Artificial intelligence exceeds humans in epidemiological job coding.

BACKGROUND: Work circumstances can substantially negatively impact health. To explore this, large occupational cohorts of free-text job descriptions are manually coded and linked to exposure. Although several automatic coding tools have been developed, accurate exposure assessment is only feasible with human intervention. METHODS: We developed OPERAS, a customizable decision support system for epidemiological job coding. Using 812,522 entries, we developed and tested classification models for the Professions et Catégories Socioprofessionnelles (PCS)2003, Nomenclature d'Activités Française (NAF)2008, International Standard Classifications of Occupation (ISCO)-88, and ISCO-68. Each code comes with an estimated correctness measure to identify instances potentially requiring expert review. Here, OPERAS' decision support enables an increase in efficiency and accuracy of the coding process through code suggestions. Using the Formaldehyde, Silica, ALOHA, and DOM job-exposure matrices, we assessed the classification models' exposure assessment accuracy. RESULTS: We show that, using expert-coded job descriptions as gold standard, OPERAS realized a 0.66-0.84, 0.62-0.81, 0.60-0.79, and 0.57-0.78 inter-coder reliability (in Cohen's Kappa) on the first, second, third, and fourth coding levels, respectively. These exceed the respective inter-coder reliability of expert coders ranging 0.59-0.76, 0.56-0.71, 0.46-0.63, 0.40-0.56 on the same levels, enabling a 75.0-98.4% exposure assessment accuracy and an estimated 19.7-55.7% minimum workload reduction. CONCLUSIONS: OPERAS secures a high degree of accuracy in occupational classification and exposure assessment of free-text job descriptions, substantially reducing workload. As such, OPERAS significantly outperforms both expert coders and other current coding tools. This enables large-scale, efficient, and effective exposure assessment securing healthy work conditions.

Work can expose us to health risks, such as asbestos and constant noise. To study these risks, job descriptions are collected and classified by experts to standard codes. This is time-consuming, expensive, and requires expert knowledge. To improve this coding, we created computer code based on Artificial Intelligence that can both automate this process and suggest codes to experts, who can then check and change it manually if needed. Our system outperforms both expert coders and other available tools. This system could make studying occupational health risks more efficient and accurate, resulting in safer work environments.

Feedback beyond accuracy: Using eye-tracking to detect comprehensibility and interest during reading.

Knowing what information a user wants is a paramount challenge to information science and technology. Implicit feedback is key to solving this challenge, as it allows information systems to learn about a user's needs and preferences. The available feedback, however, tends to be limited and its interpretation shows to be difficult. To tackle this challenge, we present a user study that explores whether tracking the eyes can unpack part of the complexity inherent to relevance and relevance decisions. The eye behavior of 30 participants reading 18 news articles was compared with their subjectively appraised comprehensibility and interest at a discourse level. Using linear regression models, the eye-tracking signal explained 49.93% (comprehensibility) and 30.41% (interest) of variance (p < .001). We conclude that eye behavior provides implicit feedback beyond accuracy that enables new forms of adaptation and interaction support for personalized information systems.

The fast continuous wavelet transformation (fCWT) for real-time, high-quality, noise-resistant time-frequency analysis.

The spectral analysis of signals is currently either dominated by the speed-accuracy trade-off or ignores a signal's often non-stationary character. Here we introduce an open-source algorithm to calculate the fast continuous wavelet transform (fCWT). The parallel environment of fCWT separates scale-independent and scale-dependent operations, while utilizing optimized fast Fourier transforms that exploit downsampled wavelets. fCWT is benchmarked for speed against eight competitive algorithms, tested on noise resistance and validated on synthetic electroencephalography and in vivo extracellular local field potential data. fCWT is shown to have the accuracy of CWT, to have 100 times higher spectral resolution than algorithms equal in speed, to be 122 times and 34 times faster than the reference and fastest state-of-the-art implementations and we demonstrate its real-time performance, as confirmed by the real-time analysis ratio. fCWT provides an improved balance between speed and accuracy, which enables real-time, wide-band, high-quality, time-frequency analysis of non-stationary noisy signals.

Virtual Reality Aids Game Navigation: Evidence from the Hypertext Lostness Measure.

Instead of traditional free movement, node-based movement can be used in virtual reality (VR) games. In node-based movement systems, players navigate by jumping to set locations. Node-based movement is similar to hypertext navigation. We show that the hypertext lostness measure can be used as a game analytic to evaluate navigational efficiency. In a randomized controlled trial with 25 adolescent participants, an immersive desktop game environment and a VR game environment were compared on the transmission of in-game educational content and navigational efficiency. Results show that the hypertext lostness measure is also valuable outside its original hypertext domain: in VR. VR did not improve players' retention of factual knowledge, but did significantly improve players' spatial knowledge and navigational efficiency. We conclude (a) the hypertext lostness measure is also valuable for node-based VR games and (b) VR games add to spatial learning, even when compared with already immersive desktop games.

A dataset of continuous affect annotations and physiological signals for emotion analysis.

From a computational viewpoint, emotions continue to be intriguingly hard to understand. In research, a direct and real-time inspection in realistic settings is not possible. Discrete, indirect, post-hoc recordings are therefore the norm. As a result, proper emotion assessment remains a problematic issue. The Continuously Annotated Signals of Emotion (CASE) dataset provides a solution as it focusses on real-time continuous annotation of emotions, as experienced by the participants, while watching various videos. For this purpose, a novel, intuitive joystick-based annotation interface was developed, that allowed for simultaneous reporting of valence and arousal, that are instead often annotated independently. In parallel, eight high quality, synchronized physiological recordings (1000 Hz, 16-bit ADC) were obtained from ECG, BVP, EMG (3x), GSR (or EDA), respiration and skin temperature sensors. The dataset consists of the physiological and annotation data from 30 participants, 15 male and 15 female, who watched several validated video-stimuli. The validity of the emotion induction, as exemplified by the annotation and physiological data, is also presented.

