Titus Kuehne

Titus Kuehne

Auto-generated profile, based on publicly available information

Titus Kuehne

Deutsches Herzzentrum der Charité, Institute of Computer-assisted Cardiovascular Medicine, Berlin, Germany

Research Focus & Metrics

Academic Career Length

23Years (2004-Present)

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

Representative Publications

View All Publications
26 October 2020 in npj Digital Medicine
Verified
Journal Article
Open access
PDF file
NRNina Rank
Nina Rank
BPBoris Pfahringer
Boris Pfahringer
JKJörg Kempfert
Jörg Kempfert
CSChristof Stamm
Christof Stamm
TKTitus Kühne
Titus Kühne
AMAlexander Meyer
Alexander Meyer
Acute kidney injury (AKI) is a major complication after cardiothoracic surgery. Early prediction of AKI could prompt preventive measures, but is challenging in the clinical routine. One important reason is that the amount of postoperative data is too massive and too high-dimensional to be effectively processed by the human operator. We therefore sought to develop a deep-learning-based algorithm that is able to predict postoperative AKI prior to the onset of symptoms and complications. Based on 96 routinely collected parameters we built a recurrent neural network (RNN) for real-time prediction of AKI after cardiothoracic surgery. From the data of 15,564 admissions we constructed a balanced training set (2224 admissions) for the development of the RNN. The model was then evaluated on an independent test set (350 admissions) and yielded an area under curve (AUC) (95% confidence interval) of 0.893 (0.862–0.924). We compared the performance of our model against that of experienced clinicians. The RNN significantly outperformed clinicians (AUC = 0.901 vs. 0.745, p < 0.001) and was overall well calibrated. This was not the case for the physicians, who systematically underestimated the risk (p < 0.001). In conclusion, the RNN was superior to physicians in the prediction of AKI after cardiothoracic surgery. It could potentially be integrated into hospitals’ electronic health records for real-time patient monitoring and may help to detect early AKI and hence modify the treatment in perioperative care.

6 April 2016 in Heart
Verified
Journal Article
Open access
PDF file
GHGeorg Hansmann
Georg Hansmann
Christian ApitzChristian Apitz
Christian Apitz
HAHashim Abdul-khaliq
Hashim Abdul-khaliq
TATero-Pekka Alastalo
Tero-Pekka Alastalo
PBPhillip Beerbaum
Phillip Beerbaum
PZPeter Zartner
Peter Zartner
The European Paediatric Pulmonary Vascular Disease (PVD) Network is a registered, non-profit organisation that strives to define and develop effective, innovative diagnostic methods and treatment options in all forms of paediatric pulmonary hypertensive vascular disease, including specific forms such as pulmonary arterial hypertension (PAH)-congenital heart disease, pulmonary hypertension (PH) associated with bronchopulmonary dysplasia, persistent PH of the newborn, and related cardiac dysfunction. Methods The writing group members conducted searches of the PubMed/MEDLINE bibliographic database (1990–2015) and held five face-to-face meetings with votings. Clinical trials, guidelines, and reviews limited to paediatric data were searched using the terms ‘pulmonary hypertensioń’ and 5–10 other keywords, as outlined in the other nine articles of this special issue. Class of recommendation (COR) and level of evidence (LOE) were assigned based on European Society of Cardiology/American Heart Association definitions and on paediatric data only, or on adult studies that included >10% children. Results A total of 9 original consensus articles with graded recommendations (COR/LOE) were developed, and are summarised here. The topics included diagnosis/monitoring, genetics/biomarker, cardiac catheterisation, echocardiography, cardiac magnetic resonance/chest CT, associated forms of PH, intensive care unit/ventricular assist device/lung transplantation, and treatment of paediatric PAH. Conclusions The multipaper expert consensus statement of the European Paediatric PVD Network provides a specific, comprehensive, detailed but practical framework for the optimal clinical care of children with PH.

5 February 2021 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
BTB. Thamsen
B. Thamsen
PYP. Yevtushenko
P. Yevtushenko
LGL. Gundelwein
L. Gundelwein
ASA. A. A. Setio
A. A. A. Setio
HLH. Lamecker
H. Lamecker
LGLeonid Goubergrits
Leonid Goubergrits
Modeling of hemodynamics and artificial intelligence have great potential to support clinical diagnosis and decision making. While hemodynamics modeling is extremely time- and resource-consuming, machine learning (ML) typically requires large training data that are often unavailable. The aim of this study was to develop and evaluate a novel methodology generating a large database of synthetic cases with characteristics similar to clinical cohorts of patients with coarctation of the aorta (CoA), a congenital heart disease associated with abnormal hemodynamics. Synthetic data allows use of ML approaches to investigate aortic morphometric pathology and its influence on hemodynamics. Magnetic resonance imaging data (154 patients as well as of healthy subjects) of aortic shape and flow were used to statistically characterize the clinical cohort. The methodology generating the synthetic cohort combined statistical shape modeling of aortic morphometry and aorta inlet flow fields and numerical flow simulations. Hierarchical clustering and non-linear regression analysis were successfully used to investigate the relationship between morphometry and hemodynamics and to demonstrate credibility of the synthetic cohort by comparison with a clinical cohort. A database of 2652 synthetic cases with realistic shape and hemodynamic properties was generated. Three shape clusters and respective differences in hemodynamics were identified. The novel model predicts the CoA pressure gradient with a root mean square error of 4.6 mmHg. In conclusion, synthetic data for anatomy and hemodynamics is a suitable means to address the lack of large datasets and provide a powerful basis for ML to gain new insights into cardiovascular diseases.

