and airway morphology. Subsequently, select patients with diagnosed OSA and healthy reference subjects were scanned during 90 minutes of wakefulness and sleep with concurrent electroencephalographic (EEG) monitoring and airway plethysmography. During non-rapid eye movement sleep, changes in metabolic parameters caused by neurovascular-metabolic uncoupling were detected, resulting in sleep-stage dependent reductions in the CMRO 2 . Spontaneous apneas were visible in airway images and confirmed plethysmographically. Recurrent apneas in patients during N1 and N2 sleep led to increased SvO 2 and CBF (hypercapnic-hypoxic response) and decreases in SaO 2 (hypoxemic response from airway closure) resulting in CMRO 2 reductions as large 60%. The results demonstrate the MRI potential of noninvasive assessment of the dynamic changes in airway anatomy and brain metabolism in OSA during sleep.


  • Open Access

    MRI ‐based quantification of whole‐organ renal metabolic rate of oxygen during free‐breathing

    by Nada Kamona, Michael C. Langham, Rajiv S. Deshpande, Brian‐Tinh D. Vu, Mahdie Hosseini, Jeffrey Dennison, Kathryn M. Jaroszynski, Felix W. Wehrli

    Purpose: Renal metabolic rate of oxygen (rMRO 2 ) reflects the kidney's metabolic efficiency, making it a potential biomarker for early‐stage kidney disease. This study introduces an ungated, free‐breathing MRI sequence in comparison to its breath‐hold counterpart to noninvasively measure whole‐organ rMRO 2 . Methods: Free‐breathing (FB) K‐MOTIVE sequence (kidney metabolism of oxygen via T 2 and interleaved velocity encoding) was developed to simultaneously measure renal blood flow rate (BFR) and T 2 of blood water using the conservation of mass. T 2 is converted to venous oxygen saturation (SvO 2 ) using a calibration curve. Compared to previous versions, FB K‐MOTIVE minimizes respiratory motion artifacts by acquiring fully sampled velocity maps with spiral readout instead of partially collecting radial views at each T 2 weighting. Healthy participants ( n = 15, 32 ± 9 years) were imaged at 3 T at the renal veins to quantify individual rMRO 2 , and at the suprarenal and infrarenal inferior vena cava to indirectly quantify bilateral rMRO 2 (the total metabolism from both kidneys). Results: Renal venous blood was highly oxygenated (SvO 2 91% ± 3%) and exhibited high BFR of 460 ± 90 mL/min per kidney. Further, total rMRO 2 of the two kidneys (160 ± 80 (μmol O 2 /min)/100 g) was statistically comparable to the indirect bilateral rMRO 2 (250 ± 120 (μmol O 2 /min)/100 g, p = 0.066). Using Lin's concordance correlation coefficient, there was good agreement between breath‐hold and free‐breathing acquisitions at the individual kidneys for SvO 2 (>0.75), BFR (>0.96), and rMRO 2 (>0.75). Conclusion: FB K‐MOTIVE is a feasible approach to estimate rMRO 2 , yielding reproducible and physiologically plausible metabolic parameters. Free‐breathing acquisition can enhance patient comfort by eliminating the need for breath‐holding.