Background and purpose: Automatic approaches are widely implemented to automate dose optimization in radiotherapy treatment planning. This study systematically investigates how to configure automatic planning in order to create the best possible plans. Materials and methods: Automatic plans were generated using protocol based automatic iterative optimization. Starting from a simple automation protocol which consisted of the constraints for targets and organs at risk (OAR), the performance of the automatic approach was evaluated in terms of target coverage, OAR sparing, conformity, beam complexity, and plan quality. More complex protocols were systematically explored to improve the quality of the automatic plans. The protocols could be improved by adding a dose goal on the outer 2 mm of the PTV, by setting goals on strategically chosen subparts of OARs, by adding goals for conformity, and by limiting the leaf motion. For prostate plans, development of an automated post-optimization procedure was required to achieve precise control over the dose distribution. Automatic and manually optimized plans were compared for 20 head and neck (H&N), 20 prostate, and 20 rectum cancer patients. Results: Based on simple automation protocols, the automatic optimizer was not always able to generate adequate treatment plans. For the improved final configurations for the three sites, the dose was lower in automatic plans compared to the manual plans in 12 out of 13 considered OARs. In blind tests, the automatic plans were preferred in 80% of cases. Conclusions: With adequate, advanced, protocols the automatic planning approach is able to create high-quality treatment plans.
DOCUMENT
The huge number of images shared on the Web makes effective cataloguing methods for efficient storage and retrieval procedures specifically tailored on the end-user needs a very demanding and crucial issue. In this paper, we investigate the applicability of Automatic Image Annotation (AIA) for image tagging with a focus on the needs of database expansion for a news broadcasting company. First, we determine the feasibility of using AIA in such a context with the aim of minimizing an extensive retraining whenever a new tag needs to be incorporated in the tag set population. Then, an image annotation tool integrating a Convolutional Neural Network model (AlexNet) for feature extraction and a K-Nearest-Neighbours classifier for tag assignment to images is introduced and tested. The obtained performances are very promising addressing the proposed approach as valuable to tackle the problem of image tagging in the framework of a broadcasting company, whilst not yet optimal for integration in the business process.
DOCUMENT
Adverse Outcome Pathways (AOPs) are conceptual frameworks that tie an initial perturbation (molecular initiat- ing event) to a phenotypic toxicological manifestation (adverse outcome), through a series of steps (key events). They provide therefore a standardized way to map and organize toxicological mechanistic information. As such, AOPs inform on key events underlying toxicity, thus supporting the development of New Approach Methodologies (NAMs), which aim to reduce the use of animal testing for toxicology purposes. However, the establishment of a novel AOP relies on the gathering of multiple streams of evidence and infor- mation, from available literature to knowledge databases. Often, this information is in the form of free text, also called unstructured text, which is not immediately digestible by a computer. This information is thus both tedious and increasingly time-consuming to process manually with the growing volume of data available. The advance- ment of machine learning provides alternative solutions to this challenge. To extract and organize information from relevant sources, it seems valuable to employ deep learning Natural Language Processing techniques. We review here some of the recent progress in the NLP field, and show how these techniques have already demonstrated value in the biomedical and toxicology areas. We also propose an approach to efficiently and reliably extract and combine relevant toxicological information from text. This data can be used to map underlying mechanisms that lead to toxicological effects and start building quantitative models, in particular AOPs, ultimately allowing animal-free human-based hazard and risk assessment.
DOCUMENT
To study the ways in which compounds can induce adverse effects, toxicologists have been constructing Adverse Outcome Pathways (AOPs). An AOP can be considered as a pragmatic tool to capture and visualize mechanisms underlying different types of toxicity inflicted by any kind of stressor, and describes the interactions between key entities that lead to the adverse outcome on multiple biological levels of organization. The construction or optimization of an AOP is a labor intensive process, which currently depends on the manual search, collection, reviewing and synthesis of available scientific literature. This process could however be largely facilitated using Natural Language Processing (NLP) to extract information contained in scientific literature in a systematic, objective, and rapid manner that would lead to greater accuracy and reproducibility. This would support researchers to invest their expertise in the substantive assessment of the AOPs by replacing the time spent on evidence gathering by a critical review of the data extracted by NLP. As case examples, we selected two frequent adversities observed in the liver: namely, cholestasis and steatosis denoting accumulation of bile and lipid, respectively. We used deep learning language models to recognize entities of interest in text and establish causal relationships between them. We demonstrate how an NLP pipeline combining Named Entity Recognition and a simple rules-based relationship extraction model helps screen compounds related to liver adversities in the literature, but also extract mechanistic information for how such adversities develop, from the molecular to the organismal level. Finally, we provide some perspectives opened by the recent progress in Large Language Models and how these could be used in the future. We propose this work brings two main contributions: 1) a proof-of-concept that NLP can support the extraction of information from text for modern toxicology and 2) a template open-source model for recognition of toxicological entities and extraction of their relationships. All resources are openly accessible via GitHub (https://github.com/ontox-project/en-tox).
DOCUMENT
During the past two decades the implementation and adoption of information technology has rapidly increased. As a consequence the way businesses operate has changed dramatically. For example, the amount of data has grown exponentially. Companies are looking for ways to use this data to add value to their business. This has implications for the manner in which (financial) governance needs to be organized. The main purpose of this study is to obtain insight in the changing role of controllers in order to add value to the business by means of data analytics. To answer the research question a literature study was performed to establish a theoretical foundation concerning data analytics and its potential use. Second, nineteen interviews were conducted with controllers, data scientists and academics in the financial domain. Thirdly, a focus group with experts was organized in which additional data were gathered. Based on the literature study and the participants responses it is clear that the challenge of the data explosion consist of converting data into information, knowledge and meaningful insights to support decision-making processes. Performing data analyses enables the controller to support rational decision making to complement the intuitive decision making by (senior) management. In this way, the controller has the opportunity to be in the lead of the information provision within an organization. However, controllers need to have more advanced data science and statistic competences to be able to provide management with effective analysis. Specifically, we found that an important skill regarding statistics is the visualization and communication of statistical analysis. This is needed for controllers in order to grow in their role as business partner..
