004 Datenverarbeitung; Informatik
Refine
Year of publication
Document Type
- Article (43)
- Preprint (10)
- Doctoral Thesis (1)
Has Fulltext
- yes (54)
Is part of the Bibliography
- no (54)
Keywords
- data science (9)
- artificial intelligence (7)
- machine-learning (6)
- digital medicine (5)
- machine learning (4)
- Data science (3)
- Machine learning (3)
- information theory (3)
- patients (3)
- Biomedical informatics (2)
Institute
- Medizin (54) (remove)
Background: Prostate cancer is a major health concern in aging men. Paralleling an aging society, prostate cancer prevalence increases emphasizing the need for efcient diagnostic algorithms.
Methods: Retrospectively, 106 prostate tissue samples from 48 patients (mean age,
66 ± 6.6 years) were included in the study. Patients sufered from prostate cancer (n = 38) or benign prostatic hyperplasia (n = 10) and were treated with radical prostatectomy or Holmium laser enucleation of the prostate, respectively. We constructed tissue microarrays (TMAs) comprising representative malignant (n = 38) and benign (n = 68) tissue cores. TMAs were processed to histological slides, stained, digitized and assessed for the applicability of machine learning strategies and open–source tools in diagnosis of prostate cancer. We applied the software QuPath to extract features for shape, stain intensity, and texture of TMA cores for three stainings, H&E, ERG, and PIN-4. Three machine learning algorithms, neural network (NN), support vector machines (SVM), and random forest (RF), were trained and cross-validated with 100 Monte Carlo random splits into 70% training set and 30% test set. We determined AUC values for single color channels, with and without optimization of hyperparameters by exhaustive grid search. We applied recursive feature elimination to feature sets of multiple color transforms.
Results: Mean AUC was above 0.80. PIN-4 stainings yielded higher AUC than H&E and
ERG. For PIN-4 with the color transform saturation, NN, RF, and SVM revealed AUC of 0.93 ± 0.04, 0.91 ± 0.06, and 0.92 ± 0.05, respectively. Optimization of hyperparameters improved the AUC only slightly by 0.01. For H&E, feature selection resulted in no increase of AUC but to an increase of 0.02–0.06 for ERG and PIN-4.
Conclusions: Automated pipelines may be able to discriminate with high accuracy between malignant and benign tissue. We found PIN-4 staining best suited for classifcation. Further bioinformatic analysis of larger data sets would be crucial to evaluate the reliability of automated classifcation methods for clinical practice and to evaluate potential discrimination of aggressiveness of cancer to pave the way to automatic precision medicine.
The human immune system is determined by the functionality of the human lymph node. With the use of high-throughput techniques in clinical diagnostics, a large number of data is currently collected. The new data on the spatiotemporal organization of cells offers new possibilities to build a mathematical model of the human lymph node - a virtual lymph node. The virtual lymph node can be applied to simulate drug responses and may be used in clinical diagnosis. Here, we review mathematical models of the human lymph node from the viewpoint of cellular processes. Starting with classical methods, such as systems of differential equations, we discuss the values of different levels of abstraction and methods in the range from artificial intelligence techniques formalism.
Background: Patients with rare diseases (RDs) are often diagnosed too late or not at all. Clinical decision support systems (CDSSs) could support the diagnosis in RDs. The MIRACUM (Medical Informatics in Research and Medicine) consortium, which is one of four funded consortia in the German Medical Informatics Initiative, will develop a CDSS for RDs based on distributed clinical data from ten university hospitals. This qualitative study aims to investigate (1) the relevant organizational conditions for the operation of a CDSS for RDs when diagnose patients (e.g. the diagnosis workflow), (2) which data is necessary for decision support, and (3) the appropriate user group for such a CDSS.
Methods: Interviews were carried out with RDs experts. Participants were recruited from staff physicians at the Rare Disease Centers (RDCs) at the MIRACUM locations, which offer diagnosis and treatment of RDs.
