1 - 4 of 4
-
PublicationDeepKneeExplainer: Explainable knee osteoarthritis diagnosis from radiographs and magnetic resonance imaging( 2021)Osteoarthritis (OA) is a degenerative joint disease, which significantly affects middle-aged and elderly people. Although primarily identified via hyaline cartilage change based on medical images, technical bottlenecks like noise, artifacts, and modality impose an enormous challenge on high-precision, objective, and efficient early quantification of OA. Owing to recent advancements, approaches based on neural networks (DNNs) have shown outstanding success in this application domain. However, due to nested non-linear and complex structures, DNNs are mostly opaque and perceived as black-box methods, which raises numerous legal and ethical concerns. Moreover, these approaches do not have the ability to provide the reasoning behind diagnosis decisions in the way humans would do, which poses an additional risk in the clinical setting. In this paper, we propose a novel explainable method for knee OA diagnosis based on radiographs and magnetic resonance imaging (MRI), which we called DeepKneeExplainer. First, we comprehensively preprocess MRIs and radiographs through the deep-stacked transformation technique against possible noises and artifacts that could contain unseen images for domain generalization. Then, we extract the region of interests (ROIs) by employing U-Net architecture with ResNet backbone. To classify the cohorts, we train DenseNet and VGG architectures on the extracted ROIs. Finally, we highlight class-discriminating regions using gradient-guided class activation maps (Grad-CAM++) and layer-wise relevance propagation (LRP), followed by providing human-interpretable explanations of the predictions. Comprehensive experiments based on the multicenter osteoarthritis study (MOST) cohorts, our approach yields up to 91% classification accuracy, outperforming comparable state-of-the-art approaches. We hope that our results will encourage medical researchers and developers to adopt explainable methods and DNN-based analytic pipelines towards an increasing acceptance and adoption of AI-assisted applications in the clinical practice for improved knee OA diagnoses.
-
PublicationFinding and analysing energy research funding data: The EnArgus system( 2021)
-
PublicationEx ante assessment of disruptive threats: Identifying relevant threats before one is disrupted( 2020)The shortening of product life-cycles accompanied by the rapid development of new products and dissolving industry boundaries are indicative of a multitude of potentially disruptive threats. The survival of incumbents depends on their capability to effectively anticipate and manage such threats. Thus, the early anticipation of disruptive threats to react or prepare for their impacts is a crucial topic in practice and academia. Although the current body of knowledge provides numerous approaches to disruption anticipation, a comprehensive conceptualisation of the evolution of disruptive threats is missing. Moreover, incumbents lack guidance on how to effectively anticipate disruptive threats. To address this gap, we propose the Disruption Evolution Framework (DEF), which conceptualises the course of disruptive threats along three phases (i.e. threat possible, apparent, and materialised) as well as distinguishes four interrelated categories of signals (i.e. context, catalyst, capability, and company signals) and threats (i.e. customer, competitor, product, and policy threats). Building on the DEF, we also propose the Disruptability Assessment Method (DAM), which enables incumbents to systematically assess disruptive threats via a step-by-step procedure. We evaluated the DAM in the Corporate Development and the Global Digital Partnerships departments of an insurance company. Overall, our work contributes to the descriptive and prescriptive knowledge on disruption anticipation.
-
PublicationInter-technology relationship networks: Arranging technologies through text mining( 2019)Ongoing advances in digital technologies - which enable new products, services, and business models - have fundamentally affected business and society through several waves of digitalization. When analyzing digital technologies, a dynamic system or an ecosystem model that represents interrelated technologies is beneficial owing to the systemic character of digital technologies. Using an assembly-based process model for situational method engineering, and following the design science research paradigm, we develop an analytical method to generate technology-related network data that retraces elapsed patterns of technological change. We consider the technological distances that characterize technologies' proximities and dependencies. We use established text mining techniques and draw from technology innovation research as justificatory knowledge. The proposed method processes textual data from different information sources into an analyzable and readable inter-technology relationship network. To evaluate the method, we use exemplary digital technologies from the big data analytics domain as an application scenario.
