The underuse of AI in the health sector: Opportunity costs, success stories, risks and recommendations.

Purpose: This contribution explores the underuse of artificial intelligence (AI) in the health sector, what this means for practice, and how much the underuse can cost. Attention is drawn to the relevance of an issue that the European Parliament has outlined as a "major threat" in 2020. At its heart is the risk that research and development on trusted AI systems for medicine and digital health will pile up in lab centers without generating further practical relevance. Our analysis highlights why researchers, practitioners and especially policymakers, should pay attention to this phenomenon. Methods: The paper examines the ways in which governments and public agencies are addressing the underuse of AI. As governments and international organizations often acknowledge the limitations of their own initiatives, the contribution explores the causes of the current issues and suggests ways to improve initiatives for digital health. Results: Recommendations address the development of standards, models of regulatory governance, assessment of the opportunity costs of underuse of technology, and the urgency of the problem. Conclusions: The exponential pace of AI advances and innovations makes the risks of underuse of AI increasingly threatening.

A historical perspective of biomedical explainable AI research.

The black-box nature of most artificial intelligence (AI) models encourages the development of explainability methods to engender trust into the AI decision-making process. Such methods can be broadly categorized into two main types: post hoc explanations and inherently interpretable algorithms. We aimed at analyzing the possible associations between COVID-19 and the push of explainable AI (XAI) to the forefront of biomedical research. We automatically extracted from the PubMed database biomedical XAI studies related to concepts of causality or explainability and manually labeled 1,603 papers with respect to XAI categories. To compare the trends pre- and post-COVID-19, we fit a change point detection model and evaluated significant changes in publication rates. We show that the advent of COVID-19 in the beginning of 2020 could be the driving factor behind an increased focus concerning XAI, playing a crucial role in accelerating an already evolving trend. Finally, we present a discussion with future societal use and impact of XAI technologies and potential future directions for those who pursue fostering clinical trust with interpretable machine learning models.

Assessment of Mitochondrial Ca2+ Uptake.

Mitochondrial Ca2+ uptake regulates mitochondrial function and contributes to cell signaling. Accordingly, quantifying mitochondrial Ca2+ signals and elaborating the mechanisms that accomplish mitochondrial Ca2+ uptake are essential to gain our understanding of cell biology. Here, we describe the benefits and drawbacks of various established old and new techniques to assess dynamic changes of mitochondrial Ca2+ concentration ([Ca2+]mito) in a wide range of applications.

Mutation-based clustering and classification analysis reveals distinctive age groups and age-related biomarkers for glioma.

BACKGROUND: Malignant brain tumor diseases exhibit differences within molecular features depending on the patient's age. METHODS: In this work, we use gene mutation data from public resources to explore age specifics about glioma. We use both an explainable clustering as well as classification approach to find and interpret age-based differences in brain tumor diseases. We estimate age clusters and correlate age specific biomarkers. RESULTS: Age group classification shows known age specifics but also points out several genes which, so far, have not been associated with glioma classification. CONCLUSIONS: We highlight mutated genes to be characteristic for certain age groups and suggest novel age-based biomarkers and targets.

Open Data for Differential Network Analysis in Glioma.

The complexity of cancer diseases demands bioinformatic techniques and translational research based on big data and personalized medicine. Open data enables researchers to accelerate cancer studies, save resources and foster collaboration. Several tools and programming approaches are available for analyzing data, including annotation, clustering, comparison and extrapolation, merging, enrichment, functional association and statistics. We exploit openly available data via cancer gene expression analysis, we apply refinement as well as enrichment analysis via gene ontology and conclude with graph-based visualization of involved protein interaction networks as a basis for signaling. The different databases allowed for the construction of huge networks or specified ones consisting of high-confidence interactions only. Several genes associated to glioma were isolated via a network analysis from top hub nodes as well as from an outlier analysis. The latter approach highlights a mitogen-activated protein kinase next to a member of histondeacetylases and a protein phosphatase as genes uncommonly associated with glioma. Cluster analysis from top hub nodes lists several identified glioma-associated gene products to function within protein complexes, including epidermal growth factors as well as cell cycle proteins or RAS proto-oncogenes. By using selected exemplary tools and open-access resources for cancer research and differential network analysis, we highlight disturbed signaling components in brain cancer subtypes of glioma.

