TY - JOUR A1 - Loh, Frank A1 - Poignée, Fabian A1 - Wamser, Florian A1 - Leidinger, Ferdinand A1 - Hoßfeld, Tobias T1 - Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming JF - Sensors N2 - Streaming video is responsible for the bulk of Internet traffic these days. For this reason, Internet providers and network operators try to make predictions and assessments about the streaming quality for an end user. Current monitoring solutions are based on a variety of different machine learning approaches. The challenge for providers and operators nowadays is that existing approaches require large amounts of data. In this work, the most relevant quality of experience metrics, i.e., the initial playback delay, the video streaming quality, video quality changes, and video rebuffering events, are examined using a voluminous data set of more than 13,000 YouTube video streaming runs that were collected with the native YouTube mobile app. Three Machine Learning models are developed and compared to estimate playback behavior based on uplink request information. The main focus has been on developing a lightweight approach using as few features and as little data as possible, while maintaining state-of-the-art performance. KW - HTTP adaptive video streaming KW - quality of experience prediction KW - machine learning Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-241121 SN - 1424-8220 VL - 21 IS - 12 ER - TY - JOUR A1 - März, Juliane A1 - Kurlbaum, Max A1 - Roche-Lancaster, Oisin A1 - Deutschbein, Timo A1 - Peitzsch, Mirko A1 - Prehn, Cornelia A1 - Weismann, Dirk A1 - Robledo, Mercedes A1 - Adamski, Jerzy A1 - Fassnacht, Martin A1 - Kunz, Meik A1 - Kroiss, Matthias T1 - Plasma Metabolome Profiling for the Diagnosis of Catecholamine Producing Tumors JF - Frontiers in Endocrinology N2 - Context Pheochromocytomas and paragangliomas (PPGL) cause catecholamine excess leading to a characteristic clinical phenotype. Intra-individual changes at metabolome level have been described after surgical PPGL removal. The value of metabolomics for the diagnosis of PPGL has not been studied yet. Objective Evaluation of quantitative metabolomics as a diagnostic tool for PPGL. Design Targeted metabolomics by liquid chromatography-tandem mass spectrometry of plasma specimens and statistical modeling using ML-based feature selection approaches in a clinically well characterized cohort study. Patients Prospectively enrolled patients (n=36, 17 female) from the Prospective Monoamine-producing Tumor Study (PMT) with hormonally active PPGL and 36 matched controls in whom PPGL was rigorously excluded. Results Among 188 measured metabolites, only without considering false discovery rate, 4 exhibited statistically significant differences between patients with PPGL and controls (histidine p=0.004, threonine p=0.008, lyso PC a C28:0 p=0.044, sum of hexoses p=0.018). Weak, but significant correlations for histidine, threonine and lyso PC a C28:0 with total urine catecholamine levels were identified. Only the sum of hexoses (reflecting glucose) showed significant correlations with plasma metanephrines. By using ML-based feature selection approaches, we identified diagnostic signatures which all exhibited low accuracy and sensitivity. The best predictive value (sensitivity 87.5%, accuracy 67.3%) was obtained by using Gradient Boosting Machine Modelling. Conclusions The diabetogenic effect of catecholamine excess dominates the plasma metabolome in PPGL patients. While curative surgery for PPGL led to normalization of catecholamine-induced alterations of metabolomics in individual patients, plasma metabolomics are not useful for diagnostic purposes, most likely due to inter-individual variability. KW - adrenal KW - pheochromocytoma KW - paraganglioma KW - targeted metabolomics KW - mass spectronomy KW - catecholamines KW - machine learning KW - feature selection Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-245710 SN - 1664-2392 VL - 12 ER - TY - JOUR A1 - Allgaier, Johannes A1 - Schlee, Winfried A1 - Langguth, Berthold A1 - Probst, Thomas A1 - Pryss, Rüdiger T1 - Predicting the Gender of Individuals with Tinnitus based on Daily Life Data of the TrackYourTinnitus mHealth Platform JF - Scientific Reports N2 - Tinnitus is an auditory phantom perception in the absence of an external sound stimulation. People with tinnitus often report severe constraints in their daily life. Interestingly, indications exist on gender differences between women and men both in the symptom profile as well as in the response to specific tinnitus treatments. In this paper, data of the TrackYourTinnitus platform (TYT) were analyzed to investigate whether the gender of users can be predicted. In general, the TYT mobile