An interview guide was developed with a category-guided deductive approach. The interviews were recorded on an audio device and then transcribed into written form. We continued data collection until all interviews were completed. Afterwards, data analysis was performed using Mayring’s qualitative content analysis approach.
Results: A total of seven experts were included in the study. The results show that medical center guides and physicians from RDC B-centers (with a focus on different RDs) are involved in the diagnostic process. Furthermore, interdisciplinary case discussions between physicians are conducted.
The experts explained that RDs exist which cannot be fully differentiated, but rather described only by their overall symptoms or findings: diagnosis is dependent on the disease or disease group. At the end of the diagnostic process, most centers prepare a summary of the patient case. Furthermore, the experts considered both physicians and experts from the B-centers to be potential users of a CDSS. The experts also have different experiences with CDSS for RDs.
Conclusions: This qualitative study is a first step towards establishing the requirements for the development of a CDSS for RDs. Further research is necessary to create solutions by also including the experts on RDs.
For medicine to fulfill its promise of personalized treatments based on a better understanding of disease biology, computational and statistical tools must exist to analyze the increasing amount of patient data that becomes available. A particular challenge is that several types of data are being measured to cope with the complexity of the underlying systems, enhance predictive modeling and enrich molecular understanding.
Here we review a number of recent approaches that specialize in the analysis of multimodal data in the context of predictive biomedicine. We focus on methods that combine different OMIC measurements with image or genome variation data. Our overview shows the diversity of methods that address analysis challenges and reveals new avenues for novel developments.
Background: Rare Diseases (RDs), which are defined as diseases affecting no more than 5 out of 10,000 people, are often severe, chronic and life-threatening. A main problem is the delay in diagnosing RDs. Clinical decision support systems (CDSSs) for RDs are software systems to support clinicians in the diagnosis of patients with RDs. Due to their clinical importance, we conducted a scoping review to determine which CDSSs are available to support the diagnosis of RDs patients, whether the CDSSs are available to be used by clinicians and which functionalities and data are used to provide decision support.
Methods: We searched PubMed for CDSSs in RDs published between December 16, 2008 and December 16, 2018. Only English articles, original peer reviewed journals and conference papers describing a clinical prototype or a routine use of CDSSs were included. For data charting, we used the data items “Objective and background of the publication/project”, “System or project name”, “Functionality”, “Type of clinical data”, “Rare Diseases covered”, “Development status”, “System availability”, “Data entry and integration”, “Last software update” and “Clinical usage”.
Results: The search identified 636 articles. After title and abstracting screening, as well as assessing the eligibility criteria for full-text screening, 22 articles describing 19 different CDSSs were identified. Three types of CDSSs were classified: “Analysis or comparison of genetic and phenotypic data,” “machine learning” and “information retrieval”. Twelve of nineteen CDSSs use phenotypic and genetic data, followed by clinical data, literature databases and patient questionnaires. Fourteen of nineteen CDSSs are fully developed systems and therefore publicly available. Data can be entered or uploaded manually in six CDSSs, whereas for four CDSSs no information for data integration was available. Only seven CDSSs allow further ways of data integration. thirteen CDSS do not provide information about clinical usage.
Conclusions: Different CDSS for various purposes are available, yet clinicians have to determine which is best for their patient. To allow a more precise usage, future research has to focus on CDSSs RDs data integration, clinical usage and updating clinical knowledge. It remains interesting which of the CDSSs will be used and maintained in the future.
Background: Clinical trial registries increase transparency in medical research by making information and results of planned, ongoing, and completed studies publicly available. However, the registration of clinical trials remains a time-consuming manual task complicated by the fact that the same studies often need to be registered in different registries with different data entry requirements and interfaces.
Objective: This study investigates how Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) may be used as a standardized format for exchanging and storing clinical trial records.
Methods: We designed and prototypically implemented an open-source central trial registry containing records from university hospitals, which are automatically exported and updated by local study management systems.
Results: We provided an architecture and implementation of a multisite clinical trials registry based on HL7 FHIR as a data storage and exchange format.
Conclusions: The results show that FHIR resources establish a harmonized view of study information from heterogeneous sources by enabling automated data exchange between trial centers and central study registries.
