Moritz Christoph Wurnig

Moritz Christoph Wurnig

Auto-generated profile, based on publicly available information

Moritz Christoph Wurnig

Institute of Diagnostic and Interventional Radiology, University Hospital Zürich, Switzerland

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 July 2017 in Investigative Radiology
Verified
Journal Article
Closed access
PDF file
ABAnton S. Becker
Anton S. Becker
MMMagda Marcon
Magda Marcon
SGSoleen Ghafoor
Soleen Ghafoor
MWMoritz C. Wurnig
Moritz C. Wurnig
Thomas FrauenfelderThomas Frauenfelder
Thomas Frauenfelder
ABAndreas Boss
Andreas Boss
Objectives: The aim of this study was to evaluate the diagnostic accuracy of a multipurpose image analysis software based on deep learning with artificial neural networks for the detection of breast cancer in an independent, dual-center mammography data set. Materials and Methods: In this retrospective, Health Insurance Portability and Accountability Act-compliant study, all patients undergoing mammography in 2012 at our institution were reviewed (n = 3228). All of their prior and follow-up mammographies from a time span of 7 years (2008–2015) were considered as a reference for clinical diagnosis. After applying exclusion criteria (missing reference standard, prior procedures or therapies), patients with the first diagnosis of a malignoma or borderline lesion were selected (n = 143). Histology or clinical long-term follow-up served as reference standard. In a first step, a breast density-and age-matched control cohort was selected (n = 143) from the remaining patients with more than 2 years follow-up (n = 1003). The neural network was trained with this data set. From the publicly available Breast Cancer Digital Repository data set, patients with cancer and a matched control cohort were selected (n = 35 × 2). The performance of the trained neural network was also tested with this external data set. Three radiologists (3, 5, and 10 years of experience) evaluated the test data set. In a second step, the neural network was trained with all cases from January to September and tested with cases from October to December 2012 (screening-like cohort). The radiologists also evaluated this second test data set. The areas under the receiver operating characteristic curve between readers and the neural network were compared. A Bonferroni-corrected P value of less than 0.016 was considered statistically significant. Results: Mean age of patients with lesion was 59.6 years (range, 35–88 years) and in controls, 59.1 years (35–83 years). Breast density distribution (A/B/C/D) was 21/59/42/21 and 22/60/41/20, respectively. Histologic diagnoses were invasive ductal carcinoma in 90, ductal in situ carcinoma in 13, invasive lobular carcinoma in 13, mucinous carcinoma in 3, and borderline lesion in 12 patients. In the first step, the area under the receiver operating characteristic curve of the trained neural network was 0.81 and comparable on the test cases 0.79 (P = 0.63). One of the radiologists showed almost equal performance (0.83, P = 0.17), whereas 2 were significantly better (0.91 and 0.94, P < 0.016). In the second step, performance of the neural network (0.82) was not significantly different from the human performance (0.77–0.87, P > 0.016); however, radiologists were consistently less sensitive and more specific than the neural network. Conclusions: Current state-of-the-art artificial neural networks for general image analysis are able to detect cancer in mammographies with similar accuracy to radiologists, even in a screening-like cohort with low breast cancer prevalence.

4 August 2013 in Nature Genetics
Verified
Journal Article
Closed access
PDF file
AKAnnika Keller
Annika Keller
AWAna Westenberger
Ana Westenberger
MSMaria J Sobrido
Maria J Sobrido
MGMaria García-Murias
Maria García-Murias
ADAloysius Domingo
Aloysius Domingo
JOJoao R M Oliveira
Joao R M Oliveira

1 August 2011 in NeuroImage
Verified
Journal Article
Open access
RBRoland Beisteiner
Roland Beisteiner
SRS. Robinson
S. Robinson
MWM. Wurnig
M. Wurnig
MHM. Hilbert
M. Hilbert
KMK. Merksa
K. Merksa
AGA. Geißler
A. Geißler
Despite there being an increasing number of installations of ultra high field MR systems (>3T) in clinical environments, no functional patient investigations have yet examined possible benefits for functional diagnostics. Here we performed presurgical localization of the primary motor hand area on 3T and 7T Siemens scanners with identical investigational procedures and comparable system specific sequence optimizations. Results from 17 patients showed significantly higher functional sensitivity of the 7T system measured via percent signal change, mean t-values, number of suprathreshold voxels and contrast to noise ratio. On the other hand, 7T data suffered from a significant increase of artifacts (ghosting, head motion). We conclude that ultra high field systems provide a clinically relevant increase of functional sensitivity for patient investigations.

31 October 2014 in Magnetic Resonance in Medicine
Verified
Journal Article
Open access
PDF file
MWMoritz C. Wurnig
Moritz C. Wurnig
ODOlivio F. Donati
Olivio F. Donati
EUErika Ulbrich
Erika Ulbrich
LFLukas Filli
Lukas Filli
DKDavid Kenkel
David Kenkel
ABAndreas Boss
Andreas Boss
Purpose: To systematically evaluate the dependence of intravoxel‐incoherent‐motion (IVIM) parameters on the b‐value threshold separating the perfusion and diffusion compartment, and to implement and test an algorithm for the standardized computation of this threshold.Methods: Diffusion weighted images of the upper abdomen were acquired at 3 Tesla in eleven healthy male volunteers with 10 different b‐values and in two healthy male volunteers with 16 different b‐values. Region‐of‐interest IVIM analysis was applied to the abdominal organs and skeletal muscle with a systematic increase of the b‐value threshold for computing pseudodiffusion D*, perfusion fraction Fp, diffusion coefficient D, and the sum of squared residuals to the bi‐exponential IVIM‐fit.Results: IVIM parameters strongly depended on the choice of the b‐value threshold. The proposed algorithm successfully provided optimal b‐value thresholds with the smallest residuals for all evaluated organs [s/mm2]: e.g., right liver lobe 20, spleen 20, right renal cortex 150, skeletal muscle 150. Mean D* [10−3 mm2/s], Fp [%], and D [10−3 mm2/s] values (±standard deviation) were: right liver lobe, 88.7 ± 42.5, 22.6 ± 7.4, 0.73 ± 0.12; right renal cortex: 11.5 ± 1.8, 18.3 ± 2.9, 1.68 ± 0.05; spleen: 41.9 ± 57.9, 8.2 ± 3.4, 0.69 ± 0.07; skeletal muscle: 21.7 ± 19.0; 7.4 ± 3.0; 1.36 ± 0.04.Conclusion: IVIM parameters strongly depend upon the choice of the b‐value threshold used for computation. The proposed algorithm may be used as a robust approach for IVIM analysis without organ‐specific adaptation. Magn Reson Med 74:1414–1422, 2015.

1 November 2015 in Clinical Radiology
Verified
Journal Article
Closed access
AMA. Meier
A. Meier
MWM. Wurnig
M. Wurnig
LDL. Desbiolles
L. Desbiolles
SLS. Leschka
S. Leschka
Thomas FrauenfelderThomas Frauenfelder
Thomas Frauenfelder
HAH. Alkadhi
H. Alkadhi

19 May 2013 in European Radiology
Verified
Journal Article
Closed access
PDF file
FMFabian Morsbach
Fabian Morsbach
MWMoritz Wurnig
Moritz Wurnig
DKDaniel M. Kunz
Daniel M. Kunz
AKAndreas Krauss
Andreas Krauss
BSBernhard Schmidt
Bernhard Schmidt
Hatem AlkadhiHatem Alkadhi
Hatem Alkadhi