Time-series forecasting through modelling sequences of temporally dependent observations has many industrial and scientific applications. While machine learning models have been widely used to... Show moreTime-series forecasting through modelling sequences of temporally dependent observations has many industrial and scientific applications. While machine learning models have been widely used to create time-series forecasting models, creating efficient and performant time-series forecasting models is a complex task for domain users. Automated Machine Learning (AutoML) is a growing field that aims to make the process of creating machine-learning models accessible for non-machine learning experts. This is achieved by optimising machine learning pipelines automatically. Time-series machine-learning pipelines include various specialised pre-processing steps that are not currently supported by existing AutoML systems. This dissertation investigates how AutoML can be extended to time-series data analysis problems such as time-series forecasting. Several challenges arise when developing specialised AutoML systems for time-series forecasting. For instance, advanced machine-learning pipelines that can extract time-series features and select well-suited machine-learning models need to be developed. Also, extra hyperparameters such as the window size, which shows how many historical data points are helpful, need to be optimised by the AutoML system. This dissertation addresses these issues. We provide a comprehensive overview of the AutoML research field, including hyperparameter optimisation techniques, neural architecture search, and existing AutoML systems. Next, we investigate the use of AutoML for short-term forecasting, single-step ahead time-series forecasting, and multi-step time-series forecasting with time-series features. Show less
AI-powered emotion recognition, typing with thoughts or eavesdropping virtual assistants: three non-fictional examples illustrate how AI may impact society. AI-related products and services... Show moreAI-powered emotion recognition, typing with thoughts or eavesdropping virtual assistants: three non-fictional examples illustrate how AI may impact society. AI-related products and services increasingly find their way into daily life. Are the EU's fundamental rights to privacy and data protection equipped to protect individuals effectively? In addressing this question, the dissertation concludes that no new legal framework is needed. Instead, adjustments are required. First, the extent of adjustments depends on the AI discipline. There is nothing like 'the AI'. AI covers various concepts, including the disciplines machine learning, natural language processing, computer vision, affective computing and automated reasoning. Second, the extent of adjustments depends on the type of legal problem: legal provisions are violated (type 1), cannot be enforced (type 2) or are not fit for purpose (type 3). Type 2 and 3 problems require either adjustments of current provisions or new judicial interpretations. Two instruments might be helpful for more effective legislation: rebuttable presumptions and reversal of proof. In some cases, the solution is technical, not legal. Research in AI should solve reasoning deficiencies in AI systems and their lack of common sense. Show less
This thesis looks at Artificial Intelligence (AI) and its potential to revolutionise the healthcare sector. The first part of this thesis focuses on the responsible development and validation of AI... Show moreThis thesis looks at Artificial Intelligence (AI) and its potential to revolutionise the healthcare sector. The first part of this thesis focuses on the responsible development and validation of AI-based clinical prediction algorithms, exploring the prime considerations in this process. The second part of this thesis addresses the opportunities for classical statistics and machine learning techniques for developing prediction algorithms. It also examines the performance, potential, and challenges of AI prediction algorithms for clinical practice. The conclusion states that cross-discipline collaboration, exchangeability of knowledge and results, and validation of AI for healthcare practice are essential for realising the potential of AI in healthcare. Show less
The societal burden of spinal conditions is vast and continues to grow with the in- creasing prevalence of patients with spinal degenerative disease, spinal metasta- ses, and spinal infections.... Show moreThe societal burden of spinal conditions is vast and continues to grow with the in- creasing prevalence of patients with spinal degenerative disease, spinal metasta- ses, and spinal infections. Recent application of artificial intelligence in healthcare have shown great promise and similar extensions in spine surgery may improve decision-making. The purpose of this thesis was to examine the utility of predictive analytics and natural language processing in spine surgery. Show less
The aim of this thesis is to determine diagnostic performance of machine learning in differentiating between atypical cartilaginous tumor (ACT) and high-grade chondrosarcoma (CS) based on radiomic... Show moreThe aim of this thesis is to determine diagnostic performance of machine learning in differentiating between atypical cartilaginous tumor (ACT) and high-grade chondrosarcoma (CS) based on radiomic features derived from magnetic resonance imaging (MRI) and computed tomography (CT). In chapter 2, the concept of radiomics of musculoskeletal sarcomas is introduced and a systematic review on radiomic feature reproducibility and validation strategies is conducted. In chapter 3, a preliminary study is performed to investigate the performance of MRI radiomics-based machine learning in discriminating ACT from high-grade CS, using a single-center cohort, in comparison with an expert radiologist. In chapter 4, the influence of interobserver segmentation variability on the reproducibility of CT and MRI radiomic features of cartilaginous bone tumors is assessed. In chapter 5, the performance of CT radiomics-based machine learning in discriminating ACT from high-grade CS of long bones is determined and validated using independent data from a multicenter cohort, compared to an expert radiologist. In chapter 6, the performance of MRI radiomics-based machine learning in differentiating between ACT and grade II CS of long bones is determined and validated using independent data from a multicenter cohort, in comparison with an expert radiologist. Finally, in chapter 7, the main results and implications of this thesis are summarized and discussed. Show less