Background: Persistent pain in breast cancer survivors is common. Psychological and sleep-related factors modulate perception, interpretation and coping with pain and may contribute to the clinical phenotype. The present analysis pursued the hypothesis that breast cancer survivors form subgroups, based on psychological and sleep-related parameters that are relevant to the impact of pain on the patients’ life.
Methods: We analysed 337 women treated for breast cancer, in whom psychological and sleep-related parameters as well as parameters related to pain intensity and interference had been acquired. Data were analysed by using supervised and unsupervised machine-learning techniques (i) to detect patient subgroups based on the pattern of psychological or sleep-related parameters, (ii) to interpret the detected cluster structure and (iii) to relate this data structure to pain interference and impact on life.
Results: Artificial intelligence-based detection of data structure, implemented as self-organizing neuronal maps, identified two different clusters of patients. A smaller cluster (11.5% of the patients) had comparatively lower resilience, more depressive symptoms and lower extraversion than the other patients. In these patients, life-satisfaction, mood, and life in general were comparatively more impeded by persistent pain.
Conclusions: The results support the initial hypothesis that psychological and sleep-related parameter patterns are meaningful for subgrouping patients with respect to how persistent pain after breast cancer treatments interferes with their life. This indicates that management of pain should address more complex features than just pain intensity. Artificial intelligence is a useful tool in the identification of subgroups of patients based on psychological factors.
Gene therapy (GT) is becoming a realistic treatment option for patients with haemophilia. Outside clinical trials, the complexity and potential complications of GT will pose unprecedented challenges to haemophilia care centres.AIM: To explore the potential use of electronic tools to improve the delivery of GT under real-world conditions.METHODS: Considering the hub-and-spoke model, the GTH working group on GT considered the entire patient pathway and reached consensus on requirements for an integrative software tool to secure documenting and sharing information between treaters, pharmacies and patients.RESULTS: Six steps of the gene therapy process were identified, each requiring completion of the previous step as a prerequisite for entry. The responsibilities of GT dosing and follow-up treatment centres, read/write access rules, and the minimum data set were outlined. Data contributed by patients through mobile devices was also considered.CONCLUSION: Important information needs to be shared between patients and treatment centres in a real-world GT hub-and-spoke model. Collecting and sharing this information in well-organised electronic applications will not only improve patient care but also enable national and international data collection in clinical registries...
Internalin B–mediated activation of the membrane-bound receptor tyrosine kinase MET is accompanied by a change in receptor mobility. Conversely, it should be possible to infer from receptor mobility whether a cell has been treated with internalin B. Here, we propose a method based on hidden Markov modeling and explainable artificial intelligence that machine-learns the key differences in MET mobility between internalin B–treated and –untreated cells from single-particle tracking data. Our method assigns receptor mobility to three diffusion modes (immobile, slow, and fast). It discriminates between internalin B–treated and –untreated cells with a balanced accuracy of >99% and identifies three parameters that are most affected by internalin B treatment: a decrease in the mobility of slow molecules (1) and a depopulation of the fast mode (2) caused by an increased transition of fast molecules to the slow mode (3). Our approach is based entirely on free software and is readily applicable to the analysis of other membrane receptors.
