Helmut Denk

Bio: Helmut Denk is an academic researcher from Medical University of Graz. The author has contributed to research in topics: Mallory body & Cholestasis. The author has an hindex of 55, co-authored 193 publications receiving 15354 citations. Previous affiliations of Helmut Denk include Yale University & Odense University Hospital.

Causability and explainability of artificial intelligence in medicine.

Abstract: Explainable artificial intelligence (AI) is attracting much interest in medicine. Technically, the problem of explainability is as old as AI itself and classic AI represented comprehensible retraceable approaches. However, their weakness was in dealing with uncertainties of the real world. Through the introduction of probabilistic learning, applications became increasingly successful, but increasingly opaque. Explainable AI deals with the implementation of transparency and traceability of statistical black-box machine learning methods, particularly deep learning (DL). We argue that there is a need to go beyond explainable AI. To reach a level of explainable medicine we need causability. In the same way that usability encompasses measurements for the quality of use, causability encompasses measurements for the quality of explanations. In this article, we provide some necessary definitions to discriminate between explainability and causability as well as a use-case of DL interpretation and of human explanation in histopathology. The main contribution of this article is the notion of causability, which is differentiated from explainability in that causability is a property of a person, while explainability is a property of a system This article is categorized under: Fundamental Concepts of Data and Knowledge > Human Centricity and User Interaction.

p62 Is a Common Component of Cytoplasmic Inclusions in Protein Aggregation Diseases

Abstract: Exposure of cells to stress, particularly oxidative stress, leads to misfolding of proteins and, if they are not refolded or degraded, to cytoplasmic protein aggregates. Protein aggregates are characteristic features of a variety of chronic toxic and degenerative diseases, such as Mallory bodies (MBs) in hepatocytes in alcoholic and non-alcoholic steatohepatitis, neurofibrillary tangles in neurons in Alzheimer's, and Lewy bodies in Parkinson's disease. Using 2D gel electrophoresis and mass spectrometry, we identified p62 as a novel MB component. p62 and cytokeratins (CKs) are major MB constituents; HSP 70, HSP 25, and ubiquitinated CKs are also present. These proteins characterize MBs as a prototype of disease-associated cytoplasmic inclusions generated by stress-induced protein misfolding. As revealed by transfection of tissue culture cells overexpressed p62 did not induce aggregation of regular CK filaments but selectively bound to misfolded and ubiquitinated CKs. The general role of p62 in the cellular response to misfolded proteins was substantiated by detection of p62 in other cytoplasmic inclusions, such as neurofibrillary tangles, Lewy bodies, Rosenthal fibers, intracytoplasmic hyaline bodies in hepatocellular carcinoma, and alpha1-antitrypsin aggregates. The presence of p62 along with other stress proteins and ubiquitin in cytoplasmic inclusions indicates deposition as aggregates as a third line of defense against misfolded proteins in addition to refolding and degradation.

Nonalcoholic fatty liver disease.

Abstract: Nonalcoholic fatty liver disease (NAFLD) is present in up to one third of the general population and in the majority of patients with metabolic risk factors such as obesity and diabetes. Insulin resistance is a key pathogenic factor resulting in hepatic fat accumulation. Recent evidence demonstrates NAFLD in turn, exacerbates hepatic insulin resistance and often precedes glucose intolerance. Once hepatic steatosis is established, other factors including oxidative stress, mitochondrial dysfunction, gut-derived lipopolysaccharide and adipocytokines, may promote hepatocellular damage, inflammation and progressive liver disease. Confirmation of the diagnosis of NAFLD can usually be achieved by imaging studies, however staging the disease requires a liver biopsy. NAFLD is associated with an increased risk of all-cause death, probably because of complications of insulin resistance such as vascular disease, as well as due to cirrhosis and hepatocellular carcinoma, which occurs in a minority of patients. NAFLD is also now recognized to account for a substantial proportion of patients previously diagnosed with 'cryptogenic cirrhosis'. Diabetes, obesity and the necroinflammatory form of NAFLD known as non-alcoholic steatohepatitis, are risk factors for progressive liver disease. Current treatment relies on weight loss and exercise, although various insulin-sensitizing medications appear promising. Further research is needed to identify which patients will achieve the most benefit from therapy.