Health crowdsensing platform was developed to demystify the daily and momentary variations of tinnitus symptoms over time. The goal of the presented investigation is a better understanding of gender-related differences in the symptom profiles of users from TYT. Based on two questionnaires of TYT, four machine learning based classifiers were trained and analyzed. With respect to the provided daily answers, the gender of TYT users can be predicted with an accuracy of 81.7%. In this context, worries, difficulties in concentration, and irritability towards the family are the three most important characteristics for predicting the gender. Note that in contrast to existing studies on TYT, daily answers to the worst symptom question were firstly investigated in more detail. It was found that results of this question significantly contribute to the prediction of the gender of TYT users. Overall, our findings indicate gender-related differences in tinnitus and tinnitus-related symptoms. Based on evidence that gender impacts the development of tinnitus, the gathered insights can be considered relevant and justify further investigations in this direction. KW - computer science KW - machine learning KW - psychology KW - signs and symptoms Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-261753 VL - 11 IS - 1 ER - TY - JOUR A1 - Dirscherl, Mariel A1 - Dietz, Andreas J. A1 - Kneisel, Christof A1 - Kuenzer, Claudia T1 - A novel method for automated supraglacial lake mapping in Antarctica using Sentinel-1 SAR imagery and deep learning JF - Remote Sensing N2 - Supraglacial meltwater accumulation on ice sheets can be a main driver for accelerated ice discharge, mass loss, and global sea-level-rise. With further increasing surface air temperatures, meltwater-induced hydrofracturing, basal sliding, or surface thinning will cumulate and most likely trigger unprecedented ice mass loss on the Greenland and Antarctic ice sheets. While the Greenland surface hydrological network as well as its impacts on ice dynamics and mass balance has been studied in much detail, Antarctic supraglacial lakes remain understudied with a circum-Antarctic record of their spatio-temporal development entirely lacking. This study provides the first automated supraglacial lake extent mapping method using Sentinel-1 synthetic aperture radar (SAR) imagery over Antarctica and complements the developed optical Sentinel-2 supraglacial lake detection algorithm presented in our companion paper. In detail, we propose the use of a modified U-Net for semantic segmentation of supraglacial lakes in single-polarized Sentinel-1 imagery. The convolutional neural network (CNN) is implemented with residual connections for optimized performance as well as an Atrous Spatial Pyramid Pooling (ASPP) module for multiscale feature extraction. The algorithm is trained on 21,200 Sentinel-1 image patches and evaluated in ten spatially or temporally independent test acquisitions. In addition, George VI Ice Shelf is analyzed for intra-annual lake dynamics throughout austral summer 2019/2020 and a decision-level fused Sentinel-1 and Sentinel-2 maximum lake extent mapping product is presented for January 2020 revealing a more complete supraglacial lake coverage (~770 km\(^2\)) than the individual single-sensor products. Classification results confirm the reliability of the proposed workflow with an average Kappa coefficient of 0.925 and a F\(_1\)-score of 93.0% for the supraglacial water class across all test regions. Furthermore, the algorithm is applied in an additional test region covering supraglacial lakes on the Greenland ice sheet which further highlights the potential for spatio-temporal transferability. Future work involves the integration of more training data as well as intra-annual analyses of supraglacial lake occurrence across the whole continent and with focus on supraglacial lake development throughout a summer melt season and into Antarctic winter. KW - Antarctica KW - Antarctic ice sheet KW - supraglacial lakes KW - ice sheet hydrology KW - Sentinel-1 KW - remote sensing KW - machine learning KW - deep learning KW - semantic segmentation KW - convolutional neural network Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-222998 SN - 2072-4292 VL - 13 IS - 2 ER - TY - JOUR A1 - Janiesch, Christian A1 - Zschech, Patrick A1 - Heinrich, Kai T1 - Machine learning and deep learning JF - Electronic Markets N2 - Today, intelligent systems that offer artificial intelligence capabilities often rely on machine learning. Machine learning describes the capacity of systems to learn from problem-specific training data to automate the process of analytical model building and solve associated tasks. Deep learning is a machine