Recent scientific evidence suggests that chronic pain phenotypes are reflected in metabolomic changes. However, problems associated with chronic pain, such as sleep disorders or obesity, may complicate the metabolome pattern. Such a complex phenotype was investigated to identify common metabolomics markers at the interface of persistent pain, sleep, and obesity in 71 men and 122 women undergoing tertiary pain care. They were examined for patterns in d = 97 metabolomic markers that segregated patients with a relatively benign pain phenotype (low and little bothersome pain) from those with more severe clinical symptoms (high pain intensity, more bothersome pain, and co-occurring problems such as sleep disturbance). Two independent lines of data analysis were pursued. First, a data-driven supervised machine learning-based approach was used to identify the most informative metabolic markers for complex phenotype assignment. This pointed primarily at adenosine monophosphate (AMP), asparagine, deoxycytidine, glucuronic acid, and propionylcarnitine, and secondarily at cysteine and nicotinamide adenine dinucleotide (NAD) as informative for assigning patients to clinical pain phenotypes. After this, a hypothesis-driven analysis of metabolic pathways was performed, including sleep and obesity. In both the first and second line of analysis, three metabolic markers (NAD, AMP, and cysteine) were found to be relevant, including metabolic pathway analysis in obesity, associated with changes in amino acid metabolism, and sleep problems, associated with downregulated methionine metabolism. Taken together, present findings provide evidence that metabolomic changes associated with co-occurring problems may play a role in the development of severe pain. Co-occurring problems may influence each other at the metabolomic level. Because the methionine and glutathione metabolic pathways are physiologically linked, sleep problems appear to be associated with the first metabolic pathway, whereas obesity may be associated with the second.
Background: Persistent postsurgical neuropathic pain (PPSNP) can occur after intraoperative damage to somatosensory nerves, with a prevalence of 29–57% in breast cancer surgery. Proteomics is an active research field in neuropathic pain and the first results support its utility for establishing diagnoses or finding therapy strategies. Methods: 57 women (30 non-PPSNP/27 PPSNP) who had experienced a surgeon-verified intercostobrachial nerve injury during breast cancer surgery, were examined for patterns in 74 serum proteomic markers that allowed discrimination between subgroups with or without PPSNP. Serum samples were obtained both before and after surgery. Results: Unsupervised data analyses, including principal component analysis and self-organizing maps of artificial neurons, revealed patterns that supported a data structure consistent with pain-related subgroup (non-PPSPN vs. PPSNP) separation. Subsequent supervised machine learning-based analyses revealed 19 proteins (CD244, SIRT2, CCL28, CXCL9, CCL20, CCL3, IL.10RA, MCP.1, TRAIL, CCL25, IL10, uPA, CCL4, DNER, STAMPB, CCL23, CST5, CCL11, FGF.23) that were informative for subgroup separation. In cross-validated training and testing of six different machine-learned algorithms, subgroup assignment was significantly better than chance, whereas this was not possible when training the algorithms with randomly permuted data or with the protein markers not selected. In particular, sirtuin 2 emerged as a key protein, presenting both before and after breast cancer treatments in the PPSNP compared with the non-PPSNP subgroup. Conclusions: The identified proteins play important roles in immune processes such as cell migration, chemotaxis, and cytokine-signaling. They also have considerable overlap with currently known targets of approved or investigational drugs. Taken together, several lines of unsupervised and supervised analyses pointed to structures in serum proteomics data, obtained before and after breast cancer surgery, that relate to neuroinflammatory processes associated with the development of neuropathic pain after an intraoperative nerve lesion.
Motivation: Gaussian mixture models (GMMs) are probabilistic models commonly used in biomedical research to detect subgroup structures in data sets with one-dimensional information. Reliable model parameterization requires that the number of modes, i.e., states of the generating process, is known. However, this is rarely the case for empirically measured biomedical data. Several implementations are available that estimate GMM parameters differently. This work aims to provide a comparative evaluation of automated GMM fitting methods.
Results and conclusions: The performance of commonly used algorithms for automatic parameterization and mode number determination was compared with respect to reproducing the ground truth of generated data derived from multiple normal distributions. Four main variants of Gaussian mode number detection algorithms and five variants of GMM parameter estimation methods were tested in a combinatory scenario. The combination of best performing mode number determination algorithms and GMM parameter estimation methods was then tested on artificial and real-live data sets known to display a GMM structure. None of the tested methods correctly determined the underlying data structure consistently. The likelihood ratio test had the best performance in identifying the mode number associated with the best GMM fit of the data distribution while the Markov chain Monte Carlo (MCMC) algorithm was best for GMM parameter estimation while. The combination of the two methods of number determination algorithms and GMM parameter estimation was consistently among the best and overall outperformed the available implementations.
Implementation: An automated tool for the detection of GMM based structures in (biomedical) datasets was created based on the present results and made freely available in the R library “opGMMassessment” at https://cran.r-project.org/package=opGMMassessment.