The role of artificial intelligence and machine learning in harmonization of high-resolution post-mortem MRI (virtopsy) with respect to brain microstructure.

Enhanced resolution of 7 T magnetic resonance imaging (MRI) scanners has considerably advanced our knowledge of structure and function in human and animal brains. Post-industrialized countries are particularly prone to an ever-increasing number of ageing individuals and ageing-associated neurodegenerative diseases. Neurodegenerative diseases are associated with volume loss in the affected brain. MRI diagnoses and monitoring of subtle volume changes in the ageing/diseased brains have the potential to become standard diagnostic tools. Even with the superior resolution of 7 T MRI scanners, the microstructural changes comprising cell types, cell numbers, and cellular processes, are still undetectable. Knowledge of origin, nature, and progression for microstructural changes are necessary to understand pathogenetic stages in the relentless neurodegenerative diseases, as well as to develop therapeutic tools that delay or stop neurodegenerative processes at their earliest stage. We illustrate the gap in resolution by comparing the identical regions of the post-mortem in situ 7 T MR images (virtual autopsy or virtopsy) with the histological observations in serial sections through the same brain. We also described the protocols and limitations associated with these comparisons, as well as the necessity of supercomputers and data management for "Big data". Analysis of neuron and/or glial number by using a body of mathematical tools and guidelines (stereology) is time-consuming, cumbersome, and still restricted to trained human investigators. Development of tools based on machine learning (ML) and artificial intelligence (AI) could considerably accelerate studies on localization, onset, and progression of neuron loss. Finally, these observations could disentangle the mechanisms of volume loss into stages of reversible atrophy and/or irreversible fatal cell death. This AI- and ML-based cooperation between virtopsy and histology could bridge the present gap between virtual reality and neuropathology. It could also culminate in the creation of an imaging-associated comprehensive database. This database would include genetic, clinical, epidemiological, and technical aspects that could help to alleviate or even stop the adverse effects of neurodegenerative diseases on affected individuals, their families, and society.

Use case driven evaluation of open databases for pediatric cancer research.

BACKGROUND: A plethora of Web resources are available offering information on clinical, pre-clinical, genomic and theoretical aspects of cancer, including not only the comprehensive cancer projects as ICGC and TCGA, but also less-known and more specialized projects on pediatric diseases such as PCGP. However, in case of data on childhood cancer there is very little information openly available. Several web-based resources and tools offer general biomedical data which are not purpose-built, for neither pediatric nor cancer analysis. Additionally, many Web resources on cancer focus on incidence data and statistical social characteristics as well as self-regulating communities. METHODS: We summarize those resources which are open and are considered to support scientific fundamental research, while we address our comparison to 11 identified pediatric cancer-specific resources (5 tools, 6 databases). The evaluation consists of 5 use cases on the example of brain tumor research and covers user-defined search scenarios as well as data mining tasks, also examining interactive visual analysis features. RESULTS: Web resources differ in terms of information quantity and presentation. Pedican lists an abundance of entries with few selection features. PeCan and PedcBioPortal include visual analysis tools while the latter integrates published and new consortia-based data. UCSC Xena Browser offers an in-depth analysis of genomic data. ICGC data portal provides various features for data analysis and an option to submit own data. Its focus lies on adult Pan-Cancer projects. Pediatric Pan-Cancer datasets are being integrated into PeCan and PedcBioPortal. Comparing information on prominent mutations within glioma discloses well-known, unknown, possible, as well as inapplicable biomarkers. This summary further emphasizes the varying data allocation. Tested tools show advantages and disadvantages, depending on the respective use case scenario, providing inhomogeneous data quantity and information specifics. CONCLUSIONS: Web resources on specific pediatric cancers are less abundant and less-known compared to those offering adult cancer research data. Meanwhile, current efforts of ongoing pediatric data collection and Pan-Cancer projects indicate future opportunities for childhood cancer research, that is greatly needed for both fundamental as well as clinical research.

In silico cancer research towards 3R.