Fast Exact Euclidean Distance (FEED): A New Class of Adaptable Distance Transforms.

A new unique class of foldable distance transforms of digital images (DT) is introduced, baptized: Fast exact euclidean distance (FEED) transforms. FEED class algorithms calculate the DT starting-directly from the definition or rather its inverse. The principle of FEED class algorithms is introduced, followed by strategies for their efficient implementation. It is shown that FEED class algorithms unite properties of ordered propagation, raster scanning, and independent scanning DT. Moreover, FEED class algorithms shown to have a unique property: they can be tailored to the images under investigation. Benchmarks are conducted on both the Fabbri et al. data set and on a newly developed data set. Three baseline, three approximate, and three state-of-the-art DT algorithms were included, in addition to two implementations of FEED class algorithms. It illustrates that FEED class algorithms i) provide truly exact Euclidean DT; ii) do no suffer from disconnected Voronoi tiles, which is a unique feature for non-parallel but fast DT; iii) outperform any other approximate and exact Euclidean DT with its time complexity O(N), even after their optimization; and iv) are unequaled in that they can be adapted to the characteristics of the image class at hand.

Representing 3D virtual objects: interaction between visuo-spatial ability and type of exploration.

We investigated individual differences in interactively exploring 3D virtual objects. 36 participants explored 24 simple and 24 difficult objects (composed of respectively three and five Biederman geons) actively, passively, or not at all. Both their 3D mental representation of the objects and visuo-spatial ability was assessed. Results show that, regardless of the object's complexity, people with a low VSA benefit from active exploration of objects, where people with a middle or high VSA do not. These findings extend and refine earlier research on interactively learning visuo-spatial information and underline the importance to take individual differences into account.

Navigating through virtual environments: visual realism improves spatial cognition.

Recent advances in computer technology have significantly facilitated the use of virtual environments (VE) for small and medium enterprises (SME). However, achieving visual realism in such VE requires high investments in terms of time and effort, while its usefulness has not yet become apparent from research. Other qualities of VE, such as the use of large displays, proved its effectiveness in enhancing the individual user's spatial cognition. The current study assessed whether the same benefits apply for visual realism in VE. Thirty-two participants were divided into two groups, who explored either a photorealistic or a nonrealistic supermarket presented on a large screen. The participants were asked to navigate through the supermarket on a predetermined route. Subsequently, spatial learning was tested in four pen-and-paper tests that assessed how accurately they had memorized the route and the environment's spatial layout. The study revealed increased spatial learning from the photorealistic compared to the nonrealistic supermarket. Specifically, participants performed better on tests that involved egocentric spatial knowledge. The results suggest visual realism is useful because it increases the user's spatial knowledge in the VE. Therefore, the current study provides clear evidence that it is worthwhile for SME to invest in achieving visual realism in VE.

Exploring workload and attention measurements with uLog mouse data.

User-system interactions (e.g., mouse clicks and movements) can be logged with the uLog computer program. A Web-based study with 20 participants was conducted to investigate the feasibility of using uLog data as an indicator of workload and attention. Eye fixation, heart rate variability (HRV), and skin conductance were used to unveil users' workload and attention and, hence, to validate uLog data as indicators of these. Results on one of the Tasks did indeed show correlations between uLog data and HRV. This is a promising first step toward the validation of uLog mouse data as indicators of workload and attention.

Considerations for emotion-aware consumer products.

Emotion-aware consumer products require reliable, short-term emotion assessment (i.e., unobtrusive, robust, and lacking calibration). To explore the feasibility of this, an experiment was conducted where the galvanic skin response (GSR) and three electromyography (EMG) signals (frontalis, corrugator supercilii, and zygomaticus major) were recorded on 24 participants who watched eight 2-min emotion inducing film fragments. The unfiltered psychophysiological signals were processed and six statistical parameters (i.e., mean, absolute deviation, standard deviation, variance, skewness, and kurtosis) were derived for each 10-s interval of the film fragment. For each physiological signal, skewness and kurtosis discriminated among affective states, accompanied by other parameters, depending on the signal. The skewness parameter also showed to indicate mixed emotions. Moreover, a mapping of events in the fragments on the signals showed the importance of short-term emotion assessment. Hence, this research identified generic features, denoted important considerations, and illustrated the feasibility of emotion-aware consumer products.

Emotional Prosody Measurement (EPM): a voice-based evaluation method for psychological therapy effectiveness.

The voice embodies three sources of information: speech, the identity, and the emotional state of the speaker (i.e., emotional prosody). The latter feature is resembled by the variability of the F0 (also named fundamental frequency of pitch) (SD F0). To extract this feature, Emotional Prosody Measurement (EPM) was developed, which consists of 1) speech recording, 2) removal of speckle noise, 3) a Fourier Transform to extract the F0-signal, and 4) the determination of SD F0. After a pilot study in which six participants mimicked emotions by their voice, the core experiment was conducted to see whether EPM is successful. Twenty-five patients suffering from a panic disorder with agoraphobia participated. Two methods (story-telling and reliving) were used to trigger anxiety and were compared with comparable but more relaxed conditions. This resulted in a unique database of speech samples that was used to compare the EPM with the Subjective Unit of Distress to validate it as measure for anxiety/stress. The experimental manipulation of anxiety proved to be successful and EPM proved to be a successful evaluation method for psychological therapy effectiveness.