Because it is associated with central nervous changes, and olfactory dysfunction has been reported with increased prevalence among persons with diabetes, this study addressed the question of whether the risk of developing diabetes in the next 10 years is reflected in olfactory symptoms. In a cross-sectional study, in 164 individuals seeking medical consulting for possible diabetes, olfactory function was evaluated using a standardized clinical test assessing olfactory threshold, odor discrimination, and odor identification. Metabolomics parameters were assessed via blood concentrations. The individual diabetes risk was quantified according to the validated German version of the “FINDRISK” diabetes risk score. Machine learning algorithms trained with metabolomics patterns predicted low or high diabetes risk with a balanced accuracy of 63–75%. Similarly, olfactory subtest results predicted the olfactory dysfunction category with a balanced accuracy of 85–94%, occasionally reaching 100%. However, olfactory subtest results failed to improve the prediction of diabetes risk based on metabolomics data, and metabolomics data did not improve the prediction of the olfactory dysfunction category based on olfactory subtest results. Results of the present study suggest that olfactory function is not a useful predictor of diabetes.
Background: The categorization of individuals as normosmic, hyposmic, or anosmic from test results of odor threshold, discrimination, and identification may provide a limited view of the sense of smell. The purpose of this study was to expand the clinical diagnostic repertoire by including additional tests. Methods: A random cohort of n = 135 individuals (83 women and 52 men, aged 21 to 94 years) was tested for odor threshold, discrimination, and identification, plus a distance test, in which the odor of peanut butter is perceived, a sorting task of odor dilutions for phenylethyl alcohol and eugenol, a discrimination test for odorant enantiomers, a lateralization test with eucalyptol, a threshold assessment after 10 min of exposure to phenylethyl alcohol, and a questionnaire on the importance of olfaction. Unsupervised methods were used to detect structure in the olfaction-related data, followed by supervised feature selection methods from statistics and machine learning to identify relevant variables. Results: The structure in the olfaction-related data divided the cohort into two distinct clusters with n = 80 and 55 subjects. Odor threshold, discrimination, and identification did not play a relevant role for cluster assignment, which, on the other hand, depended on performance in the two odor dilution sorting tasks, from which cluster assignment was possible with a median 100-fold cross-validated balanced accuracy of 77–88%. Conclusions: The addition of an odor sorting task with the two proposed odor dilutions to the odor test battery expands the phenotype of olfaction and fits seamlessly into the sensory focus of standard test batteries.
Recent advances in mathematical modelling and artificial intelligence have challenged the use of traditional regression analysis in biomedical research. This study examined artificial and cancer research data using binomial and multinomial logistic regression and compared its performance with other machine learning models such as random forests, support vector machines, Bayesian classifiers, k-nearest neighbours and repeated incremental clipping (RIPPER). The alternative models often outperformed regression in accurately classifying new cases. Logistic regression had a structural problem similar to early single-layer neural networks, which limited its ability to identify variables with high statistical significance for reliable class assignment. Therefore, regression is not always the best model for class prediction in biomedical datasets. The study emphasises the importance of validating selected models and suggests that a mixture of experts approach may be a more advanced and effective strategy for analysing biomedical datasets.
Selecting the k best features is a common task in machine learning. Typically, a few features have high importance, but many have low importance (right-skewed distribution). This report proposes a numerically precise method to address this skewed feature importance distribution in order to reduce a feature set to the informative minimum of items. Computed ABC analysis (cABC) is an item categorization method that aims to identify the most important items by partitioning a set of non-negative numerical items into subsets "A", "B", and "C" such that subset "A" contains the "few important" items based on specific properties of ABC curves defined by their relationship to Lorenz curves. In its recursive form, the cABC analysis can be applied again to subset "A". A generic image dataset and three biomedical datasets (lipidomics and two genomics datasets) with a large number of variables were used to perform the experiments. The experimental results show that the recursive cABC analysis limits the dimensions of the data projection to a minimum where the relevant information is still preserved and directs the feature selection in machine learning to the most important class-relevant information, including filtering feature sets for nonsense variables. Feature sets were reduced to 10% or less of the original variables and still provided accurate classification in data not used for feature selection. cABC analysis, in its recursive variant, provides a computationally precise means of reducing information to a minimum. The minimum is the result of a computation of the number of k most relevant items, rather than a decision to select the k best items from a list. In addition, there are precise criteria for stopping the reduction process. The reduction to the most important features can improve the human understanding of the properties of the data set. The cABC method is implemented in the Python package "cABCanalysis" available at https://pypi.org/project/cABCanalysis/.