DOCUMENT
Research into automatic text simplification aims to promote access to information for all members of society. To facilitate generalizability, simplification research often abstracts away from specific use cases, and targets a prototypical reader and an underspecified content creator. In this paper, we consider a real-world use case – simplification technology for use in Dutch municipalities – and identify the needs of the content creators and the target audiences in this scenario. The stakeholders envision a system that (a) assists the human writer without taking over the task; (b) provides diverse outputs, tailored for specific target audiences; and (c) explains the suggestions that it outputs. These requirements call for technology that is characterized by modularity, explainability, and variability. We argue that these are important research directions that require further exploration
MULTIFILE
Brochure from the Inauguration of Klaas Dijkstra, professor Computer Vision and Data Science
DOCUMENT
![I’M A.I. IMage Processing and Artificial Intelligence [Inaugurele rede]](https://publinova-harvester-content-prod.s3.amazonaws.com/thumbnails/files/previews/pdf/20250506132441657988.LS20Brochure20Professorship20CVDS20FINAL20WEB2028129-thumbnail-400x300.png)
When an adult claims he cannot sleep without his teddy bear, people tend to react surprised. Language interpretation is, thus, influenced by social context, such as who the speaker is. The present study reveals inter-individual differences in brain reactivity to social aspects of language. Whereas women showed brain reactivity when stereotype-based inferences about a speaker conflicted with the content of the message, men did not. This sex difference in social information processing can be explained by a specific cognitive trait, one's ability to empathize. Individuals who empathize to a greater degree revealed larger N400 effects (as well as a larger increase in γ-band power) to socially relevant information. These results indicate that individuals with high-empathizing skills are able to rapidly integrate information about the speaker with the content of the message, as they make use of voice-based inferences about the speaker to process language in a top-down manner. Alternatively, individuals with lower empathizing skills did not use information about social stereotypes in implicit sentence comprehension, but rather took a more bottom-up approach to the processing of these social pragmatic sentences.
MULTIFILE
Analyzing historical decision-related data can help support actual operational decision-making processes. Decision mining can be employed for such analysis. This paper proposes the Decision Discovery Framework (DDF) designed to develop, adapt, or select a decision discovery algorithm by outlining specific guidelines for input data usage, classifier handling, and decision model representation. This framework incorporates the use of Decision Model and Notation (DMN) for enhanced comprehensibility and normalization to simplify decision tables. The framework’s efficacy was tested by adapting the C4.5 algorithm to the DM45 algorithm. The proposed adaptations include (1) the utilization of a decision log, (2) ensure an unpruned decision tree, (3) the generation DMN, and (4) normalize decision table. Future research can focus on supporting on practitioners in modeling decisions, ensuring their decision-making is compliant, and suggesting improvements to the modeled decisions. Another future research direction is to explore the ability to process unstructured data as input for the discovery of decisions.
MULTIFILE
Retail industry consists of the establishment of selling consumer goods (i.e. technology, pharmaceuticals, food and beverages, apparels and accessories, home improvement etc.) and services (i.e. specialty and movies) to customers through multiple channels of distribution including both the traditional brickand-mortar and online retailing. Managing corporate reputation of retail companies is crucial as it has many advantages, for instance, it has been proven to impact generated revenues (Wang et al., 2016). But, in order to be able to manage corporate reputation, one has to be able to measure it, or, nowadays even better, listen to relevant social signals that are out there on the public web. One of the most extensive and widely used frameworks for measuring corporate reputation is through conducting elaborated surveys with respective stakeholders (Fombrun et al., 2015). This approach is valuable but deemed to be laborious and resource-heavy and will not allow to generate automatic alerts and quick and live insights that are extremely needed in this era of internet. For these purposes a social listening approach is needed that can be tailored to online data such as consumer reviews as the main data source. Online review datasets are a form of electronic Word-of-Mouth (WOM) that, when a data source is picked that is relevant to retail, commonly contain relevant information about customers’ perceptions regarding products (Pookulangara, 2011) and that are massively available. The algorithm that we have built in our application provides retailers with reputation scores for all variables that are deemed to be relevant to retail in the model of Fombrun et al. (2015). Examples of such variables for products and services are high quality, good value, stands behind, and meets customer needs. We propose a new set of subvariables with which these variables can be operationalized for retail in particular. Scores are being calculated using proportions of positive opinion pairs such as <fast, delivery> or <rude, staff> that have been designed per variable. With these important insights extracted, companies can act accordingly and proceed to improve their corporate reputation. It is important to emphasize that, once the design is complete and implemented, all processing can be performed completely automatic and unsupervised. The application makes use of a state of the art aspect-based sentiment analysis (ABSA) framework because of ABSA’s ability to generate sentiment scores for all relevant variables and aspects. Since most online data is in open form and we deliberately want to avoid labelling any data by human experts, the unsupervised aspectator algorithm has been picked. It employs a lexicon to calculate sentiment scores and uses syntactic dependency paths to discover candidate aspects (Bancken et al., 2014). We have applied our approach to a large number of online review datasets that we sampled from a list of 50 top global retailers according to National Retail Federation (2020), including both offline and online operation, and that we scraped from trustpilot, a public website that is well-known to retailers. The algorithm has carefully been evaluated by manually annotating a randomly sampled subset of the datasets for validation purposes by two independent annotators. The Kappa’s score on this subset was 80%.
MULTIFILE
