Refine
Has Fulltext
- yes (45)
Is part of the Bibliography
- yes (45)
Document Type
- Journal article (43)
- Doctoral Thesis (2)
Keywords
- machine learning (45) (remove)
Institute
- Institut für Geographie und Geologie (10)
- Institut für Informatik (8)
- Center for Computational and Theoretical Biology (5)
- Institut für Klinische Epidemiologie und Biometrie (4)
- Pathologisches Institut (4)
- Theodor-Boveri-Institut für Biowissenschaften (4)
- Betriebswirtschaftliches Institut (3)
- Klinik und Poliklinik für Nuklearmedizin (3)
- Medizinische Klinik und Poliklinik I (3)
- Medizinische Klinik und Poliklinik II (3)
Sonstige beteiligte Institutionen
Purpose
Machine learning based on radiomics features has seen huge success in a variety of clinical applications. However, the need for standardization and reproducibility has been increasingly recognized as a necessary step for future clinical translation. We developed a novel, intuitive open-source framework to facilitate all data analysis steps of a radiomics workflow in an easy and reproducible manner and evaluated it by reproducing classification results in eight available open-source datasets from different clinical entities.
Methods
The framework performs image preprocessing, feature extraction, feature selection, modeling, and model evaluation, and can automatically choose the optimal parameters for a given task. All analysis steps can be reproduced with a web application, which offers an interactive user interface and does not require programming skills. We evaluated our method in seven different clinical applications using eight public datasets: six datasets from the recently published WORC database, and two prostate MRI datasets—Prostate MRI and Ultrasound With Pathology and Coordinates of Tracked Biopsy (Prostate-UCLA) and PROSTATEx.
Results
In the analyzed datasets, AutoRadiomics successfully created and optimized models using radiomics features. For WORC datasets, we achieved AUCs ranging from 0.56 for lung melanoma metastases detection to 0.93 for liposarcoma detection and thereby managed to replicate the previously reported results. No significant overfitting between training and test sets was observed. For the prostate cancer detection task, results were better in the PROSTATEx dataset (AUC = 0.73 for prostate and 0.72 for lesion mask) than in the Prostate-UCLA dataset (AUC 0.61 for prostate and 0.65 for lesion mask), with external validation results varying from AUC = 0.51 to AUC = 0.77.
Conclusion
AutoRadiomics is a robust tool for radiomic studies, which can be used as a comprehensive solution, one of the analysis steps, or an exploratory tool. Its wide applicability was confirmed by the results obtained in the diverse analyzed datasets. The framework, as well as code for this analysis, are publicly available under https://github.com/pwoznicki/AutoRadiomics.
Deep convolutional generative adversarial networks (GAN) allow for creating images from existing databases. We applied a modified light-weight GAN (FastGAN) algorithm to cerebral blood flow SPECTs and aimed to evaluate whether this technology can generate created images close to real patients. Investigating three anatomical levels (cerebellum, CER; basal ganglia, BG; cortex, COR), 551 normal (248 CER, 174 BG, 129 COR) and 387 pathological brain SPECTs using N-isopropyl p-I-123-iodoamphetamine (123I-IMP) were included. For the latter scans, cerebral ischemic disease comprised 291 uni- (66 CER, 116 BG, 109 COR) and 96 bilateral defect patterns (44 BG, 52 COR). Our model was trained using a three-compartment anatomical input (dataset ‘A’; including CER, BG, and COR), while for dataset ‘B’, only one anatomical region (COR) was included. Quantitative analyses provided mean counts (MC) and left/right (LR) hemisphere ratios, which were then compared to quantification from real images. For MC, ‘B’ was significantly different for normal and bilateral defect patterns (P < 0.0001, respectively), but not for unilateral ischemia (P = 0.77). Comparable results were recorded for LR, as normal and ischemia scans were significantly different relative to images acquired from real patients (P ≤ 0.01, respectively). Images provided by ‘A’, however, revealed comparable quantitative results when compared to real images, including normal (P = 0.8) and pathological scans (unilateral, P = 0.99; bilateral, P = 0.68) for MC. For LR, only uni- (P = 0.03), but not normal or bilateral defect scans (P ≥ 0.08) reached significance relative to images of real patients. With a minimum of only three anatomical compartments serving as stimuli, created cerebral SPECTs are indistinguishable to images from real patients. The applied FastGAN algorithm may allow to provide sufficient scan numbers in various clinical scenarios, e.g., for “data-hungry” deep learning technologies or in the context of orphan diseases.
