@article{UsmanReimannLiedletal.2018,
  author    = {Usman, Muhammad and Reimann, Thomas and Liedl, Rudolf and Abbas, Azhar and Conrad, Christopher and Saleem, Shoaib},
  title     = {Inverse parametrization of a regional groundwater flow model with the aid of modelling and GIS: test and application of different approaches},
  series = {ISPRS International Journal of Geo-Information},
  volume    = {7},
  journal   = {ISPRS International Journal of Geo-Information},
  number    = {1},
  doi       = {10.3390/ijgi7010022},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-175721},
  pages     = {22},
  year      = {2018},
  abstract  = {The use of inverse methods allow efficient model calibration. This study employs PEST to calibrate a large catchment scale transient flow model. Results are demonstrated by comparing manually calibrated approaches with the automated approach. An advanced Tikhonov regularization algorithm was employed for carrying out the automated pilot point (PP) method. The results indicate that automated PP is more flexible and robust as compared to other approaches. Different statistical indicators show that this method yields reliable calibration as values of coefficient of determination (R-2) range from 0.98 to 0.99, Nash Sutcliffe efficiency (ME) range from 0.964 to 0.976, and root mean square errors (RMSE) range from 1.68 m to 1.23 m, for manual and automated approaches, respectively. Validation results of automated PP show ME as 0.969 and RMSE as 1.31 m. The results of output sensitivity suggest that hydraulic conductivity is a more influential parameter. Considering the limitations of the current study, it is recommended to perform global sensitivity and linear uncertainty analysis for the better estimation of the modelling results.},
  language  = {en}
}
@article{AnkenbrandShainbergHocketal.2021,
  author    = {Ankenbrand, Markus J. and Shainberg, Liliia and Hock, Michael and Lohr, David and Schreiber, Laura M.},
  title     = {Sensitivity analysis for interpretation of machine learning based segmentation models in cardiac MRI},
  series = {BMC Medical Imaging},
  volume    = {21},
  journal   = {BMC Medical Imaging},
  number    = {1},
  doi       = {10.1186/s12880-021-00551-1},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-259169},
  pages     = {27},
  year      = {2021},
  abstract  = {Background Image segmentation is a common task in medical imaging e.g., for volumetry analysis in cardiac MRI. Artificial neural networks are used to automate this task with performance similar to manual operators. However, this performance is only achieved in the narrow tasks networks are trained on. Performance drops dramatically when data characteristics differ from the training set properties. Moreover, neural networks are commonly considered black boxes, because it is hard to understand how they make decisions and why they fail. Therefore, it is also hard to predict whether they will generalize and work well with new data. Here we present a generic method for segmentation model interpretation. Sensitivity analysis is an approach where model input is modified in a controlled manner and the effect of these modifications on the model output is evaluated. This method yields insights into the sensitivity of the model to these alterations and therefore to the importance of certain features on segmentation performance. Results We present an open-source Python library (misas), that facilitates the use of sensitivity analysis with arbitrary data and models. We show that this method is a suitable approach to answer practical questions regarding use and functionality of segmentation models. We demonstrate this in two case studies on cardiac magnetic resonance imaging. The first case study explores the suitability of a published network for use on a public dataset the network has not been trained on. The second case study demonstrates how sensitivity analysis can be used to evaluate the robustness of a newly trained model. Conclusions Sensitivity analysis is a useful tool for deep learning developers as well as users such as clinicians. It extends their toolbox, enabling and improving interpretability of segmentation models. Enhancing our understanding of neural networks through sensitivity analysis also assists in decision making. Although demonstrated only on cardiac magnetic resonance images this approach and software are much more broadly applicable.},
  language  = {en}
}