learning concept based on artificial neural networks. For many applications, deep learning models outperform shallow machine learning models and traditional data analysis approaches. In this article, we summarize the fundamentals of machine learning and deep learning to generate a broader understanding of the methodical underpinning of current intelligent systems. In particular, we provide a conceptual distinction between relevant terms and concepts, explain the process of automated analytical model building through machine learning and deep learning, and discuss the challenges that arise when implementing such intelligent systems in the field of electronic markets and networked business. These naturally go beyond technological aspects and highlight issues in human-machine interaction and artificial intelligence servitization. KW - analytical model building KW - machine learning KW - deep learning KW - artificial intelligence KW - artificial neural networks Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-270155 SN - 1422-8890 VL - 31 IS - 3 ER - TY - JOUR A1 - Marquardt, André A1 - Solimando, Antonio Giovanni A1 - Kerscher, Alexander A1 - Bittrich, Max A1 - Kalogirou, Charis A1 - Kübler, Hubert A1 - Rosenwald, Andreas A1 - Bargou, Ralf A1 - Kollmannsberger, Philip A1 - Schilling, Bastian A1 - Meierjohann, Svenja A1 - Krebs, Markus T1 - Subgroup-Independent Mapping of Renal Cell Carcinoma — Machine Learning Reveals Prognostic Mitochondrial Gene Signature Beyond Histopathologic Boundaries JF - Frontiers in Oncology N2 - Background: Renal cell carcinoma (RCC) is divided into three major histopathologic groups—clear cell (ccRCC), papillary (pRCC) and chromophobe RCC (chRCC). We performed a comprehensive re-analysis of publicly available RCC datasets from the TCGA (The Cancer Genome Atlas) database, thereby combining samples from all three subgroups, for an exploratory transcriptome profiling of RCC subgroups. Materials and Methods: We used FPKM (fragments per kilobase per million) files derived from the ccRCC, pRCC and chRCC cohorts of the TCGA database, representing transcriptomic data of 891 patients. Using principal component analysis, we visualized datasets as t-SNE plot for cluster detection. Clusters were characterized by machine learning, resulting gene signatures were validated by correlation analyses in the TCGA dataset and three external datasets (ICGC RECA-EU, CPTAC-3-Kidney, and GSE157256). Results: Many RCC samples co-clustered according to histopathology. However, a substantial number of samples clustered independently from histopathologic origin (mixed subgroup)—demonstrating divergence between histopathology and transcriptomic data. Further analyses of mixed subgroup via machine learning revealed a predominant mitochondrial gene signature—a trait previously known for chRCC—across all histopathologic subgroups. Additionally, ccRCC samples from mixed subgroup presented an inverse correlation of mitochondrial and angiogenesis-related genes in the TCGA and in three external validation cohorts. Moreover, mixed subgroup affiliation was associated with a highly significant shorter overall survival for patients with ccRCC—and a highly significant longer overall survival for chRCC patients. Conclusions: Pan-RCC clustering according to RNA-sequencing data revealed a distinct histology-independent subgroup characterized by strengthened mitochondrial and weakened angiogenesis-related gene signatures. Moreover, affiliation to mixed subgroup went along with a significantly shorter overall survival for ccRCC and a longer overall survival for chRCC patients. Further research could offer a therapy stratification by specifically addressing the mitochondrial metabolism of such tumors and its microenvironment. KW - kidney cancer KW - pan-RCC KW - machine learning KW - mitochondrial DNA KW - mtDNA KW - mTOR Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-232107 SN - 2234-943X VL - 11 ER - TY - JOUR A1 - Marquardt, André A1 - Landwehr, Laura-Sophie A1 - Ronchi, Cristina L. A1 - di Dalmazi, Guido A1 - Riester, Anna A1 - Kollmannsberger, Philip A1 - Altieri, Barbara A1 - Fassnacht, Martin A1 - Sbiera, Silviu T1 - Identifying New Potential Biomarkers in Adrenocortical Tumors Based on mRNA Expression Data Using Machine Learning JF - Cancers N2 - Simple Summary Using a visual-based clustering method on the TCGA RNA sequencing data of a large adrenocortical carcinoma (ACC) cohort, we were able to classify these tumors in two distinct clusters largely overlapping with previously identified ones. As previously shown, the identified clusters also correlated with patient survival. Applying the visual clustering method to a second dataset also including benign adrenocortical