Grünflächen stellen einen der wichtigsten Umwelteinflüsse in der Wohnumwelt der Menschen dar. Einerseits wirken sie sich positiv auf die physische und mentale Gesundheit der Menschen aus, andererseits können Grünflächen auch negative Wirkungen anderer Faktoren abmildern, wie beispielsweise die im Laufe des Klimawandels zunehmenden Hitzeereignisse. Dennoch sind Grünflächen nicht für die gesamte Bevölkerung gleichermaßen zugänglich. Bestehende Forschung im Kontext der Umweltgerechtigkeit (UG) konnte bereits aufzeigen, dass unterschiedliche sozio-ökonomische und demographische Gruppen der deutschen Bevölkerung unterschiedlichen Zugriff auf Grünflächen haben. An bestehenden Analysen von Umwelteinflüssen im Kontext der UG wird kritisiert, dass die Auswertung geographischer Daten häufig auf zu stark aggregiertem Level geschieht, wodurch lokal spezifische Expositionen nicht mehr genau abgebildet werden. Dies trifft insbesondere für großflächig angelegte Studien zu. So werden wichtige räumliche Informationen verloren. Doch moderne Erdbeobachtungs- und Geodaten sind so detailliert wie nie und Methoden des maschinellen Lernens ermöglichen die effiziente Verarbeitung zur Ableitung höherwertiger Informationen.
Das übergeordnete Ziel dieser Arbeit besteht darin, am Beispiel von Grünflächen in Deutschland methodische Schritte der systematischen Umwandlung umfassender Geodaten in relevante Geoinformationen für die großflächige und hochaufgelöste Analyse von Umwelteigenschaften aufzuzeigen und durchzuführen. An der Schnittstelle der Disziplinen Fernerkundung, Geoinformatik, Sozialgeographie und Umweltgerechtigkeitsforschung sollen Potenziale moderner Methoden für die Verbesserung der räumlichen und semantischen Auflösung von Geoinformationen erforscht werden. Hierfür werden Methoden des maschinellen Lernens eingesetzt, um Landbedeckung und -nutzung auf nationaler Ebene zu erfassen. Diese Entwicklungen sollen dazu beitragen bestehende Datenlücken zu schließen und Aufschluss über die Verteilungsgerechtigkeit von Grünflächen zu bieten.
Diese Dissertation gliedert sich in drei konzeptionelle Teilschritte. Im ersten Studienteil werden Erdbeobachtungsdaten der Sentinel-2 Satelliten zur deutschlandweiten Klassifikation von Landbedeckungsinformationen verwendet. In Kombination mit punktuellen Referenzdaten der europaweiten Erfassung für Landbedeckungs- und Landnutzungsinformationen des Land Use and Coverage Area Frame Survey (LUCAS) wird ein maschinelles Lernverfahren trainiert. In diesem Kontext werden verschiedene Vorverarbeitungsschritte der LUCAS-Daten und deren Einfluss auf die Klassifikationsgenauigkeit beleuchtet. Das Klassifikationsverfahren ist in der Lage Landbedeckungsinformationen auch in komplexen urbanen Gebieten mit hoher Genauigkeit abzuleiten. Ein Ergebnis des Studienteils ist eine deutschlandweite Landbedeckungsklassifikation mit einer Gesamtgenauigkeit von 93,07 %, welche im weiteren Verlauf der Arbeit genutzt wird, um grüne Landbedeckung (GLC) räumlich zu quantifizieren.