Despite improved surgical and adjuvant treatment options, malignant brain tumors remain non-curable to date. The thin line between treatment effectiveness and patient harms underpins the importance... Show moreDespite improved surgical and adjuvant treatment options, malignant brain tumors remain non-curable to date. The thin line between treatment effectiveness and patient harms underpins the importance of tailoring clinical management to the individual brain tumor patient. Over the past decades, the volume and complexity of clinically-derived patient data (i.e., imaging, genomics, free-text etc.) is increasing exponentially. Machine learning provides a vast range of algorithms that can learn from this data and guide clinical decision-making by providing accurate patient-level predictions. The current thesis describes several studies along the continuum of the machine learning spectrum as it applies to neurosurgical oncology. Part I investigates postoperative complications and risk factors in patients operated for a primary malignant brain tumor. Part II describes de development of a model for the prediction of individual-patient survival in glioblastoma patients. Part III encompasses the development of a natural language processing framework for automated medical text analysis. Machine learning algorithms should be considered as an extension to statistical approaches and exist along a continuum determined by how much is specified by humans and how much is learnt by the machine. Although machine learning algorithms can produce highly accurate predictions based on high-dimensional data, clinicians and researchers should interpret the clinical implications of these predictions on case-by-case basis. Show less
Pharmacogenomics (PGx) is widely recognized as an important aspect in personalizedMedicine. By analyzing and interpreting one’s genetic profile dose and drug adjustmentscan be made. In this way,... Show morePharmacogenomics (PGx) is widely recognized as an important aspect in personalizedMedicine. By analyzing and interpreting one’s genetic profile dose and drug adjustmentscan be made. In this way, one can strive to improve the safety and efficacy of drugtreatments. Nonetheless, not all genetic variability in drug response can be explained withcurrent PGx. In this thesis we explore the role of additional genetic factors which can explain this missing heritability. Firstly, rare and novel variants which are unaccounted for in routine PGx panels might play a role. Secondly, the complexity of pharmacogenes can result in an inability tounravel the genetic make-up of these genes. Thirdly, haplotype phasing is generally nottaken into account in PGx. Fourthly, all genetic variants are currently summarized intoone of four metabolic categories: poor metabolizers (PM), intermediate metabolizers(IM), normal metabolizers (NM) (previously EM) and ultra-rapid metabolizers (UM).However, enzyme activity is not a matter of ‘on’ or ‘off ’, but is more of a continuous scale.Finally, the effect of a genetic variant on drug metabolism shows substrate specific effects.This substrate specificity can result in erroneous extrapolation of variant effects to theentire range of substrates. The development of novel technologies to determine one’sgenetic make-up is evolving rapidly, thereby providing opportunities for the field of PGxto address these issues. In this thesis we show that by using long-read sequencing or trio-based sequencing more information can be obtained which can lead to a better understanding of the (rare) variants and can help with haplotype phasing. Moreover, we have shown that by combining long-read sequencing with artificial intelligence a substantial increase in explained variability can be achieved. Show less
Inflammatory Bowel Diseases (IBD) such as Crohn’s disease (CD) and ulcerative colitis (UC) are chronic immunological digestive diseases with a progressive character and associated with significant... Show moreInflammatory Bowel Diseases (IBD) such as Crohn’s disease (CD) and ulcerative colitis (UC) are chronic immunological digestive diseases with a progressive character and associated with significant healthcare costs. Different solutions have been proposed such as innovation in care monitoring or implementation of electronic health (eHealth). IBD is one of many chronic diseases that could benefit from eHealth, adding smartphone applications to the toolbox for care management has the potential improve disease understanding, enhance medication adherence, improve patient-physician communications, and for earlier interventions by medical professionals when problems arise. Furthermore, the accessibility to Big Data and increased computational resources have paved the way for Artificial Intelligence (AI) to provide potential solutions for the management of prototypical complex diseases with advanced heterogeneity and alternating disease states, like IBD. In this thesis we assessed the current economic and psychosocial impact of IBD by assessing its effect on indirect costs, productivity and caregiving. Furthermore, we observed if we can proactively identify IBD patients’ needs using eHealth and Artificial Intelligence. Lastly, we analyze the impact of monitoring IBD patients using eHealth interventions in order to facilitate the delivery of high-value care. Show less
In this thesis we have studied the influence of emotion on learning. We have used computational modelling techniques to do so, more specifically, the reinforcement learning paradigm. Emotion is... Show moreIn this thesis we have studied the influence of emotion on learning. We have used computational modelling techniques to do so, more specifically, the reinforcement learning paradigm. Emotion is modelled as artificial affect, a measure that denotes the positiveness versus negativeness of a situation to an artificial agent in a reinforcement learning setting. We have done a range of different experiments to study the effect of affect on learning, including the effect on learning if affect is used to control the exploration behaviour of the agent and the effect on learning when affect is communicated by a human (though real-time analysis of that human__s facial expressions) to a simulated robot. We conclude that affect is a useful concept to consider in adaptive agents that learn based on reinforcement learning and that in some cases affect can indeed help the learning process. Further, affective modelling in this way can help understand the psychological processes that underlie influences of affect on cognition. Finally, we have developed a formal notation for a specific type of emotion theory, i.e., cognitive appraisal theory. Show less
In our modern society we rely on information and communication technology for the speed, efficiency, and security of many of our transactions and interactions. The use of these technologies almost... Show moreIn our modern society we rely on information and communication technology for the speed, efficiency, and security of many of our transactions and interactions. The use of these technologies almost always entails the keeping of electronic records. These are of interest to law enforcement and intelligence agencies since they can be used to profile and monitor (potential) suspects. While this may enhance the security of our society, it may also form a threat to privacy and liberty. Up until now, the sheer amount of data available in the information society has acted as a de facto safeguard for privacy and liberty. However, advances in the science of artificial intelligence, most notably in the area of software agent technology, are quickly removing the information overload. Software agents are intelligent programs able to perform tasks without human supervision. In this thesis the use of agents for surveillance purposes is examined. The goal of the thesis is to determine whether the legal framework for the protection of privacy and liberty in the Netherlands and the United States is still adequate in the light of agent-enabled surveillance. In particular, the thesis explores whether the right to privacy is still an adequate means of protecting liberty. Show less