Refine
Document Type
Is part of the Bibliography
- no (19)
Keywords
- Deep Learning (2)
- AI (1)
- Algorithms (1)
- BIM (1)
- BLE Beacons (1)
- Benchmark testing (1)
- Charging (1)
- Cholecystectomy (1)
- Generalized zero-shot learning (1)
- Indoor Localisation (1)
Institute
- Fakultät Informatik und Mathematik (19)
- Labor Regensburg Medical Image Computing (ReMIC) (16)
- Research Center of Biomedical Engineering - RCBE (16)
- Research Center of Health Sciences and Technology - RCHST (16)
- Research Center of Energy and Resources - RCER (3)
- Research Center for Artificial Intelligence - RCAI (1)
Begutachtungsstatus
- peer-reviewed (12)
In the engineering domain, representing real-world objects using a body of data, called a digital twin, which is frequently updated by “live” measurements, has shown various advantages over tradi- tional modelling and simulation techniques. Consequently, urban planners have a strong interest in digital twin technology, since it provides them with a laboratory for experimenting with data before making far-reaching decisions. Realizing these decisions involves the work of professionals in the architecture, engineering and construction (AEC) domain who nowadays collaborate via the methodology of building information modeling (BIM). At the same time, the citizen plays an integral role both in the data acquisition phase, while also being a beneficiary of the improved resource management strategies. In this paper, we present a prototype for a “digital energy twin” platform we designed in cooperation with the city of Regensburg. We show how our extensible platform de- sign can satisfy the various requirements of multiple user groups through a series of data processing solutions and visualizations, in- dicating valuable design and implementation guidelines for future projects. In particular, we focus on two example use cases concern- ing building electricity monitoring and BIM. By implementing a flexible data processing architecture we can involve citizens in the data acquisition process, meeting the demands of modern users regarding maximum transparency in the handling of their data.
A Wireless Low-power System for Digital Identification of Examinees (Including Covid-19 Checks)
(2022)
Indoor localization has been, for the past decade, a subject under intense development. There is, however, no currently available solution that covers all possible scenarios. Received Signal Strength Indicator (RSSI) based methods, although the most widely researched, still suffer from problems due to environment noise. In this paper, we present a system using Bluetooth Low Energy (BLE) beacons attached to the desks to localize students in exam rooms and, at the same time, automatically register them for the given exam. By using Kalman Filters (KFs) and discretizing the location task, the presented solution is capable of achieving 100% accuracy within a distance of 45cm from the center of the desk. As the pandemic gets more controlled, with our lives slowly transitioning back to normal, there are still sanitary measures being applied. An example being the necessity to show a certification of vaccination or previous disease. Those certifications need to be manually checked for everyone entering the university’s building, which requires time and staff. With that in mind, the automatic check for Covid certificates feature is also built into our system.
BACKGROUND
Differentiation of high-flow from low-flow vascular malformations (VMs) is crucial for therapeutic management of this orphan disease.
OBJECTIVE
A convolutional neural network (CNN) was evaluated for differentiation of peripheral vascular malformations (VMs) on T2-weighted short tau inversion recovery (STIR) MRI.
METHODS
527 MRIs (386 low-flow and 141 high-flow VMs) were randomly divided into training, validation and test set for this single-center study. 1) Results of the CNN's diagnostic performance were compared with that of two expert and four junior radiologists. 2) The influence of CNN's prediction on the radiologists' performance and diagnostic certainty was evaluated. 3) Junior radiologists' performance after self-training was compared with that of the CNN.
RESULTS
Compared with the expert radiologists the CNN achieved similar accuracy (92% vs. 97%, p = 0.11), sensitivity (80% vs. 93%, p = 0.16) and specificity (97% vs. 100%, p = 0.50). In comparison to the junior radiologists, the CNN had a higher specificity and accuracy (97% vs. 80%, p < 0.001; 92% vs. 77%, p < 0.001). CNN assistance had no significant influence on their diagnostic performance and certainty. After self-training, the junior radiologists' specificity and accuracy improved and were comparable to that of the CNN.
CONCLUSIONS
Diagnostic performance of the CNN for differentiating high-flow from low-flow VM was comparable to that of expert radiologists. CNN did not significantly improve the simulated daily practice of junior radiologists, self-training was more effective.
Vascular malformations (VMs) are a rare condition. They can be categorized into high-flow and low-flow VMs, which is a challenging task for radiologists. In this work, a very heterogeneous set of MRI images with only rough annotations are used for classification with a convolutional neural network. The main focus is to describe the challenging data set and strategies to deal with such data in terms of preprocessing, annotation usage and choice of the network architecture. We achieved a classification result of 89.47 % F1-score with a 3D ResNet 18.