Im zweiten konzeptionellen Teil der Arbeit steht die differenzierte Betrachtung von Grünflächen anhand des Beispiels öffentlicher Grünflächen (PGS), die häufig Gegenstand der UG-Forschung ist, im Vordergrund. Doch eine häufig verwendete Quelle für räumliche Daten zu öffentlichen Grünflächen, der European Urban Atlas (EUA), wird bisher nicht flächendeckend für Deutschland erhoben. Dieser Studienteil verfolgt einen datengetriebenen Ansatz, die Verfügbarkeit von öffentlichem Grün auf der räumlichen Ebene von Nachbarschaften für ganz Deutschland zu ermitteln. Hierfür dienen bereits vom EUA erfasste Gebiete als Referenz. Mithilfe einer Kombination von Erdbeobachtungsdaten und Informationen aus dem OpenStreetMap-Projekt wird ein Deep Learning -basiertes Fusionsnetzwerk erstellt, welche die verfügbare Fläche von öffentlichem Grün quantifiziert. Das Ergebnis dieses Schrittes ist ein Modell, welches genutzt wird, um die Menge öffentlicher Grünflächen in der Nachbarschaft zu schätzen (𝑅 2 = 0.952).
Der dritte Studienteil greift die Ergebnisse der ersten beiden Studienteile auf und betrachtet die Verteilung von Grünflächen in Deutschland unter Hinzunahme von georeferenzierten Bevölkerungsdaten. Diese exemplarische Analyse unterscheidet dabei Grünflächen nach zwei Typen: GLC und PGS. Zunächst wird mithilfe deskriptiver Statistiken die generelle Grünflächenverteilung in der Bevölkerung Deutschlands beleuchtet. Daraufhin wird die Verteilungsgerechtigkeit anhand gängiger Gerechtigkeitsmetriken bestimmt. Abschließend werden die Zusammenhänge zwischen der demographischen Komposition der Nachbarschaft und der verfügbaren Menge von Grünflächen anhand dreier exemplarischer soziodemographischer Gesellschaftsgruppen untersucht. Die Analyse zeigt starke Unterschiede der Verfügbarkeit von PGS zwischen städtischen und ländlichen Gebieten. Ein höherer Prozentsatz der Stadtbevölkerung hat Zugriff das Mindestmaß von PGS gemessen an der Vorgabe der Weltgesundheitsorganisation. Die Ergebnisse zeigen auch einen deutlichen Unterschied bezüglich der Verteilungsgerechtigkeit zwischen GLC und PGS und verdeutlichen die Relevanz der Unterscheidung von Grünflächentypen für derartige
Untersuchungen. Die abschließende Betrachtung verschiedener Bevölkerungsgruppen arbeitet Unterschiede auf soziodemographischer Ebene auf.
In der Zusammenschau demonstriert diese Arbeit wie moderne Geodaten und Methoden des maschinellen Lernens genutzt werden können bisherige Limitierungen räumlicher Datensätze zu überwinden. Am Beispiel von Grünflächen in der Wohnumgebung der Bevölkerung Deutschlands wird gezeigt, dass landesweite Analysen zur Umweltgerechtigkeit durch hochaufgelöste und lokal feingliedrige geographische Informationen bereichert werden können. Diese Arbeit verdeutlicht, wie die Methoden der Erdbeobachtung und Geoinformatik einen wichtigen Beitrag leisten können, die Ungleichheit der Wohnumwelt der Menschen zu identifizieren und schlussendlich den nachhaltigen Siedlungsbau in Form von objektiven Informationen zu unterstützen und überwachen.