1 January 2022 in Journal of Investigative Dermatology
Verified
Journal Article
Closed access
AMAndrás Maifeld
András Maifeld
Johannes WildJohannes Wild
Johannes Wild
TKTine V. Karlsen
Tine V. Karlsen
NRNatalia Rakova
Natalia Rakova
EWElisa Wistorf
Elisa Wistorf
LMLajos Marko
Lajos Marko

20 January 2020 in Journal of Cardiovascular Magnetic Resonance
Verified
Journal Article
Open access
PDF file
RKRiccardo Kranzusch
Riccardo Kranzusch
FSFabian Aus dem Siepen
Fabian Aus dem Siepen
SWStephanie Wiesemann
Stephanie Wiesemann
LZLeonora Zange
Leonora Zange
SJSarah Jeuthe
Sarah Jeuthe
DMDaniel R. Messroghli
Daniel R. Messroghli
T1 mapping using modified Look-Locker inversion recovery (MOLLI) provides quantitative information on myocardial tissue composition. T1 results differ between sites due to variations in hardware and software equipment, limiting the comparability of results. The aim was to test if Z-scores can be used to compare the results of MOLLI T1 mapping from different cardiovascular magnetic resonance (CMR) platforms. First, healthy subjects (n = 15) underwent 11 combinations of native short-axis T1 mapping (four CMR systems from two manufacturers at 1.5 T and 3 T, three MOLLI schemes). Mean and standard deviation (SD) of septal myocardial T1 were derived for each combination. T1 maps were transformed into Z-score maps based on mean and SD values using a prototype post-processing module. Second, Z-score mapping was applied to a validation sample of patients with cardiac amyloidosis at 1.5 T (n = 25) or 3 T (n = 13). In conventional T1 analysis, results were confounded by variations in field strength, MOLLI scheme, and manufacturer-specific system characteristics. Z-score-based analysis yielded consistent results without significant differences between any two of the combinations in part 1 of the study. In the validation sample, Z-score mapping differentiated between patients with cardiac amyloidosis and healthy subjects with the same diagnostic accuracy as standard T1 analysis regardless of field strength. T1 analysis based on Z-score mapping provides consistent results without significant differences due to field strengths, CMR systems, or MOLLI variants, and detects cardiac amyloidosis with the same diagnostic accuracy as conventional T1 analysis. Z-score mapping provides a means to compare native T1 results acquired with MOLLI across different CMR platforms.

9 December 2022 in Journal of Translational Medicine
Verified
Journal Article
Open access
PDF file
EPElisabeth Petter
Elisabeth Petter
CSCarmen Scheibenbogen
Carmen Scheibenbogen
PLPeter Linz
Peter Linz
CSChristian Stehning
Christian Stehning
KWKlaus Wirth
Klaus Wirth
MKMarcus Kelm
Marcus Kelm
Background: Muscle fatigue and pain are key symptoms of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS). Although the pathophysiology is not yet fully understood, there is ample evidence for hypoperfusion which may result in electrolyte imbalance and sodium overload in muscles. Therefore, the aim of this study was to assess levels of sodium content in muscles of patients with ME/CFS and to compare these to healthy controls. Methods: Six female patients with ME/CFS and six age, BMI and sex matched controls underwent 23Na-MRI of the left lower leg using a clinical 3T MR scanner before and after 3 min of plantar flexion exercise. Sodium reference phantoms with solutions of 10, 20, 30 and 40 mmol/L NaCl were used for quantification. Muscle sodium content over 40 min was measured using a dedicated plugin in the open-source DICOM viewer Horos. Handgrip strength was measured and correlated with sodium content. Results: Baseline tissue sodium content was higher in all 5 lower leg muscle compartments in ME/CFS compared to controls. Within the anterior extensor muscle compartment, the highest difference in baseline muscle sodium content between ME/CFS and controls was found (mean ± SD; 12.20 ± 1.66 mM in ME/CFS versus 9.38 ± 0.71 mM in controls, p = 0.0034). Directly after exercise, tissue sodium content increased in gastrocnemius and triceps surae muscles with + 30% in ME/CFS (p = 0.0005) and + 24% in controls (p = 0.0007) in the medial gastrocnemius muscle but not in the extensor muscles which were not exercised. Compared to baseline, the increase of sodium content in medial gastrocnemius muscle was stronger in ME/CFS than in controls with + 30% versus + 17% to baseline at 12 min (p = 0.0326) and + 29% versus + 16% to baseline at 15 min (p = 0.0265). Patients had reduced average handgrip strength which was associated with increased average muscle tissue sodium content (p = 0.0319, R2 = 0.3832). Conclusion: Muscle sodium content before and after exercise was higher in ME/CFS than in healthy controls. Furthermore, our findings indicate an inverse correlation between muscle sodium content and handgrip strength. These findings provide evidence that sodium overload may play a role in the pathophysiology of ME/CFS and may allow for potential therapeutic targeting.