BACKGROUND: Improving our understanding of cancer and other complex diseases requires integrating diverse data sets and algorithms. Intertwining in vivo and in vitro data and in silico models are paramount to overcome intrinsic difficulties given by data complexity. Importantly, this approach also helps to uncover underlying molecular mechanisms. Over the years, research has introduced multiple biochemical and computational methods to study the disease, many of which require animal experiments. However, modeling systems and the comparison of cellular processes in both eukaryotes and prokaryotes help to understand specific aspects of uncontrolled cell growth, eventually leading to improved planning of future experiments. According to the principles for humane techniques milestones in alternative animal testing involve in vitro methods such as cell-based models and microfluidic chips, as well as clinical tests of microdosing and imaging. Up-to-date, the range of alternative methods has expanded towards computational approaches, based on the use of information from past in vitro and in vivo experiments. In fact, in silico techniques are often underrated but can be vital to understanding fundamental processes in cancer. They can rival accuracy of biological assays, and they can provide essential focus and direction to reduce experimental cost. MAIN BODY: We give an overview on in vivo, in vitro and in silico methods used in cancer research. Common models as cell-lines, xenografts, or genetically modified rodents reflect relevant pathological processes to a different degree, but can not replicate the full spectrum of human disease. There is an increasing importance of computational biology, advancing from the task of assisting biological analysis with network biology approaches as the basis for understanding a cell's functional organization up to model building for predictive systems. CONCLUSION: Underlining and extending the in silico approach with respect to the 3Rs for replacement, reduction and refinement will lead cancer research towards efficient and effective precision medicine. Therefore, we suggest refined translational models and testing methods based on integrative analyses and the incorporation of computational biology within cancer research.

In silico modeling for tumor growth visualization.

BACKGROUND: Cancer is a complex disease. Fundamental cellular based studies as well as modeling provides insight into cancer biology and strategies to treatment of the disease. In silico models complement in vivo models. Research on tumor growth involves a plethora of models each emphasizing isolated aspects of benign and malignant neoplasms. Biologists and clinical scientists are often overwhelmed by the mathematical background knowledge necessary to grasp and to apply a model to their own research. RESULTS: We aim to provide a comprehensive and expandable simulation tool to visualizing tumor growth. This novel Web-based application offers the advantage of a user-friendly graphical interface with several manipulable input variables to correlate different aspects of tumor growth. By refining model parameters we highlight the significance of heterogeneous intercellular interactions on tumor progression. Within this paper we present the implementation of the Cellular Potts Model graphically presented through Cytoscape.js within a Web application. The tool is available under the MIT license at https://github.com/davcem/cpm-cytoscape and http://styx.cgv.tugraz.at:8080/cpm-cytoscape/ . CONCLUSION: In-silico methods overcome the lack of wet experimental possibilities and as dry method succeed in terms of reduction, refinement and replacement of animal experimentation, also known as the 3R principles. Our visualization approach to simulation allows for more flexible usage and easy extension to facilitate understanding and gain novel insight. We believe that biomedical research in general and research on tumor growth in particular will benefit from the systems biology perspective.

Integrated web visualizations for protein-protein interaction databases.

BACKGROUND: Understanding living systems is crucial for curing diseases. To achieve this task we have to understand biological networks based on protein-protein interactions. Bioinformatics has come up with a great amount of databases and tools that support analysts in exploring protein-protein interactions on an integrated level for knowledge discovery. They provide predictions and correlations, indicate possibilities for future experimental research and fill the gaps to complete the picture of biochemical processes. There are numerous and huge databases of protein-protein interactions used to gain insights into answering some of the many questions of systems biology. Many computational resources integrate interaction data with additional information on molecular background. However, the vast number of diverse Bioinformatics resources poses an obstacle to the goal of understanding. We present a survey of databases that enable the visual analysis of protein networks. RESULTS: We selected M=10 out of N=53 resources supporting visualization, and we tested against the following set of criteria: interoperability, data integration, quantity of possible interactions, data visualization quality and data coverage. The study reveals differences in usability, visualization features and quality as well as the quantity of interactions. StringDB is the recommended first choice. CPDB presents a comprehensive dataset and IntAct lets the user change the network layout. A comprehensive comparison table is available via web. The supplementary table can be accessed on http://tinyurl.com/PPI-DB-Comparison-2015. CONCLUSIONS: Only some web resources featuring graph visualization can be successfully applied to interactive visual analysis of protein-protein interaction. Study results underline the necessity for further enhancements of visualization integration in biochemical analysis tools. Identified challenges are data comprehensiveness, confidence, interactive feature and visualization maturing.

Assessment of mitochondrial Ca²âº uptake.