The growing reliance on Artificial Intelligence (AI) in critical domains such as healthcare demands robust mechanisms to ensure the trustworthiness of these systems, especially when faced with unexpected or anomalous inputs. This paper introduces the Open Medical Imaging Benchmarks for Out-Of-Distribution Detection (OpenMIBOOD), a comprehensive framework for evaluating out-of-distribution (OOD) detection methods specifically in medical imaging contexts. OpenMIBOOD includes three benchmarks from diverse medical domains, encompassing 14 datasets divided into covariate-shifted in-distribution, near-OOD, and far-OOD categories. We evaluate 24 post-hoc methods across these benchmarks, providing a standardized reference to advance the development and fair comparison of OOD detection methods. Results reveal that findings from broad-scale OOD benchmarks in natural image domains do not translate to medical applications, underscoring the critical need for such benchmarks in the medical field. By mitigating the risk of exposing AI models to inputs outside their training distribution, OpenMIBOOD aims to support the advancement of reliable and trustworthy AI systems in healthcare. The repository is available at this https URL.
The growing reliance on Artificial Intelligence (AI) in critical domains such as healthcare demands robust mechanisms to ensure the trustworthiness of these systems, especially when faced with unexpected or anomalous inputs. This paper introduces the Open Medical Imaging Benchmarks for Out-Of-Distribution Detection (OpenMIBOOD), a comprehensive framework for evaluating out-of-distribution (OOD) detection methods specifically in medical imaging contexts. OpenMIBOOD includes three benchmarks from diverse medical domains, encompassing 14 datasets divided into covariate-shifted in-distribution, nearOOD, and far-OOD categories. We evaluate 24 post-hoc methods across these benchmarks, providing a standardized reference to advance the development and fair comparison of OODdetection methods. Results reveal that findings from broad-scale OOD benchmarks in natural image domains do not translate to medical applications, underscoring the critical need for such benchmarks in the medical field. By mitigating the risk of exposing AI models to inputs outside their training distribution, OpenMIBOOD aims to support the advancement of reliable and trustworthy AI systems in healthcare. The repository is available at https://github.com/remic-othr/OpenMIBOOD.
Purpose
Recognizing previously unseen classes with neural networks is a significant challenge due to their limited generalization capabilities. This issue is particularly critical in safety-critical domains such as medical applications, where accurate classification is essential for reliability and patient safety. Zero-shot learning methods address this challenge by utilizing additional semantic data, with their performance relying heavily on the quality of the generated embeddings.
Methods
This work investigates the use of full descriptive sentences, generated by a Sentence-BERT model, as class representations, compared to simpler category-based word embeddings derived from a BERT model. Additionally, the impact of z-score normalization as a post-processing step on these embeddings is explored. The proposed approach is evaluated on a multi-label generalized zero-shot learning task, focusing on the recognition of surgical instruments in endoscopic images from minimally invasive cholecystectomies.
Results
The results demonstrate that combining sentence embeddings and z-score normalization significantly improves model performance. For unseen classes, the AUROC improves from 43.9% to 64.9%, and the multi-label accuracy from 26.1% to 79.5%. Overall performance measured across both seen and unseen classes improves from 49.3% to 64.9% in AUROC and from 37.3% to 65.1% in multi-label accuracy, highlighting the effectiveness of our approach.
Conclusion
These findings demonstrate that sentence embeddings and z-score normalization can substantially enhance the generalization performance of zero-shot learning models. However, as the study is based on a single dataset, future work should validate the method across diverse datasets and application domains to establish its robustness and broader applicability.
This dataset is a subset of the original CATARACTS test dataset and is used by the OpenMIBOOD framework to evaluate a specific out-of-distribution setting.
When using this dataset, it is mandatory to cite the corresponding publication (OpenMIBOOD (10.1109/CVPR52734.2025.02410)) and follow the acknowledgement and citation requirements of the original dataset (CATARACTS).
The original CATARACTS dataset (associated publication,Homepage) consists of 50 videos of cataract surgeries, split into 25 train and 25 test videos.
This subset contains the frames of the first 5 test videos. Further, black frames at the beginning of each video were removed.
This dataset is a subset of the original Cholec80 dataset and is used by the OpenMIBOOD framework to evaluate a specific out-of-distribution setting.
When using this dataset, it is mandatory to cite the corresponding publication (OpenMIBOOD) and to follow the acknowledgement and citation requirements of the original dataset (Cholec80).
The original Cholec80 dataset (associated paper,Homepage) consists of 80 cholecystectomy surgery videos recorded at 25 fps, performed by 13 surgeons. It includes phase annotations (25 fps) and tool presence labels (1 fps), with phase definitions provided by a senior surgeon. A tool is considered present if at least half of its tip is visible. The dataset categorizes tools into seven types: Grasper, Bipolar, Hook, Scissors, Clipper, Irrigator, and Specimen bag. Multiple tools may be present in each frame. Additionally, 76 of the 80 videos exhibit a strong black vignette.
For this dataset subset, frames were extracted based on tool presence labels, selecting only those containing Grasper, Bipolar, Hook, or Clipper while ensuring that only a single tool appears per frame. To enhance visual consistency, the black vignette was removed by extracting an inner rectangular region, where applicable.