In the past decades, various Earth observation-based time series products have emerged, which have enabled studies and analysis of global change processes. Besides their contribution to understanding past processes, time series datasets hold enormous potential for predictive modeling and thereby meet the demands of decision makers on future scenarios. In order to further exploit these data, a novel pixel-based approach has been introduced, which is the spatio-temporal matrix (STM). The approach integrates the historical characteristics of a specific land cover at a high temporal frequency in order to interpret the spatial and temporal information for the neighborhood of a given target pixel. The provided information can be exploited with common predictive models and algorithms. In this study, this approach was utilized and evaluated for the prediction of future urban/built-settlement growth. Random forest and multi-layer perceptron were employed for the prediction. The tests have been carried out with training strategies based on a one-year and a ten-year time span for the urban agglomerations of Surat (India), Ho-Chi-Minh City (Vietnam), and Abidjan (Ivory Coast). The slope, land use, exclusion, urban, transportation, hillshade (SLEUTH) model was selected as a baseline indicator for the performance evaluation. The statistical results from the receiver operating characteristic curve (ROC) demonstrate a good ability of the STM to facilitate the prediction of future settlement growth and its transferability to different cities, with area under the curve (AUC) values greater than 0.85. Compared with SLEUTH, the STM-based model achieved higher AUC in all of the test cases, while being independent of the additional datasets for the restricted and the preferential development areas.
Associations between periodontitis and COPD: An artificial intelligence-based analysis of NHANES III
(2022)
A number of cross-sectional epidemiological studies suggest that poor oral health is associated with respiratory diseases. However, the number of cases within the studies was limited, and the studies had different measurement conditions. By analyzing data from the National Health and Nutrition Examination Survey III (NHANES III), this study aimed to investigate possible associations between chronic obstructive pulmonary disease (COPD) and periodontitis in the general population. COPD was diagnosed in cases where FEV (1)/FVC ratio was below 70% (non-COPD versus COPD; binary classification task). We used unsupervised learning utilizing k-means clustering to identify clusters in the data. COPD classes were predicted with logistic regression, a random forest classifier, a stochastic gradient descent (SGD) classifier, k-nearest neighbors, a decision tree classifier, Gaussian naive Bayes (GaussianNB), support vector machines (SVM), a custom-made convolutional neural network (CNN), a multilayer perceptron artificial neural network (MLP), and a radial basis function neural network (RBNN) in Python. We calculated the accuracy of the prediction and the area under the curve (AUC). The most important predictors were determined using feature importance analysis. Results: Overall, 15,868 participants and 19 feature variables were included. Based on k-means clustering, the data were separated into two clusters that identified two risk characteristic groups of patients. The algorithms reached AUCs between 0.608 (DTC) and 0.953% (CNN) for the classification of COPD classes. Feature importance analysis of deep learning algorithms indicated that age and mean attachment loss were the most important features in predicting COPD. Conclusions: Data analysis of a large population showed that machine learning and deep learning algorithms could predict COPD cases based on demographics and oral health feature variables. This study indicates that periodontitis might be an important predictor of COPD. Further prospective studies examining the association between periodontitis and COPD are warranted to validate the present results.