Mitochondrial Ca(2+) uptake regulates mitochondrial function and contributes to cell signaling. Accordingly, quantifying mitochondrial Ca(2+) signals and elaborating the mechanisms that accomplish mitochondrial Ca(2+) uptake are essential to gain our understanding of cell biology. Here, we describe the benefits and drawbacks of various established old and new techniques to assess dynamic changes of mitochondrial Ca(2+) concentration ([Ca(2+)]mito) in a wide range of applications.

Mitochondrial Ca(2+) uniporter (MCU)-dependent and MCU-independent Ca(2+) channels coexist in the inner mitochondrial membrane.

A protein referred to as CCDC109A and then renamed to mitochondrial calcium uniporter (MCU) has recently been shown to accomplish mitochondrial Ca(2+) uptake in different cell types. In this study, we investigated whole-mitoplast inward cation currents and single Ca(2+) channel activities in mitoplasts prepared from stable MCU knockdown HeLa cells using the patch-clamp technique. In whole-mitoplast configuration, diminution of MCU considerably reduced inward Ca(2+) and Na(+) currents. This was accompanied by a decrease in occurrence of single channel activity of the intermediate conductance mitochondrial Ca(2+) current (i-MCC). However, ablation of MCU yielded a compensatory 2.3-fold elevation in the occurrence of the extra large conductance mitochondrial Ca(2+) current (xl-MCC), while the occurrence of bursting currents (b-MCC) remained unaltered. These data reveal i-MCC as MCU-dependent current while xl-MCC and b-MCC seem to be rather MCU-independent, thus, pointing to the engagement of at least two molecularly distinct mitochondrial Ca(2+) channels.

Molecularly distinct routes of mitochondrial Ca2+ uptake are activated depending on the activity of the sarco/endoplasmic reticulum Ca2+ ATPase (SERCA).

The transfer of Ca(2+) across the inner mitochondrial membrane is an important physiological process linked to the regulation of metabolism, signal transduction, and cell death. While the definite molecular composition of mitochondrial Ca(2+) uptake sites remains unknown, several proteins of the inner mitochondrial membrane, that are likely to accomplish mitochondrial Ca(2+) fluxes, have been described: the novel uncoupling proteins 2 and 3, the leucine zipper-EF-hand containing transmembrane protein 1 and the mitochondrial calcium uniporter. It is unclear whether these proteins contribute to one unique mitochondrial Ca(2+) uptake pathway or establish distinct routes for mitochondrial Ca(2+) sequestration. In this study, we show that a modulation of Ca(2+) release from the endoplasmic reticulum by inhibition of the sarco/endoplasmatic reticulum ATPase modifies cytosolic Ca(2+) signals and consequently switches mitochondrial Ca(2+) uptake from an uncoupling protein 3- and mitochondrial calcium uniporter-dependent, but leucine zipper-EF-hand containing transmembrane protein 1-independent to a leucine zipper-EF-hand containing transmembrane protein 1- and mitochondrial calcium uniporter-mediated, but uncoupling protein 3-independent pathway. Thus, the activity of sarco/endoplasmatic reticulum ATPase is significant for the mode of mitochondrial Ca(2+) sequestration and determines which mitochondrial proteins might actually accomplish the transfer of Ca(2+) across the inner mitochondrial membrane. Moreover, our findings herein support the existence of distinct mitochondrial Ca(2+) uptake routes that might be essential to ensure an efficient ion transfer into mitochondria despite heterogeneous cytosolic Ca(2+) rises.

Characterization of distinct single-channel properties of Ca²âº inward currents in mitochondria.

Previous studies have demonstrated several molecularly distinct players involved in mitochondrial Ca(2+) uptake. In the present study, electrophysiological recordings on mitoplasts that were isolated from HeLa cells were performed in order to biophysically and pharmacologically characterize Ca(2+) currents across the inner mitochondrial membrane. In mitoplast-attached configuration with 105 mM Ca(2+) as a charge carrier, three distinct channel conductances of 11, 23, and 80 pS were observed. All types of mitochondrial currents were voltage-dependent and essentially depended on the presence of Ca(2+) in the pipette. The 23 pS channel exhibited burst kinetics. Though all channels were sensitive to ruthenium red, their sensitivity was different. The 11 and 23 pS channels exhibited a lower sensitivity to ruthenium red than the 80 pS channel. The activities of all channels persisted in the presence of cylosporin A, CGP 37187, various K(+)-channel inhibitors, and Cl(-) channel blockers disodium 4,4'-diisothiocyanatostilbene-2,2'-disulfonate and niflumic acid. Collectively, our data identified multiple conductances of Ca(2+) currents in mitoplasts isolated from HeLa cells, thus challenging the dogma of only one unique mitochondrial Ca(2+) uniporter.