samples additionally revealed that one of the ACC clusters is more closely located to the benign samples, providing a possible explanation for the better survival of this ACC cluster. Furthermore, the subsequent use of machine learning identified new possible biomarker genes with prognostic potential for this rare disease, that are significantly differentially expressed in the different survival clusters and should be further evaluated. Abstract Adrenocortical carcinoma (ACC) is a rare disease, associated with poor survival. Several “multiple-omics” studies characterizing ACC on a molecular level identified two different clusters correlating with patient survival (C1A and C1B). We here used the publicly available transcriptome data from the TCGA-ACC dataset (n = 79), applying machine learning (ML) methods to classify the ACC based on expression pattern in an unbiased manner. UMAP (uniform manifold approximation and projection)-based clustering resulted in two distinct groups, ACC-UMAP1 and ACC-UMAP2, that largely overlap with clusters C1B and C1A, respectively. However, subsequent use of random-forest-based learning revealed a set of new possible marker genes showing significant differential expression in the described clusters (e.g., SOAT1, EIF2A1). For validation purposes, we used a secondary dataset based on a previous study from our group, consisting of 4 normal adrenal glands and 52 benign and 7 malignant tumor samples. The results largely confirmed those obtained for the TCGA-ACC cohort. In addition, the ENSAT dataset showed a correlation between benign adrenocortical tumors and the good prognosis ACC cluster ACC-UMAP1/C1B. In conclusion, the use of ML approaches re-identified and redefined known prognostic ACC subgroups. On the other hand, the subsequent use of random-forest-based learning identified new possible prognostic marker genes for ACC. KW - adrenocortical carcinoma KW - in silico analysis KW - machine learning KW - bioinformatic clustering KW - biomarker prediction Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-246245 SN - 2072-6694 VL - 13 IS - 18 ER - TY - JOUR A1 - Maron, Roman C. A1 - Haggenmüller, Sarah A1 - von Kalle, Christof A1 - Utikal, Jochen S. A1 - Meier, Friedegund A1 - Gellrich, Frank F. A1 - Hauschild, Axel A1 - French, Lars E. A1 - Schlaak, Max A1 - Ghoreschi, Kamran A1 - Kutzner, Heinz A1 - Heppt, Markus V. A1 - Haferkamp, Sebastian A1 - Sondermann, Wiebke A1 - Schadendorf, Dirk A1 - Schilling, Bastian A1 - Hekler, Achim A1 - Krieghoff-Henning, Eva A1 - Kather, Jakob N. A1 - Fröhling, Stefan A1 - Lipka, Daniel B. A1 - Brinker, Titus J. T1 - Robustness of convolutional neural networks in recognition of pigmented skin lesions JF - European Journal of Cancer N2 - Background A basic requirement for artificial intelligence (AI)–based image analysis systems, which are to be integrated into clinical practice, is a high robustness. Minor changes in how those images are acquired, for example, during routine skin cancer screening, should not change the diagnosis of such assistance systems. Objective To quantify to what extent minor image perturbations affect the convolutional neural network (CNN)–mediated skin lesion classification and to evaluate three possible solutions for this problem (additional data augmentation, test-time augmentation, anti-aliasing). Methods We trained three commonly used CNN architectures to differentiate between dermoscopic melanoma and nevus images. Subsequently, their performance and susceptibility to minor changes (‘brittleness’) was tested on two distinct test sets with multiple images per lesion. For the first set, image changes, such as rotations or zooms, were generated artificially. The second set contained natural changes that stemmed from multiple photographs taken of the same lesions. Results All architectures exhibited brittleness on the artificial and natural test set. The three reviewed methods were able to decrease brittleness to varying degrees while still maintaining performance. The observed improvement was greater for the artificial than for the natural test set, where enhancements were minor. Conclusions Minor image changes, relatively inconspicuous for humans, can have an effect on the robustness of CNNs differentiating skin lesions. By the methods tested here, this effect can be reduced, but not fully eliminated. Thus, further research to sustain the performance of AI classifiers is needed to facilitate the translation of such systems into the clinic. KW - artificial intelligence KW - machine learning KW - deep learning KW - neural networks KW - dermatology KW - skin neoplasms KW - melanoma KW - nevus Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-370245 VL - 145 ER -