Background: Oro-antral communication (OAC) is a common complication following the extraction of upper molar teeth. The Archer and the Root Sinus (RS) systems can be used to classify impacted teeth in panoramic radiographs. The Archer classes B-D and the Root Sinus classes III, IV have been associated with an increased risk of OAC following tooth extraction in the upper molar region. In our previous study, we found that panoramic radiographs are not reliable for predicting OAC. This study aimed to (1) determine the feasibility of automating the classification (Archer/RS classes) of impacted teeth from panoramic radiographs, (2) determine the distribution of OAC stratified by classification system classes for the purposes of decision tree construction, and (3) determine the feasibility of automating the prediction of OAC utilizing the mentioned classification systems. Methods: We utilized multiple supervised pre-trained machine learning models (VGG16, ResNet50, Inceptionv3, EfficientNet, MobileNetV2), one custom-made convolutional neural network (CNN) model, and a Bag of Visual Words (BoVW) technique to evaluate the performance to predict the clinical classification systems RS and Archer from panoramic radiographs (Aim 1). We then used Chi-square Automatic Interaction Detectors (CHAID) to determine the distribution of OAC stratified by the Archer/RS classes to introduce a decision tree for simple use in clinics (Aim 2). Lastly, we tested the ability of a multilayer perceptron artificial neural network (MLP) and a radial basis function neural network (RBNN) to predict OAC based on the high-risk classes RS III, IV, and Archer B-D (Aim 3). Results: We achieved accuracies of up to 0.771 for EfficientNet and MobileNetV2 when examining the Archer classification. For the AUC, we obtained values of up to 0.902 for our custom-made CNN. In comparison, the detection of the RS classification achieved accuracies of up to 0.792 for the BoVW and an AUC of up to 0.716 for our custom-made CNN. Overall, the Archer classification was detected more reliably than the RS classification when considering all algorithms. CHAID predicted 77.4% correctness for the Archer classification and 81.4% for the RS classification. MLP (AUC: 0.590) and RBNN (AUC: 0.590) for the Archer classification as well as MLP 0.638) and RBNN (0.630) for the RS classification did not show sufficient predictive capability for OAC. Conclusions: The results reveal that impacted teeth can be classified using panoramic radiographs (best AUC: 0.902), and the classification systems can be stratified according to their relationship to OAC (81.4% correct for RS classification). However, the Archer and RS classes did not achieve satisfactory AUCs for predicting OAC (best AUC: 0.638). Additional research is needed to validate the results externally and to develop a reliable risk stratification tool based on the present findings.
The identification of biomarker signatures is important for cancer diagnosis and prognosis. However, the detection of clinical reliable signatures is influenced by limited data availability, which may restrict statistical power. Moreover, methods for integration of large sample cohorts and signature identification are limited. We present a step-by-step computational protocol for functional gene expression analysis and the identification of diagnostic and prognostic signatures by combining meta-analysis with machine learning and survival analysis. The novelty of the toolbox lies in its all-in-one functionality, generic design, and modularity. It is exemplified for lung cancer, including a comprehensive evaluation using different validation strategies. However, the protocol is not restricted to specific disease types and can therefore be used by a broad community. The accompanying R package vignette runs in ~1 h and describes the workflow in detail for use by researchers with limited bioinformatics training.
Objectives
Embedded in the Collaborative Research Center “Fear, Anxiety, Anxiety Disorders” (CRC‐TRR58), this bicentric clinical study aims at identifying biobehavioral markers of treatment (non‐)response by applying machine learning methodology with an external cross‐validation protocol. We hypothesize that a priori prediction of treatment (non‐)response is possible in a second, independent sample based on multimodal markers.
Methods
One‐session virtual reality exposure treatment (VRET) with patients with spider phobia was conducted on two sites. Clinical, neuroimaging, and genetic data were assessed at baseline, post‐treatment and after 6 months. The primary and secondary outcomes defining treatment response are as follows: 30% reduction regarding the individual score in the Spider Phobia Questionnaire and 50% reduction regarding the individual distance in the behavioral avoidance test.
Results
N = 204 patients have been included (n = 100 in Würzburg, n = 104 in Münster). Sample characteristics for both sites are comparable.
Discussion
This study will offer cross‐validated theranostic markers for predicting the individual success of exposure‐based therapy. Findings will support clinical decision‐making on personalized therapy, bridge the gap between basic and clinical research, and bring stratified therapy into reach. The study is registered at ClinicalTrials.gov (ID: NCT03208400).
Neural networks have to capture mathematical relationships in order to learn various tasks. They approximate these relations implicitly and therefore often do not generalize well. The recently proposed Neural Arithmetic Logic Unit (NALU) is a novel neural architecture which is able to explicitly represent the mathematical relationships by the units of the network to learn operations such as summation, subtraction or multiplication. Although NALUs have been shown to perform well on various downstream tasks, an in-depth analysis reveals practical shortcomings by design, such as the inability to multiply or divide negative input values or training stability issues for deeper networks. We address these issues and propose an improved model architecture. We evaluate our model empirically in various settings from learning basic arithmetic operations to more complex functions. Our experiments indicate that our model solves stability issues and outperforms the original NALU model in means of arithmetic precision and convergence.
