Open Access

Bericht mit Forschungsprojekten aus verschiedenen Bereichen der OTH Regensburg mit dem Schwerpunktthema "Künstliche Intelligenz" und einem Gespräch zur "Medizin der Zukunft"

The effective use of digital technology and disruptive innovations are increasingly shaping the way companies survive in today’s markets. Consequently, the need for German Small and Medium Enterprises (SMEs) to perform a digital transformation, which implies the creation of a new business model through sophisticated technologies, is gaining significance for succeeding in the digital age. While most German SMEs have already accomplished a digitization of their collaboration and communication,a profound digital transformation is still imminent for the majority. Based on the investigation of existing literature on the subject, this paper demonstrates the suitability of design thinking methods to develop strategic thrust for a digital transformation. The key potential of design thinking is rooted in its ability to creatively solve problems and reinforce skills needed to address dynamic environments. An exemplary result of using design thinking to develop strategic thrust is presented with the fictional business model GreenCube. The concept aims to assist German SMEs in performing a digital transformation by simplifying the incorporation of sophisticated digital technologies and providing opportunities for networking and thus integrating into an ecosystem. Despite their potential, design thinking activities involve several challenges including the incongruence of pre-defined specifications and designed concepts, the gap between a focus on time-reduction and a focus on experimentation, and the lack of measurability of design thinking gains. Nevertheless, the nature of businesses’internal and external environment is becoming ever more fast-paced, making the ability to manage change increasingly critical for German SMEs to establish competitive advantage.

This paper presents a novel two-stage approach for computed tomography (CT) reconstruction, focusing on sparse-angle and low-dose setups to minimize radiation exposure while maintaining high image quality. Two-stage approaches consist of an initial reconstruction followed by a neural network for image refinement. In the initial reconstruction, we apply the backprojection (BP) instead of the traditional filtered backprojection (FBP). This enhances computational speed and offers potential advantages for more complex geometries, such as fan-beam and cone-beam CT. Additionally, BP addresses noise and artifacts in sparse-angle CT by leveraging its inherent noise-smoothing effect, which reduces streaking artifacts common in FBP reconstructions. For the second stage, we fine-tune the DRUNet proposed by Zhang et al. to further improve reconstruction quality. We call our method BP-DRUNet and evaluate its performance on a synthetically generated ellipsoid dataset alongside thewell-established LoDoPaBCT dataset. Our results show that BP-DRUNet produces competetive results in terms of PSNR and SSIM metrics compared to the FBP-based counterpart, FBPDRUNet, and delivers visually competitive results across all tested angular setups.