Studying mitochondrial Ca(2+) uptake - a revisit.

Mitochondrial Ca(2+) sequestration is a well-known process that is involved in various physiological and pathological mechanisms. Using isolated suspended mitochondria one unique mitochondrial Ca(2+) uniporter was considered to account ubiquitously for the transfer of Ca(2+) into these organelles. However, by applying alternative techniques for measuring mitochondrial Ca(2+) uptake evidences for molecularly distinct mitochondrial Ca(2+) carriers accumulated recently. Herein we compared different methodical approaches of studying mitochondrial Ca(2+) uptake. Patch clamp technique on mitoplasts from endothelial and HeLa cells revealed the existence of three and two mitoplast Ca(2+) currents (I(CaMito)), respectively. According to their conductance, these channels were named small (s-), intermediate (i-), large (l-) and extra-large (xl-) mitoplast Ca(2+) currents (MCC). i-MCC was found in mitoplasts of both cell types whereas s-MCC and l-MCC or xl-MCC were/was exclusively found in mitoplasts from endothelial cells or HeLa cells. The comparison of mitochondrial Ca(2+) signals, measured either indirectly by sensing extra-mitochondrial Ca(2+) or directly by recording changes of the matrix Ca(2+), showed different Ca(2+) sensitivities of the distinct mitochondrial Ca(2+) uptake routes. Subpopulations of mitochondria with different Ca(2+) uptake capacities in intact endothelial cells could be identified using Rhod-2/AM. In contrast, cells expressing mitochondrial targeted pericam or cameleon (4mtD3cpv) showed homogeneous mitochondrial Ca(2+) signals in response to cell stimulation. The comparison of different experimental approaches and protocols using isolated organelles, permeabilized and intact cells, pointed to cell-type specific and versatile pathways for mitochondrial Ca(2+) uptake. Moreover, this work highlights the necessity of the utilization of multiple technical approaches to study the complexity of mitochondrial Ca(2+) homeostasis.

Leucine zipper EF hand-containing transmembrane protein 1 (Letm1) and uncoupling proteins 2 and 3 (UCP2/3) contribute to two distinct mitochondrial Ca2+ uptake pathways.

Cytosolic Ca(2+) signals are transferred into mitochondria over a huge concentration range. In our recent work we described uncoupling proteins 2 and 3 (UCP2/3) to be fundamental for mitochondrial uptake of high Ca(2+) domains in mitochondria-ER junctions. On the other hand, the leucine zipper EF hand-containing transmembrane protein 1 (Letm1) was identified as a mitochondrial Ca(2+)/H(+) antiporter that achieved mitochondrial Ca(2+) sequestration at small Ca(2+) increases. Thus, the contributions of Letm1 and UCP2/3 to mitochondrial Ca(2+) uptake were compared in endothelial cells. Knock-down of Letm1 did not affect the UCP2/3-dependent mitochondrial uptake of intracellularly released Ca(2+) but strongly diminished the transfer of entering Ca(2+) into mitochondria, subsequently, resulting in a reduction of store-operated Ca(2+) entry (SOCE). Knock-down of Letm1 and UCP2/3 did neither impact on cellular ATP levels nor the membrane potential. The enhanced mitochondrial Ca(2+) signals in cells overexpressing UCP2/3 rescued SOCE upon Letm1 knock-down. In digitonin-permeabilized cells, Letm1 exclusively contributed to mitochondrial Ca(2+) uptake at low Ca(2+) conditions. Neither the Letm1- nor the UCP2/3-dependent mitochondrial Ca(2+) uptake was affected by a knock-down of mRNA levels of mitochondrial calcium uptake 1 (MICU1), a protein that triggers mitochondrial Ca(2+) uptake in HeLa cells. Our data indicate that Letm1 and UCP2/3 independently contribute to two distinct, mitochondrial Ca(2+) uptake pathways in intact endothelial cells.