An approach to aerodynamically optimizing cycling posture and reducing drag in an Ironman (IM) event was elaborated. Therefore, four commonly used positions in cycling were investigated and simulated for a flow velocity of 10 m/s and yaw angles of 0–20° using OpenFoam-based Nabla Flow CFD simulation software software. A cyclist was scanned using an IPhone 12, and a special-purpose meshing software BLENDER was used. Significant differences were observed by changing and optimizing the cyclist’s posture. Aerodynamic drag coefficient (CdA) varies by more than a factor of 2, ranging from 0.214 to 0.450. Within a position, the CdA tends to increase slightly at yaw angles of 5–10° and decrease at higher yaw angles compared to a straight head wind, except for the time trial (TT) position. The results were applied to the IM Hawaii bike course (180 km), estimating a constant power output of 300 W. Including the wind distributions, two different bike split models for performance prediction were applied. Significant time saving of roughly 1 h was found. Finally, a machine learning approach to deduce 3D triangulation for specific body shapes from 2D pictures was tested.
Predicting hypertension subtypes with machine learning using targeted metabolites and their ratios
(2022)
Hypertension is a major global health problem with high prevalence and complex associated health risks. Primary hypertension (PHT) is most common and the reasons behind primary hypertension are largely unknown. Endocrine hypertension (EHT) is another complex form of hypertension with an estimated prevalence varying from 3 to 20% depending on the population studied. It occurs due to underlying conditions associated with hormonal excess mainly related to adrenal tumours and sub-categorised: primary aldosteronism (PA), Cushing’s syndrome (CS), pheochromocytoma or functional paraganglioma (PPGL). Endocrine hypertension is often misdiagnosed as primary hypertension, causing delays in treatment for the underlying condition, reduced quality of life, and costly antihypertensive treatment that is often ineffective. This study systematically used targeted metabolomics and high-throughput machine learning methods to predict the key biomarkers in classifying and distinguishing the various subtypes of endocrine and primary hypertension. The trained models successfully classified CS from PHT and EHT from PHT with 92% specificity on the test set. The most prominent targeted metabolites and metabolite ratios for hypertension identification for different disease comparisons were C18:1, C18:2, and Orn/Arg. Sex was identified as an important feature in CS vs. PHT classification.
Background: Tinnitus is often described as the phantom perception of a sound and is experienced by 5.1% to 42.7% of the population worldwide, at least once during their lifetime. The symptoms often reduce the patient's quality of life. The TrackYourTinnitus (TYT) mobile health (mHealth) crowdsensing platform was developed for two operating systems (OS)-Android and iOS-to help patients demystify the daily moment-to-moment variations of their tinnitus symptoms. In all platforms developed for more than one OS, it is important to investigate whether the crowdsensed data predicts the OS that was used in order to understand the degree to which the OS is a confounder that is necessary to consider.