Feature selection is a common step in data preprocessing that precedes machine learning to reduce data space and the computational cost of processing or obtaining the data. Filtering out uninformative variables is also important for knowledge discovery. By reducing the data space to only those components that are informative to the class structure, feature selection can simplify models so that they can be more easily interpreted by researchers in the field, reminiscent of explainable artificial intelligence. Knowledge discovery in complex data thus benefits from feature selection that aims to understand feature sets in the thematic context from which the data set originates. However, a single variable selected from a very small number of variables that are technically sufficient for AI training may make little immediate thematic sense, whereas the additional consideration of a variable discarded during feature selection could make scientific discovery very explicit. In this report, we propose an approach to explainable feature selection (XFS) based on a systematic reconsideration of unselected features. The difference between the respective classifications when training the algorithms with the selected features or with the unselected features provides a valid estimate of whether the relevant features in a data set have been selected and uninformative or trivial information was filtered out. It is shown that revisiting originally unselected variables in multivariate data sets allows for the detection of pathologies and errors in the feature selection that occasionally resulted in the failure to identify the most appropriate variables.
The use of artificial intelligence (AI) systems in biomedical and clinical settings can disrupt the traditional doctor–patient relationship, which is based on trust and transparency in medical advice and therapeutic decisions. When the diagnosis or selection of a therapy is no longer made solely by the physician, but to a significant extent by a machine using algorithms, decisions become nontransparent. Skill learning is the most common application of machine learning algorithms in clinical decision making. These are a class of very general algorithms (artificial neural networks, classifiers, etc.), which are tuned based on examples to optimize the classification of new, unseen cases. It is pointless to ask for an explanation for a decision. A detailed understanding of the mathematical details of an AI algorithm may be possible for experts in statistics or computer science. However, when it comes to the fate of human beings, this “developer’s explanation” is not sufficient. The concept of explainable AI (XAI) as a solution to this problem is attracting increasing scientific and regulatory interest. This review focuses on the requirement that XAIs must be able to explain in detail the decisions made by the AI to the experts in the field.
Sex differences in pain perception have been extensively studied, but precision medicine applications such as sex-specific pain pharmacology have barely progressed beyond proof-of-concept. A data set of pain thresholds to mechanical (blunt and punctate pressure) and thermal (heat and cold) stimuli applied to non-sensitized and sensitized (capsaicin, menthol) forearm skin of 69 male and 56 female healthy volunteers was analyzed for data structures contingent with the prior sex structure using unsupervised and supervised approaches. A working hypothesis that the relevance of sex differences could be approached via reversibility of the association, i.e., sex should be identifiable from pain thresholds, was verified with trained machine learning algorithms that could infer a person's sex in a 20% validation sample not seen to the algorithms during training, with balanced accuracy of up to 79%. This was only possible with thresholds for mechanical stimuli, but not for thermal stimuli or sensitization responses, which were not sufficient to train an algorithm that could assign sex better than by guessing or when trained with nonsense (permuted) information. This enabled the translation to the molecular level of nociceptive targets that convert mechanical but not thermal information into signals interpreted as pain, which could eventually be used for pharmacological precision medicine approaches to pain. By exploiting a key feature of machine learning, which allows for the recognition of data structures and the reduction of information to the minimum relevant, experimental human pain data could be characterized in a way that incorporates "non" logic that could be translated directly to the molecular pharmacological level, pointing toward sex-specific precision medicine for pain.