The prediction of breeding values and phenotypes is of central importance for both livestock and crop breeding. In this study, we analyze the use of artificial neural networks (ANN) and, in particular, local convolutional neural networks (LCNN) for genomic prediction, as a region-specific filter corresponds much better with our prior genetic knowledge on the genetic architecture of traits than traditional convolutional neural networks. Model performances are evaluated on a simulated maize data panel (n = 10,000; p = 34,595) and real Arabidopsis data (n = 2,039; p = 180,000) for a variety of traits based on their predictive ability. The baseline LCNN, containing one local convolutional layer (kernel size: 10) and two fully connected layers with 64 nodes each, is outperforming commonly proposed ANNs (multi layer perceptrons and convolutional neural networks) for basically all considered traits. For traits with high heritability and large training population as present in the simulated data, LCNN are even outperforming state-of-the-art methods like genomic best linear unbiased prediction (GBLUP), Bayesian models and extended GBLUP, indicated by an increase in predictive ability of up to 24%. However, for small training populations, these state-of-the-art methods outperform all considered ANNs. Nevertheless, the LCNN still outperforms all other considered ANNs by around 10%. Minor improvements to the tested baseline network architecture of the LCNN were obtained by increasing the kernel size and of reducing the stride, whereas the number of subsequent fully connected layers and their node sizes had neglectable impact. Although gains in predictive ability were obtained for large scale data sets by using LCNNs, the practical use of ANNs comes with additional problems, such as the need of genotyping all considered individuals, the lack of estimation of heritability and reliability. Furthermore, breeding values are additive by design, whereas ANN-based estimates are not. However, ANNs also comes with new opportunities, as networks can easily be extended to account for additional inputs (omics, weather etc.) and outputs (multi-trait models), and computing time increases linearly with the number of individuals. With advances in high-throughput phenotyping and cheaper genotyping, ANNs can become a valid alternative for genomic prediction.
This paper describes the estimation of the body weight of a person in front of an RGB-D camera. A survey of different methods for body weight estimation based on depth sensors is given. First, an estimation of people standing in front of a camera is presented. Second, an approach based on a stream of depth images is used to obtain the body weight of a person walking towards a sensor. The algorithm first extracts features from a point cloud and forwards them to an artificial neural network (ANN) to obtain an estimation of body weight. Besides the algorithm for the estimation, this paper further presents an open-access dataset based on measurements from a trauma room in a hospital as well as data from visitors of a public event. In total, the dataset contains 439 measurements. The article illustrates the efficiency of the approach with experiments with persons lying down in a hospital, standing persons, and walking persons. Applicable scenarios for the presented algorithm are body weight-related dosing of emergency patients.
Ever-growing data availability combined with rapid progress in analytics has laid the foundation for the emergence of business process analytics. Organizations strive to leverage predictive process analytics to obtain insights. However, current implementations are designed to deal with homogeneous data. Consequently, there is limited practical use in an organization with heterogeneous data sources. The paper proposes a method for predictive end-to-end enterprise process network monitoring leveraging multi-headed deep neural networks to overcome this limitation. A case study performed with a medium-sized German manufacturing company highlights the method’s utility for organizations.
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.
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.
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.
(1) Background: C-X-C Motif Chemokine Receptor 4 (CXCR4) and Fibroblast Activation Protein Alpha (FAP) are promising theranostic targets. However, it is unclear whether CXCR4 and FAP positivity mark distinct microenvironments, especially in solid tumors. (2) Methods: Using Random Forest (RF) analysis, we searched for entity-independent mRNA and microRNA signatures related to CXCR4 and FAP overexpression in our pan-cancer cohort from The Cancer Genome Atlas (TCGA) database — representing n = 9242 specimens from 29 tumor entities. CXCR4- and FAP-positive samples were assessed via StringDB cluster analysis, EnrichR, Metascape, and Gene Set Enrichment Analysis (GSEA). Findings were validated via correlation analyses in n = 1541 tumor samples. TIMER2.0 analyzed the association of CXCR4 / FAP expression and infiltration levels of immune-related cells. (3) Results: We identified entity-independent CXCR4 and FAP gene signatures representative for the majority of solid cancers. While CXCR4 positivity marked an immune-related microenvironment, FAP overexpression highlighted an angiogenesis-associated niche. TIMER2.0 analysis confirmed characteristic infiltration levels of CD8+ cells for CXCR4-positive tumors and endothelial cells for FAP-positive tumors. (4) Conclusions: CXCR4- and FAP-directed PET imaging could provide a non-invasive decision aid for entity-agnostic treatment of microenvironment in solid malignancies. Moreover, this machine learning workflow can easily be transferred towards other theranostic targets.
Uplink vs. Downlink: Machine Learning-Based Quality Prediction for HTTP Adaptive Video Streaming
(2021)
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.